# Numerical Analysis

**Using MATLAB® and Excel®
**

Third Edition

Steven T. Karris

Orchard Publications www.orchardpublications.com

Numerical Analysis

Using MATLAB® and Excel® Third Edition

Students and working professionals will find Numerical Analysis Using MATLAB® and Excel®, Third Edition, to be a concise and easy-to-learn text. It provides complete, clear, and detailed explanations of the principal numerical analysis methods and well known functions used in science and engineering. These are illustrated with many real-world examples.

This text includes the following chapters and appendices: • Introduction to MATLAB • Root Approximations • Sinusoids and Complex Numbers • Matrices and Determinants • Review of Differential Equations • Fourier, Taylor, and Maclaurin Series • Finite Differences and Interpolation • Linear and Parabolic Regression • Solution of Differential Equations by Numerical Methods • Integration by Numerical Methods • Difference Equations • Partial Fraction Expansion • The Gamma and Beta Functions • Orthogonal Functions and Matrix Factorizations • Bessel, Legendre, and Chebyshev Polynomials • Optimization Methods • Difference Equations in Discrete-Time Systems • Introduction to Simulink • Ill-Conditioned Matrices Each chapter contains numerous practical applications supplemented with detailed instructions for using MATLAB and/or Excel to obtain quick solutions. Steven T. Karris is the president and founder of Orchard Publications, has undergraduate and graduate degrees in electrical engineering, and is a registered professional engineer in California and Florida. He has more than 35 years of professional engineering experience and more than 30 years of teaching experience as an adjunct professor, most recently at UC Berkeley, California.

Orchard Publications Visit us on the Internet www.orchardpublications.com or email us: info@orchardpublications.com

1-9 934404-0 04-1 1 ISBN-13: 978-1 934404-0 04-7 7 ISBN-10: 1-9

$60.00 USA

Numerical Analysis

Using MATLAB® and Excel®

Third Edition

Steven T. Karris

Orchard Publications www.orchardpublications.com

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright ” 2007 Orchard Publications. All rights reserved. Printed in the United States of America. No part of this publication may be reproduced or distributed in any form or by any means, or stored in a data base or retrieval system, without the prior written permission of the publisher. Direct all inquiries to Orchard Publications, info@orchardpublications.com Product and corporate names are trademarks or registered trademarks of the Microsoft™ Corporation and The MathWorks™, Inc. They are used only for identification and explanation, without intent to infringe.

**Library of Congress Cataloging-in-Publication Data
**

Library of Congress Control Number: 2007922100 Copyright TX 5-589-152

ISBN-13: 978-1-934404-04-1 ISBN-10: 1-934404-04-7

Disclaimer

The author has made every effort to make this text as complete and accurate as possible, but no warranty is implied. The author and publisher shall have neither liability nor responsibility to any person or entity with respect to any loss or damages arising from the information contained in this text.

Preface

Numerical analysis is the branch of mathematics that is used to find approximations to difficult problems such as finding the roots of non−linear equations, integration involving complex expressions and solving differential equations for which analytical solutions do not exist. It is applied to a wide variety of disciplines such as business, all fields of engineering, computer science, education, geology, meteorology, and others. Years ago, high−speed computers did not exist, and if they did, the largest corporations could only afford them; consequently, the manual computation required lots of time and hard work. But now that computers have become indispensable for research work in science, engineering and other fields, numerical analysis has become a much easier and more pleasant task. This book is written primarily for students/readers who have a good background of high−school algebra, geometry, trigonometry, and the fundamentals of differential and integral calculus.* A prior knowledge of differential equations is desirable but not necessary; this topic is reviewed in Chapter 5. One can use Fortran, Pascal, C, or Visual Basic or even a spreadsheet to solve a difficult problem. It is the opinion of this author that the best applications programs for solving engineering problems are 1) MATLAB which is capable of performing advanced mathematical and engineering computations, and 2) the Microsoft Excel spreadsheet since the versatility offered by spreadsheets have revolutionized the personal computer industry. We will assume that the reader has no prior knowledge of MATLAB and limited familiarity with Excel. We intend to teach the student/reader how to use MATLAB via practical examples and for detailed explanations he/she will be referred to an Excel reference book or the MATLAB User’s Guide. The MATLAB commands, functions, and statements used in this text can be executed with either MATLAB Student Version 12 or later. Our discussions are based on a PC with Windows XP platforms but if you have another platform such as Macintosh, please refer to the appropriate sections of the MATLAB’s User Guide that also contains instructions for installation. MATLAB is an acronym for MATrix LABoratory and it is a very large computer application which is divided to several special application fields referred to as toolboxes. In this book we will be using the toolboxes furnished with the Student Edition of MATLAB. As of this writing, the latest release is MATLAB Student Version Release 14 and includes SIMULINK which is a

* These topics are discussed in Mathematics for Business, Science, and Technology, Third Edition, ISBN 0−9709511− 0−8. This text includes probability and other advanced topics which are supplemented by many practical applications using Microsoft Excel and MATLAB.

software package used for modeling, simulating, and analyzing dynamic systems. SIMULINK is not discussed in this text; the interested reader may refer to Introduction to Simulink with Engineering Applications, ISBN 0−9744239−7−1. Additional information including purchasing the software may be obtained from The MathWorks, Inc., 3 Apple Hill Drive, Natick, MA 01760−2098. Phone: 508 647−7000, Fax: 508 647−7001, e−mail: info@mathwork.com and web site http://www.mathworks.com. The author makes no claim to originality of content or of treatment, but has taken care to present definitions, statements of physical laws, theorems, and problems. Chapter 1 is an introduction to MATLAB. The discussion is based on MATLAB Student Version 5 and it is also applicable to Version 6. Chapter 2 discusses root approximations by numerical methods. Chapter 3 is a review of sinusoids and complex numbers. Chapter 4 is an introduction to matrices and methods of solving simultaneous algebraic equations using Excel and MATLAB. Chapter 5 is an abbreviated, yet practical introduction to differential equations, state variables, state equations, eigenvalues and eigenvectors. Chapter 6 discusses the Taylor and Maclaurin series. Chapter 7 begins with finite differences and interpolation methods. It concludes with applications using MATLAB. Chapter 8 is an introduction to linear and parabolic regression. Chapters 9 and 10 discuss numerical methods for differentiation and integration respectively. Chapter 11 is a brief introduction to difference equations with a few practical applications. Chapters 12 is devoted to partial fraction expansion. Chapters 13, 14, and 15 discuss certain interesting functions that find wide application in science, engineering, and probability. This text concludes with Chapter 16 which discusses three popular optimization methods. New to the Third Edition This is an extensive revision of the first edition. The most notable changes are the inclusion of Fourier series, orthogonal functions and factorization methods, and the solutions to all end−of− chapter exercises. It is in response to many readers who expressed a desire to obtain the solutions in order to check their solutions to those of the author and thereby enhancing their knowledge. Another reason is that this text is written also for self−study by practicing engineers who need a review before taking more advanced courses such as digital image processing. The author has prepared more exercises and they are available with their solutions to those instructors who adopt this text for their class. Another change is the addition of a rather comprehensive summary at the end of each chapter. Hopefully, this will be a valuable aid to instructors for preparation of view foils for presenting the material to their class. The last major change is the improvement of the plots generated by the latest revisions of the MATLAB® Student Version, Release 14. Orchard Publications Fremont, California www.orchardpublications.com info@orchardpublications.com

**Table of Contents 1 Introduction to MATLAB
**

1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 1.11 1.12 1.13

1−1

Command Window.................................................................................................1−1 Roots of Polynomials...............................................................................................1−3 Polynomial Construction from Known Roots ........................................................1−4 Evaluation of a Polynomial at Specified Values .....................................................1−5 Rational Polynomials ..............................................................................................1−8 Using MATLAB to Make Plots..............................................................................1−9 Subplots.................................................................................................................1−18 Multiplication, Division and Exponentiation.......................................................1−19 Script and Function Files......................................................................................1−26 Display Formats ....................................................................................................1−31 Summary ...............................................................................................................1−33 Exercises................................................................................................................1−37 Solutions to End−of−Chapter Exercises ...............................................................1−38

MATLAB Computations: Entire chapter

2

Root Approximations 2.1 2.2 2.3 2.4 2.5 2.6

2−1

Newton’s Method for Root Approximation...........................................................2−1 Approximations with Spreadsheets........................................................................2−7 The Bisection Method for Root Approximation .................................................2−19 Summary...............................................................................................................2−27 Exercises ...............................................................................................................2−28 Solutions to End−of−Chapter Exercises ...............................................................2−29

MATLAB Computations: Pages 2−2 through 2−7, 2−14, 2−21 through 2−23, 2−29 through 2−34 Excel Computations: Pages 2−8 through 2−19, 2−24 through 2−26

3

Sinusoids and Phasors 3.1 3.2 3.3 3.4 3.5 3.6 3.7

3−1

Alternating Voltages and Currents ........................................................................ 3−1 Characteristics of Sinusoids.................................................................................... 3−2 Inverse Trigonometric Functions ......................................................................... 3−10 Phasors.................................................................................................................. 3−10 Addition and Subtraction of Phasors ................................................................... 3−11 Multiplication of Phasors...................................................................................... 3−12 Division of Phasors ............................................................................................... 3−13

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

i

3.8 3.9 3.10 3.11

Exponential and Polar Forms of Phasors ..............................................................3−13 Summary ...............................................................................................................3−24 Exercises................................................................................................................3−27 Solutions to End−of−Chapter Exercises................................................................3−28

MATLAB Computations: Pages 3−15 through 3−23, 3−28 through 3−31 Simulink Modeling: Pages 3−16 through 3−23

4

Matrices and Determinants 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10 4.11 4.12 4.13 4.14

4−1

Matrix Definition.....................................................................................................4−1 Matrix Operations ...................................................................................................4−2 Special Forms of Matrices........................................................................................4−5 Determinants ...........................................................................................................4−9 Minors and Cofactors ............................................................................................4−13 Cramer’s Rule ........................................................................................................4−18 Gaussian Elimination Method...............................................................................4−20 The Adjoint of a Matrix ........................................................................................4−22 Singular and Non−Singular Matrices ....................................................................4−22 The Inverse of a Matrix.........................................................................................4−23 Solution of Simultaneous Equations with Matrices ..............................................4−25 Summary ................................................................................................................4−32 Exercises ................................................................................................................4−36 Solutions to End−of−Chapter Exercises ................................................................4−38

MATLAB Computations: Pages 4−3, 4−5 through 4−8, 4−10, 4−12, 4−3, 4−5, 4−19 through 4−20, 4−24, 4−26, 4−28, 4−30, 4−38, 4−41, 4−43 Excel Computations: Pages 4−28 through 4−29, 4−42 through 4−43

5

Differential Equations, State Variables, and State Equations 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 5.10 5.11 5.12

5−1

Simple Differential Equations..................................................................................5−1 Classification............................................................................................................5−2 Solutions of Ordinary Differential Equations (ODE) .............................................5−6 Solution of the Homogeneous ODE ...................................................................... 5−8 Using the Method of Undetermined Coefficients for the Forced Response ........ 5−10 Using the Method of Variation of Parameters for the Forced Response ............. 5−20 Expressing Differential Equations in State Equation Form.................................. 5−24 Solution of Single State Equations ....................................................................... 5−27 The State Transition Matrix ................................................................................ 5−28 Computation of the State Transition Matrix ...................................................... 5−30 Eigenvectors.......................................................................................................... 5−38 Summary .............................................................................................................. 5−42

ii

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

5.13 Exercises ............................................................................................................... 5−47 5.14 Solutions to End−of−Chapter Exercises............................................................... 5−49 MATLAB Computations: Pages 5−11, 5−13 through 5−14, 5−16 through 5−17, 5−19, 5−23, 5−33 through 5−35, 5−37, 5−49 through 5−53, 5−55 Fourier, Taylor, and Maclaurin Series 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 6.10 6.11 6.12 6.13 6−1 Wave Analysis ........................................................................................................6−1 Evaluation of the Coefficients ...............................................................................6−2 Symmetry ...............................................................................................................6−7 Waveforms in Trigonometric Form of Fourier Series .........................................6−12 Alternate Forms of the Trigonometric Fourier Series .........................................6−25 The Exponential Form of the Fourier Series .......................................................6−29 Line Spectra .........................................................................................................6−33 Numerical Evaluation of Fourier Coefficients .....................................................6−36 Power Series Expansion of Functions ..................................................................6−40 Taylor and Maclaurin Series ................................................................................6−41 Summary ..............................................................................................................6−48 Exercises ..............................................................................................................6−51 Solutions to End−of−Chapter Exercises ..............................................................6−53

6

MATLAB Computations: Pages 6−35, 6−45, 6−58 through 6−61 Excel Computations: Pages 6−37 through 6−39

7

Finite Differences and Interpolation 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 7.10 7.11

7−1

Divided Differences ............................................................................................... 7−1 Factorial Polynomials ............................................................................................. 7−6 Antidifferences ................................................................................................... 7−12 Newton’s Divided Difference Interpolation Method ......................................... 7−15 Lagrange’s Interpolation Method ........................................................................ 7−17 Gregory−Newton Forward Interpolation Method .............................................. 7−19 Gregory−Newton Backward Interpolation Method ........................................... 7−21 Interpolation with MATLAB ............................................................................. 7−24 Summary ............................................................................................................. 7−39 Exercises ............................................................................................................. 7−44 Solutions to End−of−Chapter Exercises ............................................................. 7−45

MATLAB Computations: Pages 7−8 through 7−9, 7−13 through 7−15, 7−26 through 7−38, 7−45 through 7−46, 7−48, 7−50, 7−52 Excel Computations: Pages 7−17 through 7−19, 7−22 through 7−25, 7−49, 7−52

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

iii

8

Linear and Parabolic Regression 8.1 8.2 8.3 8.4 8.5 8.6 8.7

8−1

Curve Fitting ..........................................................................................................8−1 Linear Regression ...................................................................................................8−2 Parabolic Regression ..............................................................................................8−7 Regression with Power Series Approximations ....................................................8−14 Summary ..............................................................................................................8−24 Exercises ...............................................................................................................8−26 Solutions to End−of−Chapter Exercises ...............................................................8−28

MATLAB Computations: Pages 8−11 through 8−14, 8−17 through 8−23, 8−30 through 8−34 Excel Computations: Pages 8−5 through 8−10, 8−15 through 8−19, 8−28 through 8−32

9

Solution of Differential Equations by Numerical Methods 9.1 9.2 9.3 9.4 9.5 9.6 9.7

9−1

Taylor Series Method ............................................................................................ 9−1 Runge−Kutta Method ............................................................................................ 9−5 Adams’ Method ................................................................................................... 9−13 Milne’s Method .................................................................................................... 9−15 Summary .............................................................................................................. 9−17 Exercises .............................................................................................................. 9−20 Solutions to End−of−Chapter Exercises .............................................................. 9−21

MATLAB Computations: Pages 9−5, 9−9 through 9−12, 9−21 through 9−23 Excel Computations: Page 9−2, 9−14, 9−22 through 9−26

10

Integration by Numerical Methods 10.1 10.2 10.3 10.4 10.5

10−1

The Trapezoidal Rule .......................................................................................... 10−1 Simpson’s Rule ..................................................................................................... 10−6 Summary ............................................................................................................ 10−14 Exercises ............................................................................................................ 10−15 Solution to End−of−Chapter Exercises .............................................................. 10−16

MATLAB Computations: Pages 10−3 through 10−6, 10−9 through 10−13, 10−16, 10−18 through 10−21 Excel Computations: Pages 10−10, 10−19 through 10−21

11

Difference Equations

11−1

11.1 Introduction ......................................................................................................... 11−1 11.2 Definition, Solutions, and Applications .............................................................. 11−1 11.3 Fibonacci Numbers .............................................................................................. 11−7

iv

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

11.4 Summary .............................................................................................................11−11 11.5 Exercises ............................................................................................................. 11−13 11.6 Solutions to End−of−Chapter Exercises .............................................................11−14

12

Partial Fraction Expansion 12.1 12.2 12.3 12.4 12.5

12−1

Partial Fraction Expansion ..................................................................................12−1 Alternate Method of Partial Fraction Expansion ..............................................12−13 Summary ............................................................................................................12−19 Exercises ............................................................................................................12−22 Solutions to End−of−Chapter Exercises ............................................................12−23

MATLAB Computations: Pages 12−3 through 12−5, 12−9 through 12−12, 12−16 through 12-18, 12−23 through 12−28

13

The Gamma and Beta Functions and Distributions 13.1 13.2 13.3 13.4 13.5 13.6 13.7

13−1

The Gamma Function .........................................................................................13−1 The Gamma Distribution ..................................................................................13−16 The Beta Function .............................................................................................13−17 The Beta Distribution ........................................................................................13−20 Summary ............................................................................................................13−22 Exercises ............................................................................................................13−24 Solutions to End−of−Chapter Exercises ............................................................13−25

MATLAB Computations: Pages 13−3, 13−5, 13−10, 13−19, 13−25 Excel Computations: Pages 13−5, 13−10, 13−16 through 13−17, 13−19, 13−21

14

Orthogonal Functions and Matrix Factorizations 14.1 14.2 14.3 14.4 14.5 14.6 14.7 14.8 14.9 14.10 14.11

14−1

Orthogonal Functions ......................................................................................14−1 Orthogonal Trajectories ...................................................................................14−2 Orthogonal Vectors ..........................................................................................14−4 The Gram−Schmidt Orthogonalization Procedure ..........................................14−7 The LU Factorization .......................................................................................14−9 The Cholesky Factorization ............................................................................14−23 The QR Factorization .....................................................................................14−25 Singular Value Decomposition .......................................................................14−28 Summary .........................................................................................................14−30 Exercises .........................................................................................................14−32 Solutions to End−of−Chapter Exercises .........................................................14−34

MATLAB Computations: Pages 14−8 through 14−9, 14−11 through 14−29, 14−36, 14−38 through 14−39

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

v

15

Bessel, Legendre, and Chebyshev Functions 15.1 15.2 15.3 15.4 15.5 15.6 15.7

15−1

The Bessel Function ............................................................................................15−1 Legendre Functions ...........................................................................................15−10 Laguerre Polynomials .........................................................................................15−21 Chebyshev Polynomials .....................................................................................15−22 Summary ............................................................................................................15−27 Exercises .............................................................................................................15−32 Solutions to End−of−Chapter Exercises ............................................................15−33

MATLAB Computations: Pages 15−3 through 15−4, 15−6, 15−9, 14−19 through 15−22, 15−25, 15−33, 15−35 through 15−37 Excel Computations: Pages 15−5, 15−9

16

Optimization Methods 16.1 16.2 16.3 16.4 16.5 16.6

16−1

Linear Programming ........................................................................................... 16−1 Dynamic Programming ........................................................................................16−4 Network Analysis ...............................................................................................16−14 Summary ............................................................................................................16−19 Exercises .............................................................................................................15−20 Solutions to End−of−Chapter Exercises ............................................................15−22

MATLAB Computations: Pages 16−3 Excel Computations: Pages 16−4, 16−23, 16−25 through 16−27

A B C

Difference Equations in Discrete−Time Systems

A−1

A.1 Recursive Method for Solving Difference Equations........................................... A−1 A.2 Method of Undetermined Coefficients ................................................................A−1 MATLAB Computations: Pages A−4, A−7, A−9 Introduction to Simulink® B.1 B.2 B−1

Simulink and its Relation to MATLAB ............................................................... B−1 Simulink Demos .................................................................................................. B−20

MATLAB Computations and Simulink Modeling: Entire Appendix B Ill-Conditioned Matrices C−1

C.1 The Norm of a Matrix ...........................................................................................C−1 C.2 Condition Number of a Matrix .............................................................................C−2 C.3 Hilbert Matrices ....................................................................................................C−3

vi

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

MATLAB Computations: Pages C−1. Third Edition Copyright © Orchard Publications vii . C−4 through C−5 References Index R−1 IN−1 Numerical Analysis Using MATLAB® and Excel®.

This takes us to the Editor Window where we can type our script (list of statements) for a new file.1 Command Window To distinguish the screen displays from the user commands. Numerical Analysis Using MATLAB® and Excel®. we choose New and click on M−File. we will use the following conventions: Click: Click the left button of the mouse Courier Font: Screen displays Helvetica Font: User inputs at MATLAB’s command window prompt EDU>>* Helvetica Bold: MATLAB functions Bold Italic: Important terms and facts. comment lines. We must save our program with a file name which starts with a letter. This prompt is displayed also after execution of a command. procedures for naming and saving the user generated files. Third Edition Copyright © Orchard Publications 1−1 . access to MATLAB’s Editor/ Debugger. and making plots. important terms and MATLAB functions. or open a previously saved file.Chapter 1 Introduction to MATLAB T his chapter is an introduction of the basic MATLAB commands and functions. Important! MATLAB is case sensitive. 1. and return to the command screen to execute the program as explained below. These bars will not be shown whenever an example begins at the top of a page or at the bottom of a page. we see the toolbar on top of the command screen and the prompt EDU>>. that is. save it. Several examples are provided with detailed explanations.m. From the File menu on the toolbar. for example.m. myfile01. Also. notes. MATLAB now waits for a new command from the user. and file names When we first start MATLAB. finding the roots of a polynomial. The files that we create are saved with the file name we use and the extension . Throughout this text. a left justified horizontal bar will denote the beginning of an example. t and T are two different characters in MATLAB language. the right justified bar will be omitted. We can use the Editor/Debugger to write our program. when one example follows immediately after a previous example. It is a good * EDU>> is the MATLAB prompt in the Student Version. To use the Editor/Debugger: 1. Thus. and a right justified horizontal bar will denote the end of the example. it distinguishes between upper− and lower−case letters.

The MATLAB User’s Guide provides more information on this topic. For example. 2. To execute a program. we type help matlab graphics. or j to denote the imaginary part of a complex number. we should use the clear command. we type demo and we see the MATLAB Demos menu. In other words. we ought to name and save that file as matrices01. For example. it will be understood that each input command is typed after the EDU>> prompt and followed by the <enter> key. we can use the clc command. the statement z=3−4j 1−2 Numerical Analysis Using MATLAB® and Excel®.Chapter 1 Introduction to MATLAB practice to save the script in a file name that is descriptive of our script content. but leave only the EDU>> prompt on the upper left of the screen. We should also use a separate disk to backup our files. Third Edition Copyright © Orchard Publications . and equations without exiting. But if we want to clear all previous values. For instance. variables and equations which we have already used. If we have saved our file in drive a or any other drive. we type quit or exit. One of the most powerful features of MATLAB is the ability to do computations involving complex numbers. To get help with other MATLAB topics. to get information on graphics. Once the script is written and saved as an m−file. we may exit the Editor/Debugger window by clicking on Exit Editor/Debugger of the File menu. All text after the % (percent) symbol is interpreted by MATLAB as a comment line and thus it is ignored during the execution of a program. if the script performs some matrix operations. For instance. The MATLAB User’s Guide contains numerous help topics. variables. we must make sure that it is added it to the desired directory in MATLAB’s search path. When we are done and want to leave MATLAB. we type the file name without the . such as 3 – 4i or 3 – 4j . we press <enter> and observe the execution and the values obtained from it. A comment can be typed on the same line as the function or command or as a separate line.m or any other similar name.m extension at the EDU>> prompt. 3. We can do this periodically to become familiar with them. we click on the Close button. then. The command help matlab iofun will display input/output information. Whenever we want to return to the command window. To appreciate MATLAB’s capabilities. it is like exiting MATLAB and starting it again. and MATLAB returns to the command window. Henceforth. We can also get help from the Help pull− down menu. The clc command clears the screen but MATLAB still remembers all values. if we want MATLAB to retain all previously entered commands.q) % performs multiplication of polynomials p and q % The next statement performs partial fraction expansion of p(x) / q(x) are both correct. the statements conv(p. This command erases everything. We can use either i . we can type help followed by any topic from the displayed menu.

if the imaginary part is a function or variable such as cos ( x ) . Third Edition Copyright © Orchard Publications 1−3 .0000 3. We can find the roots of any polynomial with the roots(p) function where p is a row vector containing the polynomial coefficients in descending order. Numerical Analysis Using MATLAB® and Excel®.0000 We observe that MATLAB displays the polynomial coefficients as a row vector. 1. that is.4. a multiplication (*) sign between 4 and j was not necessary because the complex number consists of numerical constants.1) Solution: The roots are found with the following two statements. we must type cos(x)*j or j*cos(x).2 Roots of Polynomials In MATLAB. The elements a i of this vector are the coefficients of the polynomial in descending order. We must include terms whose coefficients are zero. a polynomial is expressed as a row vector of the form [ a n a n – 1 a 2 a 1 a 0 ] . and the roots as a column vector. Example 1. We have denoted the polynomial as p1.0000 2. we must use the multiplication sign. However. and the roots as roots_ p1.0000 . p1=[1 −10 35 −50 24] % Specify the coefficients of p1(x) p1 = 1 -10 35 -50 24 roots_ p1=roots(p1) % Find the roots of p1(x) roots_p1 = 4.1 Find the roots of the polynomial p 1 ( x ) = x – 10x + 35x – 50x + 24 4 3 2 (1.0000i In the example above.Roots of Polynomials displays z = 3.0000 1.

the conjugate of a complex number A = a + jb is A∗ = a – jb 1−4 Numerical Analysis Using MATLAB® and Excel®. and the roots of this polynomial as roots_ p2. 1.5014 2. we find the coefficients with the poly(r) function as shown below. then. complex roots always occur in complex conjugate* pairs. Compute the coefficients of this polynomial.2) Solution: There is no cube term. with the given roots as elements of this vector. p2=[1 −7 0 16 25 52] p2 = 1 -7 0 16 25 52 roots_ p2=roots(p2) roots_ p2 = 6. and two complex roots. The roots are found with the statements below where we have defined the polynomial as p2.3366 + 1.5711 -0. * By definition. we must enter zero as its coefficient. Third Edition Copyright © Orchard Publications .3202i -0. and 4 .3202i The result indicates that this polynomial has three real roots.2 Find the roots of the polynomial p 2 ( x ) = x – 7x + 16x + 25x + 52 5 4 2 (1.3366 . therefore. 3. Example 1. say r3 . Solution: We first define a row vector.3 It is known that the roots of a polynomial are 1.3 Polynomial Construction from Known Roots We can compute the coefficients of a polynomial from a given set of roots with the poly(r) function where r is a row vector containing the roots. 2.7428 -1.Chapter 1 Introduction to MATLAB Example 1.1. Of course.

r4=[ −1 −2 −3 4+5j 4−5j ] r4 = Columns 1 through 4 -1.5. Solution: We form a row vector.Evaluation of a Polynomial at Specified Values r3=[1 2 3 4] % Specify the roots of the polynomial r3 = 1 2 3 4 poly_r3=poly(r3) % Find the polynomial coefficients poly_r3 = 1 -10 35 -50 24 We observe that these are the coefficients of the polynomial p 1 ( x ) of Example 1. and 4 – j5 . 4 + j5. Numerical Analysis Using MATLAB® and Excel®. the polynomial is p 4 ( x ) = x + 14x + 100x + 340x + 499x + 246 (1.x) function evaluates a polynomial p ( x ) at some specified value of the independent variable x .0000 -2.0000 .0000 Column 5 -4.1. Find the coefficients of this polynomial.4 Evaluation of a Polynomial at Specified Values The polyval(p. – 2. and we find the polynomial coefficients with the poly(r) function as shown below. Example 1.0000 + 5. Third Edition Copyright © Orchard Publications 1−5 . with the given roots.0000i poly_r4 = 1 14 100 340 5 4 499 3 246 2 Therefore.0000i poly_r4=poly(r4) -4. – 3. say r4 .3) 1.4 It is known that the roots of a polynomial are – 1.0000 -3.

Chapter 1 Introduction to MATLAB Example 1. polyder(p) − produces the coefficients of the derivative of a polynomial p.) after the right bracket suppresses the display of the row vector % that contains the coefficients of p5.p2) p1p2 = 2 -6 Therefore.5) Compute the product p 1 ⋅ p 2 with the conv(a. No semicolon is used here % because we want the answer to be displayed p 5 ( x ) = x – 3x + 5x – 4x + 3x + 2 6 5 3 2 (1. Solution: p5=[1 −3 0 5 −4 3 2]. -8 34 18 -24 -74 -88 78 166 174 108 1−6 Numerical Analysis Using MATLAB® and Excel®. Solution: p1=[1 −3 0 5 7 9].5 Evaluate the polynomial at x = – 3 . % These are the coefficients % The semicolon (.4) val_minus3 = 1280 Other MATLAB functions used with polynomials are the following: conv(a.d) −divides polynomial c by polynomial d and displays the quotient q and remainder r. −3)% Evaluate p5 at x=−3.6 Let p 1 = x – 3x + 5x + 7x + 9 p 2 = 2x – 8x + 4x + 10x + 12 6 4 2 5 4 2 (1. Third Edition Copyright © Orchard Publications . p1p2=conv(p1. p2=[2 0 −8 0 4 10 12].b) − multiplies two polynomials a and b [q. % val_minus3=polyval(p5. Example 1.b) function.r]=deconv(c.

6) q = 0.8 Let p 5 = 2x – 8x + 4x + 10x + 12 Compute the derivative dp 5 ⁄ dx using the polyder(p) function. Commas will display the results whereas semicolons will suppress the display.7) der_p5 = 12 Therefore.7 Let p 3 = x – 3x + 5x + 7x + 9 p 4 = 2x – 8x + 4x + 10x + 12 Compute the quotient p 3 ⁄ p 4 using the deconv(p. the quotient q ( x ) and remainder r ( x ) are q ( x ) = 0. Example 1. Solution: p3=[1 0 −3 0 5 7 9].Evaluation of a Polynomial at Specified Values p 1 Þ p 2 = 2x 11 – 6x 5 10 – 8x + 34x + 18x – 24x 4 3 2 9 8 7 6 – 74x – 88x + 78x + 166x + 174x + 108 We can write MATLAB statements in one line if we separate them by commas or semicolons. Solution: p5=[2 0 −8 0 4 10 12].q) function. der_p5=polyder(p5) 6 4 2 r ( x ) = 4x – 3x + 3x + 2x + 3 5 4 2 (1. [q.5 Example 1. p4=[2 −8 0 0 4 10 12].5000 r = 0 4 -3 0 3 2 3 Therefore.p4) 6 5 2 7 5 3 (1.r]=deconv(p3. 0 -32 0 8 10 Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications 1−7 .

1633 ) ⋅ ( x + 0.9870 ) ⋅ ( x + 0. using the roots(p) function.9870i -1.0000 As expected. roots_den=roots(den) % Display num and den roots roots_num = 2. the complex roots occur in complex conjugate pairs.4186 – j1. Solution: num=[1 −3 0 5 7 9].8) where some of the terms in the numerator and/or denominator may be zero. Third Edition Copyright © Orchard Publications .2108 – j 0.6760 + 0.0712 ) ⋅ ( x – 2.9 Let 5 4 2 p num x – 3x + 5x + 7x + 9 R ( x ) = ----------.3370 + j0.2108 .9) Express the numerator and denominator in factored form.9870 ) ⋅ ( x + 1.0712i -0.1.4186 .9961 ) and for the denominator. that is.4922 ) ⋅ ( x – 1.9961 ) ⋅ ( x + 0. as bn x + bn – 1 x + bn – 2 x + … + b1 x + b0 (x) = -----------------------------------------------------------------------------------------------------------------------R ( x ) = Num -------------------m m–1 m–2 Den ( x ) am x + am – 1 x + am – 2 x + … + a1 x + a0 n n–1 n–2 (1.6760 .4922i -0.4186 + j1.9870i 2.3370 – j0.4922 ) ⋅ ( x + 1.= ---------------------------------------------------------------------6 4 2 p den 2x – 8x + 4x + 10x + 12 (1.0.5 Rational Polynomials Rational Polynomials are those which can be expressed in ratio form.9961i 1.2108 + 0.3370 + 0.% Do not display num and den coefficients roots_num=roots(num). den=[2 0 −8 0 4 10 12].0.2108 + j0.0712i -0. we have the factored form p num = ( x – 2.6760 + j0. we have p den = ( x – 1. We can find the roots of the numerator and denominator with the roots(p) function as before.Chapter 1 Introduction to MATLAB dp 5 ⁄ dx = 12x – 32x + 4x + 8x + 10 5 3 2 1.3370 .9304 ) ⋅ ( x + 0.1633 -1.9304 -1. Example 1.4922i -0.6760 – j0. For the numerator.9961i roots_den = 1.0000 ) 1−8 Numerical Analysis Using MATLAB® and Excel®.0712 ) ⋅ ( x + 1.0.4186 + 1.

Using MATLAB to Make Plots We can also express the numerator and denominator of this rational function as a combination of linear and quadratic factors. Plot the magnitude of the impedance.0712i)*(2. that is. Solution: We cannot type ω (omega) in the MATLAB command window. we want to plot a set of ordered pairs.2108−0. The ammeter readings were then recorded for each frequency.6760+0.1058 (1. The magnitude of the impedance Z was computed as Z = V ⁄ A and the data were tabulated in Table 1. we form the coefficient b by addition of the complex conjugate roots and this is done by inspection.3370+0. while the product of the roots is equal to the constant term c . x 1 ⋅ x 2 = c . Here. so we will use the English letter w instead. the negative sum of the roots is equal to the coefficient b of the x term.10 Consider the electric circuit of Figure 1. Third Edition Copyright © Orchard Publications 1−9 .6760−0.1.2108+0. x is the horizontal axis (abscissa) and y is the vertical axis (ordinate). where the radian frequency ω (radians/second) of the applied voltage was varied from 300 to 3000 in steps of 100 radians/second. This is a very easy task with the MATLAB plot(x.9870i) ans = 1.1. while the amplitude was held constant.9961i) % Form the product of the 2nd set of complex conjugates ans = 1. – ( x 1 + x 2 ) = b . that is. Example 1. Accordingly.4186+1.4186 −1.9870i)*(−0.4922i) ans = 3.4922i)*(1.6 Using MATLAB to Make Plots Quite often. Numerical Analysis Using MATLAB® and Excel®.0712i) % Form the product of the 1st set of complex conjugates ans = 6. that is.9971 (−0. then we multiply the complex conjugate roots to obtain the constant term c using MATLAB as indicated below.3370−0.0186 1.9961i)*(−0. We recall that in a quadratic equation of the form x 2 + bx + c = 0 whose roots are x 1 and x 2 . (2. Z versus radian frequency ω .y) command which plots y versus x .0512 (−0.

. it can be continued to the next line by typing three or more periods.111 ω (rads/s) 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 3000 Z Ohms 90.10 TABLE 1.589 71.665 140.938 469.83 266.286 44. then pressing <enter> to start a new line. This is illustrated below for the data of w and z.717 46.10 ω (rads/s) 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 Z Ohms 39.437 222. or a row vector is too long to fit in one line.056 1014..845 If a statement.432 38. % Use 4 periods to continue 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500. The data are entered as follows: w=[300 400 500 600 700 800 900 1000 1100 1200 1300 1400. Also.603 81. as mentioned before.353 103.184 97.122 42.428 48.. and continue to enter data.032 187. Electric circuit for Example 1.Chapter 1 Introduction to MATLAB A R1 R2 C V L Figure 1. we use the semicolon (.182 436. Third Edition Copyright © Orchard Publications .) to suppress the display of numbers which we do not care to see on the screen.339 52.1 Table for Example 1..088 73.751 120..481 58.611 51..513 62.1.052 145.182 40.240 54.588 67. 1−10 Numerical Analysis Using MATLAB® and Excel®.

we select MATLAB Command Window.. % Use semicolon to suppress display of these numbers % z=[39. if we want to see the values of w or z or both. The default* is off.182 40. When this command is executed.286 44. we use plot(w. or any plot.437 222.104 97.588 67. For this example.845].111. 1200 1000 800 600 400 200 0 0 500 1000 1500 2000 2500 3000 Figure 1.. The command grid toggles them.088 73.789 71. * Default is a particular value for a variable or condition that is assigned automatically by an operating system.. more presentable with the following commands: grid on: This command adds grid lines to the plot. we click on the Window pull−down menu.y) command... and we select Figure. Third Edition Copyright © Orchard Publications 1−11 .603 81.611 51.122 42. To return to the command window.665 140... Numerical Analysis Using MATLAB® and Excel®. We can make the above. and we press <enter>. 48.056. that is.240 54.. changes from off to on or vice versa. MATLAB displays the plot on MATLAB’s graph screen. The grid off command removes the grid.468.Using MATLAB to Make Plots 2600 2700 2800 2900 3000].10 This plot is referred to as the amplitude frequency response of the circuit. Of course.751 120.2.938 469. 1014.481 58..353 103. To see the graph again. Plot of impedance z versus frequency ω for Example 1. and remains in effect unless canceled or overridden by the operator.032 187. we press any key.339 52. This plot is shown in Figure 1. or from the Window pull−down menu.830 266. we use the plot(x.717 46.432 38.052 145. 90.182 436.2.z).513 62. we simply type w or z. To plot z ( y – axis ) versus w ( x – axis ).

the logarithm to the base 2 is expressed as log2(x). or current. 1−12 Numerical Analysis Using MATLAB® and Excel®. ylabel('|Z| in Ohms') After execution of these commands. The command box toggles them. Likewise. it will be understood that log is the common (base 10) logarithm. grid. Radian Frequency').y) command. Third Edition Copyright © Orchard Publications . whereas the common logarithm is expressed as log10(x).3. and the x −axis as a linear scale. and the y −axis as a linear scale. the x −axis of the frequency response is more often shown in a logarithmic scale. Modified frequency response plot of Figure 1.2. Likewise. our plot is as shown in Figure 1. 1200 1000 800 |Z| in Ohms 600 400 200 0 2 10 Magnitude of Impedance vs. except that the x −axis is represented as a log scale.y) command that is similar to the plot(x. Radian Frequency 10 w in rads/sec 3 10 4 Figure 1.3. however.y) command uses logarithmic scales for both axes.y) command. and box on restores the box. We can use the semilogx(x. xlabel(‘string’) and ylabel(‘string’) are used to label the x − and y −axis respectively. by adding the following statements: semilogx(w. the semilogy(x. We must remember. and the y −axis in dB (decibels) scale. except that the y −axis is represented as a log scale. The default is on. xlabel('w in rads/sec'). title(‘string’): This command adds a line of the text string (label) at the top of the plot. Throughout this text. If the y −axis represents power.z) command title('Magnitude of Impedance vs. A review of the decibel unit follows.Chapter 1 Introduction to MATLAB box off: This command removes the box (the solid lines which enclose the plot). and ln is the natural (base e) logarithm. voltage. The loglog(x. Let us now redraw the plot with the above options. The amplitude frequency response is usually represented with the x −axis in a logarithmic scale. the function log(x) in MATLAB is the natural logarithm.z).y) command is similar to the plot(x. % Replaces the plot(w.

000 60 dB represents a power ratio of 1. the dB values for voltage and current ratios become Numerical Analysis Using MATLAB® and Excel®. voltage.25 ) = ratio of 2. By definition. 10 dB represents a power ratio of 10 10n dB represents a power ratio of 10 It is very useful to remember that: 20 dB represents a power ratio of 100 30 dB represents a power ratio of 1. 000 Also. 1 dB represents a power ratio of approximately 1.25 then.Using MATLAB to Make Plots The ratio of any two values of the same quantity (power. 000. or a transmission line has a power loss of 7 dB (or gain – 7 dB ). we say that an amplifier has 10 dB power gain.10) Therefore. or current) can be expressed in decibels (dB). 4 dB = 3 dB + 1 dB and since 3 dB ≅ power ratio of 2 and 1 dB ≅ power ratio of 1. Third Edition Copyright © Orchard Publications 1−13 . If the gain (or loss) is 0 dB the output is equal to the input. 4 dB ≅ ratio of ( 2 × 1. For instance. Thus.5 Likewise.25 3 dB represents a power ratio of approximately 2 7 dB represents a power ratio of approximately 5 From these.= I Z Z 2 2 n if we let Z = 1 . 27 dB = 20 dB + 7 dB and this is equivalent to a power ratio of approximately 100 × 5 = 500 Using the relations y = log x = 2 log x and 2 P = V -----. we can estimate other values. P out dB = 10 log --------P in (1.

dB). and colors.4. The script for the 3−phase plot is as follows: x=linspace(0.’string’) is similar to gtext(‘string’). and colors for our plots. However.95.y.11) (1. For instance. it starts with blue and cycles through the first seven colors listed in Table 1. Also. last_value. These. 'sin(x)').65. number_of_values) This command specifies the number of data points but not the increments between data points. we can use the command gtext(‘Impedance |Z| versus Frequency’).75. and this will place a cross−hair in the Figure window. An alternate command uses the colon notation and has the format x=first: increment: last This format specifies the increments between points but not the number of data points. MATLAB allows us to specify various line types.85. v=sin(x+4*pi/3).02: 2)*pi instead. % we could have used x=0:0. plot(x. % The x−axis must be specified for each function grid on.Chapter 1 Introduction to MATLAB V out dB v = 10 log --------V in 2 and V out = 20 log --------V in (1. The command text(x.65. can be added with the plot(x. Third Edition Copyright © Orchard Publications . % turn grid and axes box on text(0. and displays a cross−hair which can be moved around with the mouse.z) command with semilogx(w. The command gtext(‘string’) switches to the current Figure Window. and we replace the semilogx(w. where s is a character string containing one or more characters shown on the three columns of Table 1. MATLAB has no default color. we add the relation dB=20*log10(v). It places a label on a plot in some specific location specified by x and y.x. and string is the label which we want to place at that location. plot symbols. Then. 'sin(x+4*pi/3)') These three waveforms are shown on the same plot of Figure 1. % pi is a built−in function in MATLAB.y.u. using the mouse. 0. we can move the cross−hair to the position where we want our label to begin. The first line of the script below has the form linspace(first_value. we did not specify line styles.= 20 log ------I in I in To display the voltage v in a dB scale on the y – axis .y. u=sin(x+2*pi/3).02*pi:2*pi or x = (0: 0.65. 0. y=sin(x). In our previous examples. markers. 'sin(x+2*pi/3)'). there is no default marker.2.2 for each additional line in the plot. text(2. We will illustrate its use with the following example which plots a 3−phase sinusoidal waveform. or a combination of these. 0. and we press <enter>.v). 60). box on. no markers are 1−14 Numerical Analysis Using MATLAB® and Excel®. text(4.s) command. 2*pi.x.12) I out 2 I out dB i = 10 log ------.

For additional information we can type help plot in MATLAB’s command screen.2 Styles. Three−phase waveforms TABLE 1. 1 sin(x) 0. If we want to connect the data points with a solid line. the command plot(x.'m*:') plots a magenta dotted line with a star at each data point.y.4. The default line is the solid line.5 -1 0 1 2 3 4 5 6 7 Figure 1. colors. and plot(x. and markets used in MATLAB Symbol b g r c m y k w Color blue green red cyan magenta yellow black white Symbol Marker point circle x−mark plus star square diamond triangle down triangle up triangle left triangle right pentagram hexagram Symbol Line Style solid line dotted line dash−dot line dashed line .'rs') plots a red square at each data point. −− ⁄ Ÿ < > p h For example. Third Edition Copyright © Orchard Publications 1−15 . Numerical Analysis Using MATLAB® and Excel®.'rs−').5 sin(x+2*pi/3) sin(x+4*pi/3) 0 -0.Using MATLAB to Make Plots drawn unless they are selected. o x + * s d − : −.y. we must type plot(x.y. but does not draw any line because no line was selected.

s2. and sn are strings specifying color. dot division.13) The three−dimensional plot is shown in Figure 1. that is. These operations will be explained in Section 1.11 * This statement uses the so called dot multiplication. y= x. where x ..y1.*y. and zn are vectors or matrices.8. MATLAB has also a three−dimensional (three−axes) capability and this is discussed next.y2.) where xn. x= −10: 0.*x. and z are three vectors of the same length. y .z2.s3. 3000 2000 1000 0 -1000 -2000 10 5 0 -5 -10 -10 -5 0 5 10 Figure 1.z3. they are drawn on two axes. The general format is plot3(x1. plot3(x... y . Three dimensional plot for Example 1.^2−1.11 Plot the function z = – 2x + x + 3y – 1 Solution: We arbitrarily choose the interval (length) shown with the script below. These strings are the same as those of the two−dimensional plots. The command plot3(x.Chapter 1 Introduction to MATLAB The plots which we have discussed thus far are two−dimensional. or line style.z1.5. marker symbol.5: 10.y. Page 1−19. 1−16 Numerical Analysis Using MATLAB® and Excel®.x3. z= −2. yn. Example 1. Third Edition Copyright © Orchard Publications .x2.s1. and dot exponentiation where these operations are preceded by a dot (period).z).z) plots a line in 3−space through the points whose coordinates are the elements of x .^3+x+3. grid % Length of vector x % Length of vector y must be same as x % Vector z is function of both x and y* 3 2 (1.y.y3. and z .5.

Solution: The volume of the cone is a function of both the radius r and the height h . In a three−dimensional plot. we can use the zlabel(‘string’) command in addition to the xlabel(‘string’) and ylabel(‘string’). h ) The three−dimensional plot is created with the following MATLAB script where.y. say z = f ( x. and the matrix Y whose columns are copies of the vector y. Page 1−19. connected with a solid line.z. To produce a mesh plot of a function of two variables.z) command displays a three−dimensional plot.y. in a three−dimensional plot we can set the limits of all three axes with the axis([xmin xmax ymin ymax zmin zmax]) command. We can generate the matrices X and Y with the [X. We can also use the mesh(x.12 The volume V of a right circular cone of radius r and height h is given by 2 V = 1 -. draws contour lines for n levels.π r h 3 (1. n ] = size ( Z ) .y. Likewise. Third Edition Copyright © Orchard Publications 1−17 . this topic will be explained in Section 1. T h e s e m u s t b e d e f i n e d a s length ( x ) = n a n d length ( y ) = m w h e r e [ m.y.y) function which creates the matrix X whose rows are copies of the vector x. In a two−dimensional plot.14) Plot the volume of the cone as r and h vary on the intervals 0 ≤ r ≤ 4 and 0 ≤ h ≤ 6 meters. the vertices of the mesh lines are the triples { x ( j ). Example 1. and y corresponds to the rows of Z . in the second line we have used the dot multiplication. We observe that x corresponds to the columns of Z . or on the same line without first executing the plot command.z. that is. The three−dimensional text(x. contour(Z.’string’) command will place string beginning at the co−ordinate ( x.y) or plot3(x. y ) . In this case. This must be done for each plot. y.Using MATLAB to Make Plots The command plot3(x.Y]=meshgrid(x. we must first generate the X and Y matrices which consist of repeated rows and columns over the range of the variables x and y . z ) on the plot. as in the previous example.'bd−') will display the plot in blue diamonds.n). division. j ) } . It must be placed after the plot(x. V = f ( r. Another command. As mentioned in the footnote of the previous page. and exponentiation. For three−dimensional plots.y. Z ( i. we can set the limits of the x − and y − axes with the axis([xmin xmax ymin ymax]) command. Numerical Analysis Using MATLAB® and Excel®. y ( i ). The mesh(x. grid on and box off are the default states.z) command with two vector argum e n t s .8.z) commands.

radius r (meters) Figure 1. volume (cubic meters)'). H. and p . n .^ 2 . radius r (meters)').5. This command divides the window into an m × n matrix of plotting areas and chooses the pth area to be active. Volume of a right circular cone.p) command following the discussion on multiplication.* H) . This. We will illustrate the use of the subplot(m. zlabel('z−axis.Chapter 1 Introduction to MATLAB [R.n. volume (cubic meters) 150 100 50 0 6 4 2 0 0 1 2 3 4 y-axis.p).6. shows how the volume of the cone increases as the radius and height are increased.n. No spaces or commas are required between the three integers m . MATLAB can generate very sophisticated and impressive three−dimensional plots. 1. and the plot of Figure 1.7 Subplots MATLAB can display up to four windows of different plots on the Figure window using the command subplot(m. are rudimentary. ylabel('y−axis. 0: 6). The possible combinations are shown in Figure 1. division and exponentiation that follows.* R . title('Volume of Right Circular Cone'). Volume of Right Circular Cone z-axis. altitude h (meters)').6. box on The three−dimensional plot of Figure 1. altitude h (meters) x-axis. Third Edition Copyright © Orchard Publications . V) xlabel('x−axis. mesh(R. % Creates R and H matrices from vectors r and h V=(pi . 1−18 Numerical Analysis Using MATLAB® and Excel®.H]=meshgrid(0: 4./ 3. The MATLAB User’s manual contains more examples.7.

To distinguish between row and column vectors. 11] or as b=[−1 3 6 11]' MATLAB produces the same display with either format as shown below. These are the matrix multiplication.7. division. Thus. We recall that the elements of a row vector are separated by spaces. and exponentiation. An easier way to construct a column vector. such a those that contained the coefficients of polynomials. 6. consisted of one row and multiple columns.2. it is called a column vector. the elements of a column vector must be separated by semicolons. and exponentiation. 3. They are explained in the following paragraphs. Possible subpot arrangements in MATLAB 1. and exponentiation. b=[−1. is to write it first as a row vector. the arrays [ a b c … ] . division. If an array has one column and multiple rows. In Section 1. and the element−by−element multiplication.8 Multiplication. Division and Exponentiation 111 Full Screen 211 212 221 222 212 221 223 211 223 224 222 224 221 223 122 Default 121 122 222 224 121 Figure 1. division. Third Edition Copyright © Orchard Publications 1−19 . 3. and then transpose it into a column vector. 6. MATLAB uses the single quotation character (¢) to transpose a vector.Multiplication. 11] b = -1 3 6 11 b=[−1 3 6 11]' b = -1 3 Numerical Analysis Using MATLAB® and Excel®. and thus are called row vectors. Division and Exponentiation MATLAB recognizes two types of multiplication. a column vector can be written either as b=[−1.

1. B=[ −2 6 −3 8 7]'. multiplication of the row vector A by the column vector B . if and A = [1 2 3 4 5] B = [ – 2 6 – 3 8 7 ]' the matrix multiplication A*B produces the single value 68. A=[1 2 3 4 5]. let us suppose that both A and B are row vectors. B=[−2 6 −3 8 7]. that is. A*B % Observe transpose operator (‘) in B (B. MATLAB displays the following message: ??? Error using ==> * Inner matrix dimensions must agree. Matrix Multiplication (multiplication of row by column vectors) Let A = [ a1 a2 a3 … an ] and B = [ b 1 b 2 b 3 … b n ]' be two vectors. We observe that A is defined as a row vector whereas B is defined as a column vector. Here. A∗ B = 1 × ( – 2 ) + 2 × 6 + 3 × ( – 3 ) + 4 × 8 + 5 × 7 = 68 and this is verified with the MATLAB script A=[1 2 3 4 5]. A*B % No transpose operator (‘) here When these statements are executed. MATLAB expects 1−20 Numerical Analysis Using MATLAB® and Excel®. is performed with the matrix multiplication operator (*). Third Edition Copyright © Orchard Publications . as indicated by the transpose operator (′). and we attempt to perform a row−by− row multiplication with the following MATLAB statements. Here. Then.Chapter 1 Introduction to MATLAB 6 11 We will now define Matrix Multiplication and Element−by−Element multiplication.15) ans = 68 Now. A*B = [ a 1 b 1 + a 2 b 2 + a 3 b 3 + … + a n b n ] = sin gle value For example. because we have used the matrix multiplication operator (*) in A*B.

whereas dot division (. as illustrated with the examples above.Multiplication.16) This product is another row vector with the same number of elements. Division and Exponentiation vector B to be a column vector. D=[−2 6 −3 8 7]. and warns us that we cannot perform this multiplication using the matrix multiplication operator (*). C. It recognizes that B is a row vector. Accordingly. As an example. Element−by−Element Multiplication (multiplication of a row vector by another row vector) Let C = [ c1 c2 c3 … cn ] and D = [ d1 d2 d3 … dn ] be two row vectors. We must remember that no space is allowed between the dot (. Here. This operator is defined below.) is never required with the plus (+) and minus (−) operators.∗ D = [ c 1 d 1 c2 d2 c3 d3 … cn dn ] (B.*). multiplication of the row vector C by the row vector D is performed with the dot multiplication operator (.) and the multiplication (*). There is no space between the dot and the multiplication symbol. and exponentiation (^) operators. let and C = [1 2 3 4 5] D = [ –2 6 –3 8 7 ] Dot multiplication of these two row vectors produce the following result. Note: A dot (.∗ D = 1 × ( – 2 ) 2 × 6 3 × ( – 3 ) 4 × 8 5 × 7 = – 2 12 – 9 32 35 Check with MATLAB: C=[1 2 3 4 5]. not a row vector. division ( /).^) are used for element− by−element division and exponentiation. Numerical Analysis Using MATLAB® and Excel®. C. Third Edition Copyright © Orchard Publications 1−21 . we must perform this type of multiplication with a different operator./) and dot exponentiation (. Thus. are used for matrix division and exponentiation. the division (/) and exponentiation (^) operators.*D % Vectors C and D must have % same number of elements % We observe that this is a dot multiplication ans = -2 12 -9 32 35 Similarly. C. as the elements of C and D . 2.

17) in the 0 ≤ t ≤ 5 seconds interval.* cos(5 ./ (t+1).01 increments y=3 .* exp(−4 .13 Write the MATLAB script that produces a simple plot for the waveform defined as y = f ( t ) = 3e –4 t 2 cos 5t – 2e –3 t t sin 2t + -----------t+1 (1. which is a number approximately equal to 2.13 Had we. we use the special MATLAB function eps. and exponentiation.13 0 1 2 t 3 4 5 Figure 1. Plot for Example 1. division. the eps number.y).* exp(−3 .01: 5. xmax. To avoid division by zero.8 shows the plot for this example. ymin and ymax. say as – 3 ≤ t ≤ 3 . Third Edition Copyright © Orchard Publications .13') Figure 1. title('Plot for Example 1. It will be used with the next example. This is because the last term (the rational fraction) of the given expression. in the example above.* t)−2 . and also MATLAB’s capability of 1−22 Numerical Analysis Using MATLAB® and Excel®. grid.* t) . The following example illustrates the use of the dot multiplication. 6 4 y=f(t) 2 0 -2 Plot for Example 1. % Define t−axis in 0. xlabel('t'). ylabel('y=f(t)'). is divided by zero when t = – 1 . There are no commas between these four arguments. Solution: The MATLAB script for this example is as follows: t=0: 0.^2 .Chapter 1 Introduction to MATLAB Example 1.* t) . the axis([xmin xmax ymin ymax]) command. defined the time interval starting with a negative value equal to or less than – 1 .2 × 10 – 16 .* sin(2 . This command must be placed after the plot command and must be repeated for each plot.8. The command axis([xmin xmax ymin ymax]) scales the current plot to the values specified by the arguments xmin. plot(t. MATLAB would have displayed the following message: Warning: Divide by zero.* t) + t .

% Interval with 100 data points % add eps to avoid division by zero % upper left of four subplots % upper right of four subplots % lower left of four subplots % lower right of four subplots These subplots are shown in Figure 1.v).9.1 0 0 2 4 6 0 0 2 4 6 Figure 1. plot(x.^ 2).Multiplication.3]).2 400 v=(sinx)2/(cosx)2 200 0. z = cos 2x. axis([0 2*pi 0 0.z). axis([0 2*pi 0 400]).* z. axis([0 2*pi 0 1]). plot(x.5 0.5 0 0 2 2 4 2 6 0 0 2 4 6 w=(sinx) *(cosx) 0. subplot(221). title('w=(sinx)^2*(cosx)^2'). title('v=(sinx)^2/(cosx)^2').14 Plot the functions y = sin 2x. Subplots for the functions of Example 1.18) in the interval 0 ≤ x ≤ 2 π using 100 data points. Third Edition Copyright © Orchard Publications 1−23 . axis([0 2*pi 0 1]). w=y . 2*pi. Example 1.14 Numerical Analysis Using MATLAB® and Excel®./ (z+eps).w). plot(x.9.y). subplot(224). title('y=(sinx)^2'). w = sin 2x ⋅ cos 2x. title('z=(cosx)^2'). plot(x. v = sin 2x ⁄ cos 2x (1. Division and Exponentiation displaying up to four windows of different plots. subplot(223). Solution: The MATLAB script to produce the four subplots is as follows: x=linspace(0.100). subplot(222).^ 2). y=(sin(x) . v=y . Use the subplot command to display these functions on four windows on the same graph. z=(cos(x) . 1 y=(sinx)2 1 z=(cosx)2 0.

w=0: 1: 2000.* w −10.1 w – 10 ⁄ w ) a. theta is the angle in radians. and the round(n) function that rounds a number to its nearest integer. the impedance Z ab as a function of the radian frequency ω can be computed from the following expression.19) 1−24 Numerical Analysis Using MATLAB® and Excel®.* (0.r) function that produces a plot in polar coordinates. Plot Im { Z } (the imaginary part of the impedance Z ) versus frequency ω . Example 1. where r is the magnitude.15 b With the given values of resistance. and plots these as two separate graphs (parts a & b). Solution: The MATLAB script below computes the real and imaginary parts of Z ab that is.Chapter 1 Introduction to MATLAB The next example illustrates MATLAB’s capabilities with imaginary numbers.10.^5. Plot the impedance Z versus frequency ω in polar coordinates.10. % Define interval with one radian interval z=(10+(10 . It also produces a polar plot (part c). the abs(z). a 10 Ω 10 Ω 10 μF 0. and capacitance. Electric circuit for Example 1. inductance. Plot Re { Z } (the real part of the impedance Z ) versus frequency ω .^ 4 −j . for simplicity. c. and the angle(z) functions that compute the absolute value (magnitude) and phase angle of the complex quantity z = x + iy = r –θ .1 H Z ab Figure 1. We will introduce the real(z) and imag(z) functions which display the real and imaginary parts of the complex quantity z = x + iy. 4 6 (1.* 10 . Third Edition Copyright © Orchard Publications . We will also use the polar(theta. 10 – j ( 10 ⁄ w ) Z ab = Z = 10 + ------------------------------------------------------5 10 + j ( 0./ (w+eps)) .15 Consider the electric circuit of Figure 1.^ 6 ./ (w+eps)))).1 . denoted as z . b./ (10 + j .

and 1. plot(w.Multiplication. Division and Exponentiation % % The first five statements (next two lines) compute and plot Re{z} real_part=real(z). ylabel('Polar Plot of Z').rndz). xlabel('radian frequency w'). ylabel('Imaginary part of Z'). The real. % % The next five statements (next two lines) compute and plot Im{z} imag_part=imag(z).% Computes the phase angle of impedance Z polar(theta.imag_part).% Computes |Z| rndz=round(abs(z)).11. 1200 1000 800 Real part of Z 600 400 200 0 0 200 400 600 800 1000 1200 radian frequency w 1400 1600 1800 2000 Figure 1. grid. xlabel('radian frequency w').15 Numerical Analysis Using MATLAB® and Excel®. grid.% Angle is the first argument grid. Plot for the real part of Z in Example 1. and polar plots are shown in Figures 1.13 respectively. ylabel('Real part of Z'). Third Edition Copyright © Orchard Publications 1−25 .% Rounds |Z| to read polar plot easier theta=angle(z). plot(w. 1. % The last six statements (next six lines) below produce the polar plot of z mag=abs(z).12.11. imaginary.real_part).

and flexible MATLAB is. Third Edition Copyright © Orchard Publications . fast. make up a script file.15 clearly illustrates how powerful.15 Example 1.Chapter 1 Introduction to MATLAB 600 400 Imaginary part of Z 200 0 -200 -400 -600 0 200 400 600 800 1000 1200 radian frequency w 1400 1600 1800 2000 Figure 1.15 90 120 150 Polar Plot of Z 1500 1000 60 30 500 180 0 210 240 270 300 330 Figure 1.m extension.9 Script and Function Files MATLAB recognizes two types of files: script files and function files. accurate. Plot for the imaginary part of Z in Example 1. the script for each of the examples we discussed earlier.13. Thus. 1. Both types are referred to as m−files since both require the .12. a script file is one which was generated and saved as an m−file with an editor such as the MATLAB’s 1−26 Numerical Analysis Using MATLAB® and Excel®. Generally. Polar plot of Z in Example 1. A script file consists of two or more built−in functions such as those we have discussed thus far.

where f is a string containing the name of a real−valued function of a single real variable. or inability to produce a result as described on the next paragraph.x) tries to find a zero of a function of one variable. The first line of a function file must contain the word function. To save it.2 ) + 0.5 Solution: We first plot this function to observe the approximate zeros. or returns NaN if the search fails. We use function files for repetitive tasks.m. maxima and minima of the function 1 1 . but the file name must have the extension . For example. maxima. Important: We must remember that we use roots(p) to find the roots of polynomials only. The function name and file name must be the same. We can avoid division by zero using the eps number.+ ----------------------------------------.lims) − plots the function specified by the string fcn between the x−axis limits specified by lims = [xmin xmax]. fplot(fcn. Example 1. followed by the output argument. and minima using the following script: (1. it is MATLAB’s response to an undefined expression such as 0 ⁄ 0 . The function fzero(f. A function file is a user−defined function using MATLAB.04 in the interval – 1.1 ) + 0. We will use the following MATLAB functions with the next example.^ 3 + cos(3 .20) Numerical Analysis Using MATLAB® and Excel®. we choose New and click on M−File. and the input argument enclosed in parentheses. the equal sign ( = ).– 10 f ( x ) = ----------------------------------------2 2 ( x – 0.01 ( x – 1. This takes us to the Editor Window where we type these two lines and we save it as myfunction.Script and Function Files Editor/Debugger.16 Find the zeros. from the File menu of the command window.* x) is a function file and must be saved.2. such as those in Examples 1. Third Edition Copyright © Orchard Publications 1−27 . NaN (Not−a−Number) is not a function. and returns that value. MATLAB searches for a value near a point where the function f changes sign. which we mentioned earlier. Using lims = [xmin xmax ymin ymax] also controls the y−axis limits. the function file consisting of the two lines below function y = myfunction(x) y=x . The string fcn must be the name of an m−file function or a string with variable x .m.1 and 1. ∞ ⁄ ∞ .5 ≤ x ≤ 1.

/ ((x−1.m function m−file with the following script: function y=funczero01(x) % Finding the zeros of the function shown below y=1/((x−0. y max = 90 .5 0 0.14. Third Edition Copyright © Orchard Publications . The maximum occurs at approximately x = 0.5 1 1.2)^2+0.01)−1/((x−1.1).16 using the plot command The roots (zeros) of this function appear to be in the neighborhood of x = – 0.3 . from the File drop menu on the Command Window.5 -1 -0. grid The plot is shown in Figure 1.^ 2 + 0.5 1. 100 80 60 40 20 0 -20 -40 -1. we can use the fplot(fcn.14 which was obtained with the plot(x. [−1./ ((x−0.^ 2 + 0.y) command as shown in Figure 1.01: 1.m to it. grid This plot is shown in Figure 1. we type the script above and we save it as funczero01. 1−28 Numerical Analysis Using MATLAB® and Excel®.01) −1. Now.y). approximately. y min = – 34 .lims) command to plot f ( x ) as follows: fplot('funczero01'.Chapter 1 Introduction to MATLAB x=−1. Plot for Example 1. plot(x. and the minimum occurs at approximately x = 1. To save this file.04) −10.04)−10.14. and when the Editor Window appears.1)^2+0.5]).14.5 Figure 1. MATLAB appends the extension . we choose New. approximately. Next. y=1.5: 0.5.15.2 and x = 0. we define and save f(x) as the funczero01.2).2 where. As expected.1 where. this plot is identical to the plot of Figure 1.

5 Figure 1.5 0 0. x2= fzero('funczero01'. 0.x2) and with this function we could compute both a minimum of some function f ( x ) or a maximum of f ( x ) since a maximum of f ( x ) is equal to a minimum of – f ( x ) . ymin=1/((x−0.x) function to compute the roots of f ( x ) in Equation (1. From elementary calculus.x1.5 -1 -0. This function is no longer used in MATLAB and thus we will compute the maxima and minima from the derivative of the given function.. The MATLAB script below will accomplish this.3788 The earlier MATLAB versions included the function fmin(f.16 using the fplot command We will use the fzero(f..2).20) more precisely. zmin=diff(ymin) zmin = -1/((x-1/10)^2+1/100)^2*(2*x-1/5)+1/((x-6/5)^2+1/25)^2*(2*x-12/5) When the command Numerical Analysis Using MATLAB® and Excel®.1)^2+0. x1= fzero('funczero01'. For this example we use the diff(x) function which produces the approximate derivative of a function.04)−10.2)^2+0. x1).3). fprintf(' and r2= %3.5 1 1. we recall that the maxima or minima of a function y = f ( x ) can be found by setting the first derivative of a function equal to zero and solving for the independent variable x . Plot for Example 1. −0. Third Edition Copyright © Orchard Publications 1−29 . fprintf('The roots (zeros) of this function are r1= %3. This can be visualized by flipping the plot of a function f ( x ) upside−down..Script and Function Files 100 80 60 40 20 0 -20 -40 -1. x2) MATLAB displays the following: The roots (zeros) of this function are r1= -0.1919 and r2= 0.4f'.15. Thus. we use the following MATLAB script: syms x ymin zmin.4f \n'.01)−1/((x−1.

^2+0. 40 20 0 -20 -40 -60 -80 -100 -1.01)−1/((x−1. ymin=1 / ((x−0.2012. x=1. ymax=−(1/((x−0.2012 The real value 1. Plot of – f ( x ) for Example 1. we substitute it into f ( x ) . that is.16. Third Edition Copyright © Orchard Publications . zmax=diff(ymax) 1−30 Numerical Analysis Using MATLAB® and Excel®. MATLAB displays a very long expression which when copied at the command prompt and executed.1)^2+0.5 Figure 1.5.5:0.01) −1 / ((x−1. ymax=−1.2012 above is the value of x at which the function y has its minimum value as we observe also in the plot of Figure 1.2) ^ 2 + 0.1812 We can find the maximum value from – f ( x ) whose plot is produced with the script x=−1.3437i ans = 0.16.01:1.01)+1.6585 + 0. produces the following: ans = 0.^2+0.6585 ./((x−1. syms x ymax zmax.2).1) ^ 2 + 0.Chapter 1 Introduction to MATLAB solve(zmin) is executed.5 -1 -0.ymax). This is accomplished with the MATLAB script below.0.1).04)−10).3437i ans = 1.5 0 0.2)^2+0.5 1 1./((x−0.16 Next we compute the first derivative of – f ( x ) and we solve for x to find the value where the maximum of ymax occurs.04)+10. grid and the plot is shown in Figure 1. To find the value of y corresponding to this value of x. plot(x.15.04) −10 ymin = -34.

Third Edition Copyright © Orchard Publications 1−31 .10 Display Formats MATLAB displays the results on the screen in integer format without decimals if the result is an integer number.6585 + 0.3437i ans = 0. Accordingly. produces the following: ans = 0. or in short floating point format with four decimals if it a fractional number.2) ^ 2 + 0. MATLAB performs all computations with accuracy up to 16 decimal places. The format displayed has nothing to do with the accuracy in the computations. All computations in MATLAB are done in double precision.Display Formats zmax = 1/((x-1/10)^2+1/100)^2*(2*x-1/5)-1/((x-6/5)^2+1/25)^2*(2*x-12/5) solve(zmax) When the command solve(zmax) is executed.15 and 1. x=0.0999.16.0.04) −10 ymax = 89.0999 From the values above we choose x = 0.2012 ans = 0. we execute the following script to obtain the value of ymin . The available formats can be displayed with the help format command as follows: help format FORMAT Set output format. FORMAT may be used to switch between different output display formats as follows: Numerical Analysis Using MATLAB® and Excel®. % Using this value find the corresponding value of ymax ymax=1 / ((x−0.2000 1.3437i ans = 1. The output format can changed with the format command.1) ^ 2 + 0.6585 . MATLAB displays a very long expression which when copied at the command prompt and executed.01) −1 / ((x−1.0999 which is consistent with the plots of Figures 1.

FORMAT + The symbols +.33 two decimal digits format + only + or − or zero are printed format rat 100/3 rational approximation 1−32 Numerical Analysis Using MATLAB® and Excel®. SHORT Scaled fixed point format with 5 digits. FORMAT LONG G Best of fixed or floating point format with 15 digits. LONG Scaled fixed point format with 15 digits. LONG E Floating point format with 15 digits. Some examples with different format displays age given below. SHORT E Floating point format with 5 digits. Same as SHORT. FORMAT RAT Approximation by ratio of small integers. FORMAT LOOSE Puts the extra line-feeds back in.3333e+01 Four decimal digits plus exponent format short g 33. .Chapter 1 Introduction to MATLAB FORMAT FORMAT FORMAT FORMAT FORMAT FORMAT Default.and blank are printed for positive. FORMAT HEX Hexadecimal format.33333333333334 16 digits format short e 3. SHORT G Best of fixed or floating point format with 5 digits. FORMAT BANK Fixed format for dollars and cents. format short 33. Third Edition Copyright © Orchard Publications . Spacing: FORMAT COMPACT Suppress extra line-feeds.333 Better of format short or format short e format bank 33. negative and zero elements.3335 Four decimal digits (default) format long 33. Imaginary parts are ignored.

the command help matlab\iofun will display input/output information. • We can write MATLAB statements in one line if we separate them by commas or semicolons. • We can compute the coefficients of a polynomial from a given set of roots with the poly(r) function where r is a row vector containing the roots. To access it.x) function evaluates a polynomial p ( x ) at some specified value of the independent variable x . • We type quit or exit when we are done and want to leave MATLAB. The elements a i of this vector are the coefficients of the polynomial in descending order. and equations without exiting. For example.d) function divides polynomial c by polynomial d and displays the quotient q and remainder r. Third Edition Copyright © Orchard Publications 1−33 . Commas will display the results whereas semicolons will suppress the display. variables. A comment can be typed on the same line as the function or command or as a separate line. • The polyder(p) function produces the coefficients of the derivative of a polynomial p. • The conv(a.11 Summary • We can get help with MATLAB topics by typing help followed by any topic available. we click on the Close button. Numerical Analysis Using MATLAB® and Excel®.Summary 1. and help matlab graphics will display help on graphics. • We find the roots of any polynomial with the roots(p) function where p is a row vector containing the polynomial coefficients in descending order. variables and equations which we have already used. • We use the clear command if we want to clear all previous values. or j to denote the imaginary part of the complex number. • In MATLAB.b) function multiplies the polynomials a and b. • The clc command clears the screen but MATLAB still remembers all values. • The polyval(p. Whenever we want to return to the command window. a polynomial is expressed as a row vector of the form [ a n a n – 1 a 2 a 1 a 0 ] . • The MATLAB Demos menu displays MATLAB’s capabilities. • All text after the % (percent) symbol is interpreted by MATLAB as a comment line and thus it is ignored during the execution of a program. We must include terms whose coefficients are zero.r]=deconv(c. we type demo and we see the different topics. • The [q. • For computations involving complex numbers we can use either i .

and ylabel(‘string’). and continue to enter data. last_value. box. Likewise. • The command linspace(first_value. the logarithm to the base 2 is expressed as log2(x). An alternate command uses the colon notation and has the format x=first: increment: last. is normally expressed in decibels (dB) and by definition. title(‘string’). • If a statement. then pressing <enter> to start a new line. it can be continued to the next line by typing three or more periods. Likewise. it places a label on a plot in some specific location specified by x and y. This command plots y versus x where x is the horizontal axis (abscissa). n n–1 n–2 1−34 Numerical Analysis Using MATLAB® and Excel®.’string’) is similar to gtext(‘string’).y.y) command is similar to the plot(x. and the y −axis as a linear scale. • We use the MATLAB command plot(x. number_of_values) specifies the number of data points but not the increments between data points.y) command. typically power. • The ratio of any two values of the same quantity. • The semilogx(x. that is. and displays a cross−hair which can be moved around with the mouse. This format specifies the increments between points but not the number of data points. and the x −axis as a linear scale. The command text(x. and string is the label which we want to place at that location. except that the y −axis is represented as a log scale. Normally. • The function log(x) in MATLAB is the natural logarithm.y) command uses logarithmic scales for both axes. whereas the common logarithm is expressed as log10(x). or a row vector is too long to fit in one line. Third Edition Copyright © Orchard Publications . the semilogy(x. The loglog(x.y) to make two−dimensional plots. as bn x + bn – 1 x + bn – 2 x + … + b1 x + b0 (x) = -----------------------------------------------------------------------------------------------------------------------R ( x ) = Num -------------------m m – 1 m – 2 Den ( x ) am x + am – 1 x + am – 2 x + … + a1 x + a0 where some of the terms in the numerator and/or denominator may be zero. For a three−dimensional plot. we can also use the zlabel(‘string’) command. P out dB = 10 log --------P in • The command gtext(‘string’) switches to the current Figure Window. except that the x −axis is represented as a log scale.y) command. we express the numerator and denominator of a rational function as a combination of linear and quadratic factors.y) command is similar to the plot(x.Chapter 1 Introduction to MATLAB • Rational Polynomials are those which can be expressed in ratio form. and y is the vertical axis (ordinate). xlabel(‘string’). • We can make a two−dimensional plot more presentable with the commands grid.

or on the same line without first executing the plot command. • The mesh(x.z. whereas element−by−element multiplication is performed with the dot multiplication operator (. We observe that x corresponds to the columns of Z . j ) } . Similarly. y. It must be placed after the plot(x.y) function which creates the matrix X whose rows are copies of the vector x. n ] = size ( Z ) . y . Third Edition Copyright © Orchard Publications 1−35 . no markers are drawn unless they are selected. Numerical Analysis Using MATLAB® and Excel®. and y corresponds to the rows of Z . This command divides the window into an m × n matrix of plotting areas and chooses the pth area to be active. in a three−dimensional plot we can set the limits of all three axes with the axis([xmin xmax ymin ymax zmin zmax]) command.Summary • MATLAB has no default color. Likewise.y. y ) .’string’) command will place string beginning at the co−ordinate ( x.2 × 10 – 16 .n). ymin and ymax. • The command axis([xmin xmax ymin ymax]) scales the current plot to the values specified by the arguments xmin. contour(Z. In this case.z) command displays a three−dimensional plot. • To avoid division by zero. and z are three vectors of the same length. there is no default marker.y.y.z) command with two vector arguments.^) are used for element−by−element division and exponentiation. • In a two−dimensional plot.p). whereas dot division (. we can set the limits of the x − and y −axes with the axis([xmin xmax ymin ymax]) command. We can also use the mesh(x. matrix multiplication (multiplication of row by column vectors) is performed with the matrix multiplication operator (*). We can generate the matrices X and Y with the [X. z ) on the plot. xmax.Y]=meshgrid(x. The default line is the solid line.z) commands.n. There are no commas between these four arguments. • The plot3(x. • MATLAB can display up to four windows of different plots on the Figure window using the command subplot(m. Z ( i. The three − dimensional text(x.y.z) command plots a line in 3−space through the points whose coordinates are the elements of x .*). y. These must be defined as length ( x ) = n and length ( y ) = m where [ m. This must be done for each plot.y) or plot3(x. draws contour lines for n levels. where x. we must first generate the X and Y matrices which consist of repeated rows and columns over the range of the variables x and y . • With MATLAB. we use the special MATLAB function eps. and z . the division (/) and exponentiation (^) operators. y ( i )./) and dot exponentiation (. say z = f ( x. which is a number approximately equal to 2. To produce a mesh plot of a function of two variables. the vertices of the mesh lines are the triples { x ( j ). Also.y. and the matrix Y whose columns are copies of the vector y. This command must be placed after the plot command and must be repeated for each plot. are used for matrix division and exponentiation. Another command. it starts with blue and cycles through seven colors.

m.lims) command plots the function specified by the string fcn between the x −axis limits specified by lims = [xmin xmax]. Both types are referred to as m−files. • The MATLAB fplot(fcn. and the angle(z) functions compute the absolute value (magnitude) and phase angle of the complex quantity z = x + iy = r –θ . It uses specifiers to indicate where and in which format the values would be displayed and printed.x2) function minimizes a function of one variable. • The MATLAB fmin(f. and the abs(z). where r is the magnitude. Third Edition Copyright © Orchard Publications . The format displayed has nothing to do with the accuracy in the computations. • MATLAB recognizes two types of files: script files and function files. Generally. With these commands only the real part of each parameter is processed. • MATLAB displays the results on the screen in integer format without decimals if the result is an integer number.Chapter 1 Introduction to MATLAB • The real(z) and imag(z) functions display the real and imaginary parts of the complex quantity z = x + iy. and if %e is used. but the file name must have the extension . the values will be displayed and printed in fixed decimal format. the values will be displayed and printed in scientific notation format.x1. and theta is the angle in radians. The function name and file name must be the same. A function file is a user−defined function using MATLAB.array) command used above displays and prints both text and arrays. A script file consists of two or more built−in functions. It attempts to return a value of x where f ( x ) is minimum in the interval x 1 < x < x 2 . MATLAB performs all computations with accuracy up to 16 decimal places.r) function produces a plot in polar coordinates. Thus. a script file is one which was generated and saved as an m−file with an editor such as the MATLAB’s Editor/ Debugger. followed by the output argument. • The MATLAB fprintf(format. and the input argument enclosed in parentheses. or in short floating point format with four decimals if it a fractional number. The first line of a function file must contain the word function. The polar(theta. if %f is used. the equal sign ( = ). Using lims = [xmin xmax ymin ymax] also controls the y −axis limits. The string fcn must be the name of an m−file function or a string with variable x . We use function files for repetitive tasks. 1−36 Numerical Analysis Using MATLAB® and Excel®. The string f contains the name of the function to be minimized.

5 1200 102. Use MATLAB to derive the polynomials having the following roots: a. – 6.75 4.000 and three roots at x = – 3. Numerical Analysis Using MATLAB® and Excel®.5708 – 0. Use MATLAB to evaluate the polynomials below at the specified values. p ( x ) = x + 8x + 10x + 4 b.8 1600 94.000 3. the applied voltage V S was kept constant and the voltage V C across the capacitor was measured and recorded at several frequencies as shown on the table below.0 1300 100. Third Edition Copyright © Orchard Publications 1−37 .9 Plot V C (in dB scale) versus ω (in common log scale) and label the axes appropriately.9 900 105.3132 5 4 3 2 3 2 b. R1 R2 VS C L VC 5 4 3 2 3 2 Capacitor voltage versus radian frequency ω VC ω VC 500 88.Exercises 1.3132 – 0.7146 + j0. Two roots at x = – 2.9 1100 103.5 1000 104.9 1400 98. In the electric circuit below.12 Exercises 1.7146 – j 0.4 700 103. Use MATLAB to compute the roots of the following polynomials: a. p ( y ) = y + 7y + 19y + 25y + 16y + 4 at y = – 3.25 b.3 1500 96. p ( x ) = x + 8x + 10x + 4 at x = 1.7 800 104. p ( y ) = y + 7y + 19y + 25y + 16y + 4 2. a.8 600 98.

make an honest effort to find the solutions to the exercises without first looking at the solutions that follow. For the exercises that you are uncertain. You should follow this practice with the rest of the exercises of this book.Chapter 1 Introduction to MATLAB 1. 1−38 Numerical Analysis Using MATLAB® and Excel®.13 Solutions to End−of−Chapter Exercises Dear Reader: The remaining pages on this chapter contain the solutions to the exercises. Refer to the solutions as a last resort and rework those exercises at a later date. It is recommended that first you go through and work out those you feel that you know. You must. for your benefit. Third Edition Copyright © Orchard Publications . review this chapter and try again.

25) value = 30. −3. value=polyval(Pv.0. ans = -6. poly_r2=poly(r2) poly_r2 = 1 5 13 4 67 3 171 2 216 108 p ( z ) = z + 13z + 67z + 171z + 216z + 108 3. Pw=[1 7 19 25 16 4].7146−0. a.0000 -1.3132i -0. r2=[−2 −2 −3 −3 −3].9997 4. poly_r1=poly(r1) poly_r1 = 3 1. Pv=[1 8 10 4].3132j −0.5708 −0. Px=[1 8 10 4].3132j].7146 + 0.0000 + 0.0000 p ( x ) = x + 8x + 10x + 4 b.5708 -0.0000 9. 1.0000 -2.7146+0. value=polyval(Pw. a.0000 .9531 b.Solutions to End−of−Chapter Exercises 1.0000 2 8. roots(Px) b.0.3132i Py=[1 7 19 25 16 4].0000i -1. r1=[−6. Third Edition Copyright © Orchard Publications 1−39 .7146 .0000i 2. a.0000 -1.75) value = -63. roots(Py) ans = -2.6904 Numerical Analysis Using MATLAB® and Excel®.

8 103.4 100.9 98. w 10 w in rads/sec 3 10 4 1−40 Numerical Analysis Using MATLAB® and Excel®. dB=20*log10(Vc). grid.Chapter 1 Introduction to MATLAB 4. ylabel('|Vc| in volts') 40.7 98.. semilogx(w..9 96.5 103 104. xlabel('w in rads/sec').9]. title('Magnitude of Vc vs. w').8 102.5 2 10 Magnitude of Vc vs. Vc=[88.3 104. w=[5 6 7 8 9 10 11 12 13 14 15 16]*100.5 94..5 39 38. Third Edition Copyright © Orchard Publications .9 105.dB).5 40 |Vc| in volts 39.

0 – f ( x1 ) = f ' ( x1 ) ( x2 – x1 ) f ( x1 ) x 2 = x 1 – -------------f ' ( x1 ) (2.1).1. f ( x 1 ) } { x 1. By substitution. Since this point [ x 2. Then. f ( x 1 ) } x ( x 2. Several examples are presented to illustrate practical solutions using MATLAB and Excel spreadsheets. Newton’s method for approximating real roots of a function We assume that the slope is neither zero nor infinite. f ( x 2 ) ] = ( x 2.2) and in general.Chapter 2 Root Approximations T his chapter is an introduction to Newton’s and bisection methods for approximating roots of linear and non−linear equations. y y = f(x ) • • Tangent line (slope) to the curve y = f ( x ) at point { x 1.1 Newton’s Method for Root Approximation Newton’s (or Newton−Raphson) method can be used to approximate the roots of any linear or non−linear equation of any degree.3) Numerical Analysis Using MATLAB® and Excel®.1) The slope crosses the x – axis at x = x 2 and y = 0 . 2. This is an iterative (repetitive procedure) method and it is derived with the aid of Figure 2. it satisfies (2. 0 ) lies on the slope line.1. the slope (first derivative) at x = x 1 is y – f ( x1 ) f ' ( x 1 ) = -------------------x – x1 y – f ( x1 ) = f ' ( x1 ) ( x – x 1 ) (2. 0 ) Figure 2. Third Edition Copyright © Orchard Publications 2−1 . f ( xn ) x n + 1 = x n – -------------f ' ( xn ) (2.

x) function evaluates the polynomial p at some value x. and waits for an input from the user. c. G. we must start with a reasonable approximation of the root value.Chapter 2 Root Approximations Example 2. Knowing the polynomial p and the first approximation x 0 .\b. We start with MATLAB and will discuss the steps for using a spreadsheet afterwards. Therefore.y). We will now introduce some new MATLAB functions and review some which are discussed in Chapter 1. optional width and precision fields.x0)/polyval(q. under control of the specified format string. o. we can compute the next iteration for approximating a root with Newton’s method using these functions. 4. To apply Newton’s method. and conversion characters d. x.x0) We’ve used the fprintf command in Chapter 1. Use \\ to produce a backslash character and %% to produce the percent character. grid % Specifies 100 values between -4 and 4 % The dot exponentiation is a must 2−2 Numerical Analysis Using MATLAB® and Excel®. we will use it many more times.\t. e. This can be done easily with a simple plot using MATLAB or a spreadsheet. q=polyder(p) x1=x0−polyval(p.1 Use Newton’s method to approximate the positive root of to four decimal places. optional subtype specifier. input(‘string’): It displays the text string. and formfeed characters respectively. Therefore. u. we plot the curve of (2. Solution: As a first step. g. let us review it again. The following two lines of script will display the graph of the given equation in the interval – 4 ≤ x ≤ 4 . Third Edition Copyright © Orchard Publications . plot(x.\r. this can best be done by plotting f ( x ) versus x with the familiar statements below. f. X. and writes it to the file associated with file identifier fid and contains C language conversion specifications. y=x . E. It formats the data in the real part of matrix A (and in any additional matrix arguments). The special formats \n. carriage return.\f can be used to produce linefeed. backspace. We must enclose f ( x ) = x2 – 5 (2. i. These specifications involve the character %. See the Language Reference Guide or a C manual for complete details. x=linspace(−4. We recall that the polyder(p) function displays the row vector whose values are the coefficients of the first derivative of the polynomial p. optional flags. The polyval(p. 100).4) the text in single quotation marks.^ 2 − 5. The following description was extracted from the help fprintf function.4) to find out where it crosses the x – axis . tab. we can use the following script for the next approximation x 1 . In all cases. and s.

observe the x – axis crossings. where the interval may not be so obvious. and we compute the next value x 1 as 2 f ( x0 ) ( –1 ) (2) – 5 .= 2. so we take x 0 = 2 as our first approximation. The plot is shown in Figure 2. fprintf('\n').e.x0)/polyval(q. % Calculates the derivative of p(x) x1=x0−polyval(p. i.6) We will use the following MATLAB script to verify (2.1 As expected. Plot for the curve of Example 2.25 – ( 4.25 ) f ' ( x1 ) (2..= 2.xxxxxx. we expect this value to be a value between 2 and 3 . and then redefine the interval.= 2. For other functions.= 2 – ---------. and Numerical Analysis Using MATLAB® and Excel®.25 ) – 5 --------------. 15 10 5 0 -5 -4 -2 0 2 4 Figure 2. x0=input('Enter starting value: ').x0).0625 x 2 = x 1 – -------------.2361 ------------------------.25 .6f. % Approximation of a root of a polynomial function p(x) % Do not forget to enclose the coefficients in brackets [ ] p=input('Enter coefficients of p(x) in descending order: ').= 2. with 9 digits where 6 of these digits % are to the right of the decimal point such as xxx.5 2 ( 2. q=polyder(p). Third Edition Copyright © Orchard Publications 2−3 . the curve shows one crossing between x = 2 and x = 3 . we can choose a larger interval.5) and (2.= 2 – -----------------x 1 = x 0 – -------------f ' ( x0 ) 2(2) 4 (2.Newton’s Method for Root Approximation We chose this interval because the given equation asks for the square root of 5 .2.5) The second approximation yields 2 f ( x1 ) 2. % Inserts a blank line % % The next function displays the value of x1 in decimal format as indicated % by the specifier %9.25 – 0.6).2.

In Chapter 1 we discussed script files and function files. % Inserts another blank line % fprintf('Rerun the program using this value as your next. the function file consisting of the two lines below function y = myfunction(x) y=x . x1) fprintf('\n'). whose general form is while expression commands ..6).250000 Rerun the program using this value as your next approximation Enter polynomial coefficients in descending order: [1 0 −5] Enter starting value: 2. in the second line are executed as long as all elements in expression of the first line are true. end where the commands .^ 3 + cos(3 . The function name and file name must be the same but the file name must have the extension . the equal sign (=).* x) is a function file and must be saved as myfunction. For example... Enter coefficients of P(x) in descending order: [1 0 −5] Enter starting value: 2 The next approximation is: 2.Chapter 2 Root Approximations % \n prints a blank line before printing x1 fprintf('The next approximation is: %9. We recall that a function file is a user− defined function using MATLAB.. approximation \n')... and the input argument enclosed in parentheses..m We will use the while end loop. The following lines show MATLAB’s inquiries and our responses (inputs) for the first two approximations. The first line of a function file must contain the word function followed by the output argument.6f \n'.25 The next approximation is: 2. Third Edition Copyright © Orchard Publications .236111 We observe that this approximation is in close agreement with (2. We use function files for repetitive tasks.m. We will also be using the following commands: 2−4 Numerical Analysis Using MATLAB® and Excel®.

sprintf('%d'. y=x . 100). Example 2.round(pi)) ans = 3 where the format script %d specifies an integer. the text is displayed. Likewise. function y=funcnewt01(x) % Approximating roots with Newton's method y=x .pi) ans = 3. This takes us to the Editor Window where we type the following three lines and we save it as funcnewt01. For example. pi.A): Formats the data in the real part of matrix A under control of the specified format string. plot(x. disp(v) displays 12. we generate the function funcnewt01 and we save it as an m−file. Third Edition Copyright © Orchard Publications 2−5 . The plot shows that one real root is approximately at x = – 1 . from the File menu of the command window.* cos(x). Next.* cos(x).Newton’s Method for Root Approximation disp(x): Displays the array x without printing the array name. we choose New and click on M−File.* x + 3 + sin(x) − x . If x is a string.3f'. To save it. we sketch the curve to find out where the curve crosses the x – axis . Numerical Analysis Using MATLAB® and Excel®.* x + 3 + sin(x) − x . if v = 12 .142 where the format script %4. Solution: As a first step.3. and disp(‘volts’) displays volts. so we will use this value as our first approximation. For example.7) to four decimal places using Newton’s method. sprintf(format. sprintf('%4. We generate the plot with the script below.m. x=linspace(−pi. grid The plot is shown in Figure 2.3f specifies a fixed format of 4 digits where 3 of these digits are allocated to the fractional part.2 Approximate one real root of the non−linear equation f ( x ) = x + 4x + 3 + sin x – x cos x 2 (2.^ 2 + 4 .y).^ 2 + 4 .

function y=funcnewt02(x) % Finding roots by Newton's method % The following is the first derivative of the function defined as funcnewt02 y=2 . Third Edition Copyright © Orchard Publications . To save it.m. x = input('Enter starting value: ').Chapter 2 Root Approximations 30 20 10 0 -10 -4 -3 -2 -1 0 1 2 3 4 Figure 2. however. we choose New and click on M−File. we can let MATLAB do the differentiation. y1=diff(y) % Dot operations are not necessary with % symbolic expressions.* sin(x). This function performs differentiation of symbolic expressions. Our script for finding the next approximation with Newton’s method follows. just as a check. The syms function is used to define one or more symbolic expressions. xnext = x−fx/fprimex. and to introduce the diff(s) function. and we save it as m−file. Plot for the equation of Example 2. % Find the derivative of y y1 = 2*x+4+x*sin(x) Now. we generate the function funcnewt02. This takes us to the Editor Window where we type these two lines and we save it as funcnewt02.2 We also need the first derivative of y. fprimex = funcnewt02(x). syms x y = x^2+4*x+3+sin(x)−x*cos(x). This is y' = 2x + 4 + x sin x The computation of the derivative for this example was a simple task. but correct % answer will be displayed if they are used. fx = funcnewt01(x). x = xnext. from the File menu of the command window.* x + 4 + x . 2−6 Numerical Analysis Using MATLAB® and Excel®.3.

The string f contains the name of a real−valued function of a single real variable.894010 Next approximation? (<enter>=no.x) function. or returns NaN if the search fails. xnext=x−fx/fprimex.1=yes)').1=yes) <enter> We can also use the fzero(f. This function tries to find a zero of a function of one variable. MATLAB searches for a value near a point where the function f changes sign and returns that value. disp(sprintf('First approximation is x = %9. will be those of Excel. we will go through several examples to illustrate the procedure of using a spreadsheet such as Excel* to approximate the real roots of linear and non−linear equations. and Quattro can also be used. x)) while input('Next approximation? (<enter>=no. As we recall. x)) MATLAB produces the following result with – 1 as a starting value. it is included here for convenience. disp(sprintf('Next approximation is x = %9. fprimex=funcnewt02(x). A cubic equation of the form y + py + qy + r = 0 3 2 (2.6f \n'. Enter starting value: −1 First approximation is: -0. 2.895225 Next approximation? (<enter>=no. all spreadsheet commands and formulas that we will be using. Third Edition Copyright © Orchard Publications 2−7 .2 Approximations with Spreadsheets In this section.Approximations with Spreadsheets fx = funcnewt01(x). fprimex = funcnewt02(x).9) can be reduced to the simpler form x + ax + b = 0 3 where * We will illustrate our examples with Excel. fx=funcnewt01(x).8) (2. x=xnext. disp(sprintf('%9. Henceforth. It was introduced in Chapter 1. although others such as Lotus 1−2−3.6f \n'. x)) end.6f \n'. Numerical Analysis Using MATLAB® and Excel®.1=yes)1 -0. We recall that there is a standard procedure for finding the roots of a cubic equation.

we can use a spreadsheet to find the real root since in any cubic equation there is at least one real root. This procedure is illustrated with the examples that follow.+ ----B = – ----. Also.+ ---4 27 2 3 2 3 (2. and so on. For real roots.11) Then.12) (2. we can plot y versus x to observe the values of x that make f ( x ) = 0 . Third Edition Copyright © Orchard Publications .13) If the coefficients p . Therefore. and r are real. it will be understood that the <enter> key was pressed. cell C11. then b .–3 .< 0 the roots will be real and unequal If ---4 27 a b .= 0 there will be three real roots with at least two equal If ---4 27 2 3 2 3 2 3 While MATLAB handles complex numbers very well. spreadsheets do not.10) For the solution it is convenient to let 3 A = a –b b . Thus B3.–3 x 3 = – -------------– 2 2 (2.+ ---4 27 2 2 3 a –b b .( 2p 3 – 9pq + 27r ) b = ----27 (2. y = f ( x ) = x – 7x + 16x – 12 3 2 (2. after an entry has been made.( 3q – p 2 ) a = -3 1 .Chapter 2 Root Approximations p x = y + -3 1 .14) 2−8 Numerical Analysis Using MATLAB® and Excel®. Then. However. we can use a spreadsheet to define a range of x values with small increments and compute the corresponding values of y = f ( x ) . the values of x for which the cubic equation of (2.+ --------. unless we know that the roots are all real.> 0 one root will be real and the other two complex conjugates If ---4 27 a b .+ ------------x 2 = – ------------2 2 A–B A + B ------------. Note: In our subsequent discussion we will omit the word cell and the key <enter>.+ ----.11) is equal to zero are x1 = A + B –B A+B A . C11.+ ----. and so on will be understood to be cell B3.3 Compute the roots of the polynomial using Excel. Example 2.+ ----a . q . we should not use a spreadsheet to find the roots of a cubic equation by substitution in the above formulas.

a pop−up note shows the cell entry for the last value in the range. Next. If it does not appear on the spreadsheet. we point the mouse to the fill handle and we observe that the mouse pointer appears as a small cross. We repeat these steps for Column B and we choose 3 decimal places. a selected cell is surrounded by a heavy border. We select a cell by moving the thick hollow white cross pointer to the desired cell and we click. we type the equation formula with the = sign in front of it. that is. and we click on Copy. we drag the mouse down to A3 so that both these two cells are highlighted. and the numbers in Column B will be fixed with three decimal places. select the Edit tab. we click on Edit on the main taskbar. We observe that B2 displays the value – 12. * To use this feature.05 in A3. We click. Alternately. This is the value of f ( x ) when x = 0. For this example. Next. that is. We observe that on the lower right corner of A3. We observe that this number is displayed exactly as it was typed.Approximations with Spreadsheets Solution: We start with a blank worksheet. When properly done. we release the mouse button. from the Tools drop menu of the Menu bar. we highlight cells A2 and A3. we type = A2^3-7*A2^2 + 16*A2-2 where A2 represents the first value of x = 0. we select B1. this becomes the next selected cell. we want to copy this formula to the range B3:B102 (the colon : means B3 through B102). We observe that after pressing the <enter> key. We will enter more values in column A. Third Edition Copyright © Orchard Publications 2−9 . Next.05 increments up to 5. as we drag the fill handle. and we observe that this range is now highlighted. Next. Column A now contains 100 values of x from 0. the next cell moves downwards to A2. this is called the fill handle.00. With B2 still selected.00 in increments of 0.00 in A2. We type 0. all numbers that we will type in Column A will be fixed numbers with two decimal places. We observe that.00 to 5.000 . and we release the mouse button.00 . hold down the mouse button. We click on Edit. and to make all values look uniform. We observe that the entire column is now highlighted. we type 0. In an Excel worksheet. We observe that this value is displayed just as 0. Then. To continue. we first select A1 and we type x. We click on the Fixed Decimal check box to place a check mark and we choose 2 as the number of decimal places. then on Paste and we observe that this range is now filled with the values of f ( x ) . We will use values in 0. Numerical Analysis Using MATLAB® and Excel®. We select the range B3:B102 with the mouse. the background on the monitor has changed from white to black.00 Next. and place a check mark on the Drag and Drop setting. we click and holding the mouse left button down.05. we click on letter A on top of column A. then we release the mouse button. there is a small black square. and we type f(x). we choose Options and we click on the Edit tab. we can use the Copy and Paste icons of the taskbar. we can make it visible by performing the sequential steps Tools>Options. we drag it down to A102. that is. We will now use the AutoFill* feature to fill−in the other values of x in Column A. In B2. we select A2. A2 will have a white background but A3 will have a black background. without decimals.

25 -8.0 Major unit: 1.30 -7. We will modify this plot to make it more presentable.422 0. on the main taskbar and below it in a box where next to it is another small box with the hand icon. on the Format axis menu. We click on the Series 1 box to select it. Then.4 appears.05 -11. Then. and we delete it by pressing the Delete key. This is done with the following steps: 1. We click on the Patterns tab to select it. and we place a check mark on Chart.126 0.151 f(x) 20 15 10 5 0 -5 -10 -15 0 1 2 3 4 5 6 Figure 2.00 -12. From the Chart menu box (below the main taskbar).217 0. Next. and we click on Low on the Tick mark labels (lower right box). when it is enclosed in black square handles.625 0. only when the graph box is selected. Regular style.40 -6. x f(x) 0. Plot of the equation of Example 2. we see it enclosed in six black square handles. Note: The Chart menu appears on the main taskbar and on the box below it. From the View menu. 2. A chart similar to the one on Figure 2. we click on Next. and Finish.5 We click on the Number tab. and we click on the small box next to it (the box with the hand icon). we click on the Scale tab and we make the following entries: Minimum: 0. we choose the one on top of the right side (the smooth curves without connection points). We click on OK to return to the graph. we select Number from the Category column. the roots of f ( x ) .803 0. The Chart menu appears in two places. and we type 0 in the Decimal places box. and to see more precisely the x – axis crossing(s).50 -5. From the Chart menu box we select Value (Y) axis and we click on the small box next to it (the 2−10 Numerical Analysis Using MATLAB® and Excel®. We click on the Font tab.3. Then.072 0.656 0.469 0. we click on Toolbars.754 0.55 -5.000 0.15 -9. Third Edition Copyright © Orchard Publications .0 Minor unit: 0.10 -10. we select any font. We click anywhere inside the graph box. that is. Size 9.35 -7. we click on the Chart Wizard icon of the Standard Toolbar. Next. that is. 3.215 0.Chapter 2 Root Approximations To plot f ( x ) versus x .45 -6. and on the Chart type column we click on XY (Scatter). we select Value (X) axis.0 Maximum: 5.20 -9. From the displayed charts.4.

and we click on Outside on the Major tick mark type (upper right box). we click on the white square which is immediately below the gray box. The given polynomial has two roots at x = 2 .Approximations with Spreadsheets box with the hand icon). The plot now resembles the one shown in Figure 2. we place check marks on Major gridlines of both Value (X) axis and Value (Y) axis. we select Chart Area from the Chart menu.xls is the default extension for file names saved in Excel. Now. we click on the Scale tab. we click on the Titles tab and we make the following entries: Chart title: f(x) = the given equation (or whatever we wish) Value (X) axis: x (or whatever we wish) Value (Y) axis: y=f(x) (or whatever we wish) 5.0 Major unit: 0.0 Maximum: 1. and on the Area side of the Patterns tab.25 Minor unit: 0. 4. to change Chart Area square corners to round. Size 9. say poly01. We click on the box next to it (the box with the hand icon). We click on the Font tab. and on the Chart Options. and we select 2 in the Decimal places box. we click at any point near it and we observe that several black square handles appear along the curve. therefore. Then.xls where. 7. The plot area on the chart now appears on white background. 6. Numerical Analysis Using MATLAB® and Excel®. We click on OK to return to the graph. We click on the small box next to it. On the Format axis menu. it is highly recommended that this file is saved with any name. From the Chart menu box below the main task bar. and the third root is x = 3 . We click on the Patterns tab. we select Plot Area and we observe that the gray background of the plot area is surrounded by black square handles. select any font. We click on Chart on the main taskbar. Finally. and on the Patterns tab we place a check mark on the Round corners box. From the Weight selections we select the first of the thick lines. Third Edition Copyright © Orchard Publications 2−11 . and on the Patterns tab. and we make the following entries: Minimum: −1. Regular style. we select Number from the Category column.05 We click on the Number tab.5 where we have shown partial lists of x and f ( x ) . To make the line of the curve f ( x ) thicker. Series 1 appears on the Chart menu box. we will change the background of the plot area from gray to white. We will follow the same procedure for generating the graphs of the other examples which follow. We click on Gridlines.

887 0.002 3.081 1.00 0.xls. poly01.50 0. This is done by first opening the file poly01.516 0.95 -0. and save it with another name such as poly02.25 -0.656 0.25 -8.803 0.3.541 0. Example 2.422 0.000 f(x) = x3 .283 0.20 -9.215 0.217 0.168 0.90 -2.15) 2−12 Numerical Analysis Using MATLAB® and Excel®.009 3.121 f(x) =0 at x=2 (double root) and at x=3 Figure 2.000 3. then.50 -5.35 -7.70 -3.00 0.5.45 -6.625 0. we perform the following steps: 1. we choose the Save as option.xls.00 -12.754 0.15 -9.75 0.55 -5.843 0.12 1.40 -6.10 -10. the highest power of the polynomial is 5 (odd number).30 -7. the spreadsheet of the previous example still exists as poly01.85 -2.000 2.65 -4.50 -0.072 0. and since we know that complex roots occur in conjugate pairs.05 -0.011 2.Chapter 2 Root Approximations x f(x) 0. Modified plot of the equation of Example 2.7x2 + 16x .00 0.003 2.704 0. we arbitrarily y = f ( x ) = 3x – 2x + 6x – 8 5 3 (2. or any other name.60 -4.000 0. Since we do not know where a real root is in the x−axis interval. we invoke (open) the spreadsheet of the previous example.10 -0.05 0. Solution: To save lots of unnecessary work.95 -2.90 -0.75 -3. that is.00 0 1 2 3 4 5 f(x) x x f(x) x f(x) 1.xls.xls (or any other file name that was assigned to it).95 -0.260 1. Third Edition Copyright © Orchard Publications .80 -3.4 Find a real root of the polynomial using Excel. we save it as poly02.10 0.126 0. and from the File drop down menu. For this example. When this is done.00 -2.469 0.xls. we expect that this polynomial will have at least one real root.045 Roots 2.25 0.75 -1.151 0.00 -0.90 -0. Next.05 -11.055 2.

6. reveal some useful information. and we press the Delete key. Partial table for Example 2. Then. we fill−in the range A4:A22. =3*A2^5−2*A2^3+6*A2−8 We copy this formula to B3:B22. On the View menu. on the Format axis menu. A2 contains 1. Next.01 Numerical Analysis Using MATLAB® and Excel®. A part of the table is shown in Figure 2.02 Minor unit: 0.000 175735. we click on Toolbars and place a check mark on Chart. From the Chart menu box (below the main taskbar) we select Value (X) axis.000 -8. We observe that the chart has changed shape to conform to the new data.1 Major unit: 0.00 10. and so on. To obtain a good approximation of the real root in that interval.10 .000 -1.00 9.00 f(x) -298068. and we click on the small box next to it (the box with the hand). and we have the interval from −10 to 10 in increments of 1. Let us then redefine our interval of the x values as 1 ≤ x ≤ 2 in increments of 0. are now displayed.00 0.000 84. and thus we expect that a real root exists between x = 1.Approximations with Spreadsheets choose the interval – 10 ≤ x ≤ 10 .4 This table shows that f ( x ) changes sign somewhere in the interval from x = 1 and x = 2 . and the plot shows that the curve crosses the x−axis somewhere between x = 1 and x = 2. x -10. and we observe the square handles surrounding it.05. Our spreadsheet now shows that there is a sign change from B3 to B4.000 -175751. We select the graph box by clicking inside it.05 and x = 1. We must now delete all rows starting with 23 and downward. we enter −10 and −9 in A2 and A3 respectively.00 2. Columns A (values of x).00 -9. Third Edition Copyright © Orchard Publications 2−13 . The Chart menu on the main taskbar and the Chart menu box below it.05.e..00. Columns A and B now contain values of x and f ( x ) respectively. 2.000 Sign Change Figure 2. Using the AutoFill feature. to get better approximations. When this is done A1 contains 1.0 Maximum: 1.00 1. we perform Steps 2 through 4 below.000 298052. and B (values of f ( x ) ). we click on the Scale tab and make the following entries: Minimum: 1. Now we select B2 where we enter the formula for the given equation. We do this by highlighting the range A23:B102.6. i.

047 13.021 47.5 Minor unit: 0.970 41.02 1. we suspect that the others are complex conjugates.770 1.85 1.30 1. roots_p=roots(p) 2−14 Numerical Analysis Using MATLAB® and Excel®.8 1.20 f(x) = − 0.455 84.031 20.90 1. on the Format axis menu.447 63.631 11.823 55.00 1. Then. Graph for Example 2.7 and we see that one real root is approximately 1. x 1.2x 3 + 6x .545 8.50 -1.10 1. We click on the Titles tab and make the following entries: Chart title: f(x) = the given equation (or whatever we wish) Value (X) axis: x (or whatever we wish) Value (Y) axis: y=f(x) (or whatever we wish) Our spreadsheet now should look like the one in Figure 2.209 4.40 1.95 2.80 1.00 -0.08 1.05 1.832 17.0 Maximum: 1.692 24.186 0.00 1. From the Chart menu we select Value (Y) axis.892 3.06 f(x) -1.000 f(x) = 3x 5 .000 -0.05 Real Root between 1.965 73.15 1.0 Major unit: 0. and we click on the small box next to it.45 1.06 1.65 1. we click on the Scale tab and make the following entries: Minimum: −1.4 Since no other roots are indicated on the plot.60 1. We confirm this with MATLAB as follows: p = [ 3 0 −2 0 6 −8].10 x x 1.209 Figure 2.04 1.1 4.00 f(x) -1.70 1.25 1.55 1.15 1.865 29.770 1.007 at x = 1.892 3.50 f(x) 0.50 1.000 -0.20 1.186 0.35 1.605 34.75 1.10 1. Third Edition Copyright © Orchard Publications .7.Chapter 2 Root Approximations 3.00 0.00 1.749 6.00 1.06.

Approximations with Spreadsheets roots_p = -1.6113 0. we expect two more real roots.138 -0.1415 -1. we will find that the other two roots are approximately x = 1. If we redefine the x – axis range as 1 to 2.30 interval and the other in the 2. Since we do not know where real roots (if any) are in the x−axis interval.194 Sign Change Sign Change We observe two sign changes.041 -0.90 in A2 and A3 respectively.00 and −0.1415 0.30 f(x) 0.9476i 0. Therefore. this is found by substitution of zero into the given equation.24 .5 Compute the real roots of the trigonometric function using Excel.xls or poly02. we select B2 and we enter the formula for the given equation.20 2. We observe that Columns A and B contain the following sign changes (only a part of the table is shown): x 1.9476i (2. such as poly03.20 ≤ x ≤ 1.8 where the curve crosses the x – axis .20 1. Using the AutoFill feature.16) and we copy this formula to B3:B62. Solution: We invoke (open) the spreadsheet of one of the last two examples.20 ≤ x ≤ 2. i. that is. we enter −1. =COS(2*A2)+SIN(2*A2)+A2−1 y = f ( x ) = cos 2x + sin 2x + x – 1 + + - 0.e. we arbitrarily choose the interval – 1 ≤ x ≤ 6 .10.5.6113 1. we fill−in the range A4:A72 and thus we have the interval from −1 to 6 in increments of 0. one in the 1.8212i 0.xls. and save it with another name. Then. Next.30 2.. Numerical Analysis Using MATLAB® and Excel®. There is a root at x = 0 . Approximate values of these roots can also be observed on the plot of Figure 2.8212i 0.30 and x = 2.xls.0604 Example 2. poly01. Third Edition Copyright © Orchard Publications 2−15 .30 interval.059 0.

we make the following entries: Set cell: K1 To value: 0 By changing cell: J1 3.30 2. find the value of the independent variable x which satisfies that goal.138 -0.8.80 0. we will use it to find better approximations for the non−zero roots of Example 2.60 0. but we do not know the input value which satisfies that result. Thus. In the last three examples our goal was to find the values of x for which y = f ( x ) = 0 .70 -0.5.70 0.829 -2. so that J1 contains 1. We increase the accuracy of Columns J and K to 5 decimal places by clicking on Format.319 0.279 0. say J1 and K1.10 0. From the Tools drop menu. Third Edition Copyright © Orchard Publications . To illustrate the Goal Seek feature.325 -3.041 if range A25:B25 was copied).90 f(x) -3.20 1.90 -0. We do this with the following steps: 1.1 6 4 f(x) 2 0 -2 -4 -1 0 1 2 3 4 5 6 x Real Root at Real Root between Real Root between x 0. Cells.20 -0. Graph for Example 2.000 0. we will observe the changes in J1 and K1.5 We can obtain more accurate approximations using Excel’s Goal Seek feature.421 -1.041 -0.138 (or 1.30 0. We use Goal Seek when we know the desired result of a single formula.40 0. 2. Numbers tab.690 0.20 and K1 contains 0.60 -0.000 0.510 0. These indicate that for 2−16 Numerical Analysis Using MATLAB® and Excel®.515 -2.30 -0.894 0.059 0.00 -0.30 and −0.Chapter 2 Root Approximations x -1.039 -0. When this is done properly. and when the Goal Seek dialog box appears.770 0. if we have the function y = f ( x ) .50 -0.80 -0.50 0.10 0.00 1.40 -0.194 Figure 2.647 f(x) = cos2x + sin2x + x .801 -1.30 f(x) 0. we can use Goal Seek to set the dependent variable y to the desired value (goal) and from it.101 -2.20 0. we click on Goal Seek.20 2.00 0.814 0.882 0. We copy range A24:B24 (or A25:B25) to two blank cells.170 -1.668 -0.855 0.

9. in the To value box we enter 0. This is illustrated with the next example. then. 1. The Goal Seek dialog box then appears where the Set cell shows B24.5. which uses Newton’s method to approximate a real root of any polynomial with real coefficients up to the seventh power. A square box appears and displays Series 1. Example 2. Another method of using the Goal Seek feature. we click on the handle near the x = 1. such as finding the roots of polynomials. We click anywhere near the curve. We will illustrate the procedure with the chart of Example 2.28 and B24 displays 0.27647 . Next.30 point.00002 . 2.Approximations with Spreadsheets x = 1. * There exists a numerical procedure.30 point. Solution: 1. y = f ( x ) = 0. We observe now that A24 displays 1. we drag it towards the x−axis to change its value. We point the mouse at the curve where it intersects the x−axis.20 . near the x = 1. it can be found in advanced numerical analysis textbooks. y = f ( x ) = 0. in the By changing cell we enter A24 and we click on OK.00020 . We observe that other points are also displayed as the mouse is moved at different points near the curve. Numerical Analysis Using MATLAB® and Excel®.17) given that one real root lies in the 4 ≤ x ≤ 6 interval. We begin with a blank spreadsheet and we make the entries shown in Figure 2. that we can use to find the complex roots of a polynomial with real coefficients. For repetitive tasks. is with a chart such as those we’ve created for the last three examples. (1. with Excel. it is prudent to construct a template (model spreadsheet) with the appropriate formulas and then enter the coefficients of the polynomial to find its real roots*. and we observe that five handles (black square boxes) are displayed along different points on the curve. and when the cross symbol appears. use it to compute a root of the polynomial y = f ( x ) = x – 6x + 5x – 4x + 3x – 2x + x – 15 7 6 5 4 3 2 (2.6 Construct a template (model spreadsheet).041). We repeat the above steps for the next root near x = 2. We will not discuss this method here. Then. known as Bairstow’s method. −0.000.22515 . Third Edition Copyright © Orchard Publications 2−17 .30. 4. and we verify that for x = 2.

10. and the coefficients of the first derivative. we terminate the approximation steps there.10. is the familiar Newton’s formula which also appears in Row 14. Next. We also enter the polynomial coefficients. We observe that B16 now displays #DIV/0! (this is a warning that some value is being divided by zero). Since we are told that one real root is between 4 and 6.xls. Then. The formula in B16 of Figure 2. We save the spreadsheet of Figure 2. say template. 3. Model spreadsheet for finding real roots of polynomials. we save it with a different name. 2−18 Numerical Analysis Using MATLAB® and Excel®. therefore. but this will change once we enter the polynomial coefficients. and in B16 we type the formula =A16-($A$7*A16^7+$B$7*A16^6+$C$7*A16^5+$D$7*A16^4 +$E$7*A16^3+$F$7*A16^2+$G$7*A16^1+$H$7)/ ($B$12*A16^6+$C$12*A16^5+$D$12*A16^4+$E$12*A16^3 +$F$12*A16^2+$G$12*A16^1+$H$12) The use of the dollar sign ($) is explained in Paragraph 4 below. Powers of x and corresponding coefficients of given polynomial p(x) Enter coefficients of p(x) in Row 7 x6 x5 x4 x7 x3 x2 x Constant Coefficients of the derivative p'(x) Enter coefficients of p'(x) in Row 12 x5 x4 x6 x3 x2 x Constant Approximations: xn+1 = xn − p(xn)/p'(xn) Initial (x0) 1st (x1) 2nd (x2) 3rd (x3) 4th (x4) 5th (x5) 6th (x6) 7th (x7) Figure 2. say Example_2_6.xls. and the coefficients of the first derivative in Rows 7 and 12 respectively. We observe that there is no change in the values of E16 and F16. we copy B16 to C16:F16 and the spreadsheet now appears as shown in the spreadsheet of Figure 2.9.Chapter 2 Root Approximations A 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 B C D E F G H Spreadsheet for finding approximations of the real roots of polynomials up the 7th power by Newton's Method. This value is our first (initial) approximation. we take the average 5 and we enter it in A16. Third Edition Copyright © Orchard Publications . 2.9 with a name.

* This is a step−by−step problem−solving procedure.20409 5. and $B$16 is an absolute cell. we observe the successive approximations in Row 16. Thus. and so on. absolute so that the formula of B16 can be copied to C16. then. D16 and so on. and in front of the row number. by first placing the cursor in front. We can now use this template with any other polynomial by just entering the coefficients of the new polynomial in row 7 and the coefficients of its derivative in Row 12.3 The Bisection Method for Root Approximation The Bisection (or interval halving) method is an algorithm* for locating the real roots of a function. Numerical Analysis Using MATLAB® and Excel®. In this example. changes to C16. we made all cells. Powers of x and corresponding coefficients of given polynomial p(x) Enter coefficients of p(x) in Row 7 x6 x5 x4 x7 1 -6 5 -4 x3 3 x2 -2 x 1 Constant -15 Coefficients of the derivative p'(x) Enter coefficients of p'(x) in Row 12 x5 x4 x6 7 -36 25 x3 -16 x2 9 x -4 Constant 1 Approximations: xn+1 = xn − p(xn)/p'(xn) Initial (x0) 1st (x1) 2nd (x2) 3rd (x3) 4th (x4) 5th (x5) 6th (x6) 7th (x7) 5.10. have dollar signs ($) in front of the column letter. when copied to the next column to the right. The relative cell A16. except A16. at the end.163194 5. without changing its value. recursive computational procedure for solving a problem in a finite number of steps.0 5. A cell that is not absolute is said to be relative cell.16507 5. The contents of a relative cell changes when it is copied from one location to another. when copied to the next column. except A16. changes to B16. We can easily convert a relative cell to absolute or vice versa. we press the function key F4. then. These cells are said to be absolute. All cells in the formula of B16. 4. Spreadsheet for Example 2. especially an established.163190 Figure 2.The Bisection Method for Root Approximation A 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 B C D E F G H Spreadsheet for finding approximations of the real roots of polynomials up the 7th power by Newton's Method.163190 5. B16 is a relative cell. or between the letters and numbers of the cell. 2. Third Edition Copyright © Orchard Publications 2−19 . The value of an absolute cell does not change when it is copied from one position to another.6.

Sketches to illustrate the bisection method when f ( x 1 ) and f ( x m ) have same sign 2. In this case.12.11. so that f ( x 1 ) and f ( x 2 ) have opposite signs. either f ( x 1 ) > 0 and f ( x 2 ) < 0 .11. f ( x m ) ⋅ f ( x 1 ) > 0 . If f ( x m ) and f ( x 1 ) have opposite signs. f ( x m ) ⋅ f ( x 1 ) < 0 . f ( x 1 ) and f ( x m ) have opposite signs and thus their product is negative f ( x 1 ) and f ( x m ) have opposite signs and thus their product is negative • • x•1 x m x2 x1 • • • xm x2 Figure 2. Sketches to illustrate the bisection method when f ( x 1 ) and f ( x m ) have opposite signs After making the appropriate substitution. the above process is repeated until the root we are seeking has a specified tolerance.Chapter 2 Root Approximations The objective is to find two values of x. specify a tolerance on the error of f ( x ) 2−20 Numerical Analysis Using MATLAB® and Excel®. Then. In this case. or f ( x 1 ) < 0 and f ( x 2 ) > 0 . This indicates that x m and x 1 are on the left side of the x−axis crossing as shown in Figure 2.12. If any of these two conditions is satisfied. f ( x 1 ) and f ( x m ) are both positive and thus their product is positive • x•1 x•m x 2 f ( x 1 ) and f ( x m ) are both negative and thus their product is positive • • x x•1 x m 2 Figure 2. that is. If f ( x m ) and f ( x 1 ) have the same sign. we can compute the midpoint xm of the interval x 1 ≤ x ≤ x 2 with x1 + x2 x m = ---------------2 (2. Third Edition Copyright © Orchard Publications .18) Knowing x m . we replace x 2 with x m . we replace x 1 with x m . that is. To terminate the iterations. say x 1 and x 2 . we can find f ( x m ) . the following decisions are made: 1. This indicates that x m and x 2 are on the right side of the x−axis crossing as in Figure 2. their product will be negative. specify a number of iterations b. that is. we either: a. their product will be positive.

%9.19) a. end Before we write the program script.6f and %13.* x − 8. The syntax is: for x = array commands.7 Use the Bisection Method with MATLAB to approximate one of the roots of by y = f ( x ) = 3x – 2x + 6x – 8 5 3 (2. function y= funcbisect01(x). Here.. evaluate all commands down to the end command. a. After the k = 16 iteration. ….^ 3 + 6 . The for end loop allows a group of functions to be repeated a fixed and predetermined number of times.fm) line.^ 5 − 2 . by specifying 16 iterations. we must define a function assigned to the given polynomial and save it as a function m−file. 6 of which will be to the right of the decimal point. The special format \n specifies a linefeed.4. we can type help fprintf. We will discuss other special formats as they appear in subsequent examples. %s for string format.* x . For more information. with a total of 9 digits.00001 tolerance for f ( x ) . Third Edition Copyright © Orchard Publications 2−21 . by specifying 0. Let us also review the meaning of the fprintf('%9. the loop ends and any commands after the end are computed and displayed as commanded. Numerical Analysis Using MATLAB® and Excel®. and using a for end loop MATLAB program b.6f %13. fm must be expressed in decimal format with a total of 13 digits. that is. Likewise. and %d for integer format. it prints everything specified up to that point and starts a new line.6f are referred to as format specifiers or format scripts. % otherwise our script will fill the screen with values of y On the script below.6f \n'. We will define this function as funcbisect01 and will save it as funcbisect01.The Bisection Method for Root Approximation We will illustrate the Bisection Method with examples using both MATLAB and Excel.* x .. Example 2. Some other specifiers are %e for scientific format. the statement for k = 1:16 says for k = 1.m. % We must not forget to type the semicolon at the end of the line above. xm. y = 3 . and using a while end loop MATLAB program Solution: This is the same polynomial as in Example 2. the first specifies that the value of xm must be expressed in decimal format also called fixed point format. k = 2. 6 of which will be to the right of the decimal point. k = 16 .

. The while end loop evaluates a group of commands an indefinite number of times.031250 -0.038318 1.6 disp(' xm fm') % xm is the average of x1 and x2.125000 1. x1=1.4. xm=(x1+x2) / 2.058594 -0. b.015168 1. The syntax is: while expression commands.103195 1.001844 1.6f %13.Chapter 2 Root Approximations The script for the first part of Example 2. f2=funcbisect01(x2). fm=funcbisect01(xm).fm) % Prints xm and fm on same line.6f \n'.054688 -0. 2−22 Numerical Analysis Using MATLAB® and Excel®.060394 -0.7 is given below. if (f1*fm<0) x2=xm.006289 1. xm.002604 1. x2=2. fm is f(xm) disp('-------------------------') % insert line under xm and fm for k=1:16.060547 0. else x1=xm.060410 0. MATLAB displays the following: xm fm ------------------------1. Third Edition Copyright © Orchard Publications .060303 -0..062500 0.749023 1. Figure 2.500000 17.000380 1.060425 0. end end When this program is executed. fprintf('%9.059570 -0.000176 1.506944 1.000732 1. % We know this interval from Example 2.032885 1.250000 4.241184 1.031250 1.308441 1.046875 -0.060364 -0.000102 We observe that the values are displayed with 6 decimal places as we specified.060059 -0. but for the integer part unnecessary leading zeros are not displayed. f1=funcbisect01(x1).

059570 -0.001844 1.The Bisection Method for Root Approximation end The commands between while and end are executed as long as all elements in expression are true. Third Edition Copyright © Orchard Publications 2−23 .125000 1.062500 0.058594 -0.500000 17.060059 -0.015168 1.000003 Numerical Analysis Using MATLAB® and Excel®.6f \n'.000037 1.046875 -0.fm).054688 -0.060402 -0. The script should be written so that eventually a false condition is reached and the loop then terminates. Now we type and execute the following while end loop program.060406 0.060364 -0.060303 -0.060547 0. fprintf('%9.000176 1.506944 1. we will use the same as in part a.000380 1. f1=funcbisect01(x1).749023 1.031250 -0. fm=funcbisect01(xm).000102 1. x2=2.060394 -0. end end When this program is executed. xm=(x1+x2)/2.060425 0. tol=0. disp(' xm fm').031250 1.00001. x2=xm. MATLAB displays the following: xm fm ------------------------1.060404 -0.060410 0.032885 1.103195 1.6f %13.038318 1. x1=1.006289 1.002604 1. disp('-------------------------') while (abs(x1-x2)>2*tol). if (f1*fm<0).250000 4. There is no need to create another function m−file.241184 1.308441 1.000032 1. xm. f2=funcbisect01(x2). else x1=xm.000732 1.

value_if_false) where logical_test: any value or expression that can be evaluated to true or false.5 . 1200):If the value in D22 is less than the value of E22. This requires repeated use of the IF function which has the following syntax. A13: 2 B13: 3 C13: =(A13+B13)/2 2 5 within 2−24 Numerical Analysis Using MATLAB® and Excel®. false is returned. =IF(D22<E22. we make the following entries in rows 13 and 14. 800. L8/24):If the value in M8 is not equal to the value in N17. We will create a template as we did in Example 2.00001 accuracy. Example 2. Value_if_true can be another formula. =IF(M8<>N17. If logical_test is true and value_if_true is omitted. use the value in A15.value_if_true. value_if_false is the value that is returned if logical_test is false . we will use an Excel spreadsheet to construct a template that approximates a real root of a function with the bisection method.13.8 Use the bisection method with an Excel spreadsheet to approximate the value of 0. B15):If the value in C11 is greater than or equal to 1500. otherwise assign the number 1200. We expect the positive root to be in the 2 < x < 3 interval so we assign x 1 = 2 and x 2 = 3 . We start with a blank spreadsheet and we make the entries in rows 1 through 12 as shown in Figure 2. assign the number 800. Now. Third Edition Copyright © Orchard Publications . otherwise use the value in L8 divided by 24. Value_if_false can be another formula. The average of these values is x m = 2. value_if_true: the value that is returned if logical_test is true. true is returned. Solution: Finding the square root of 5 is equivalent to finding the roots of x – 5 = 0 . =IF(C11>=1500. If logical_test is false and value_if_false is omitted. otherwise use the value in B15.6 so we can use it with any polynomial equation. =IF(logical_test. These statements may be clarified with the following examples. K7*12.A15.Chapter 2 Root Approximations Next. use the value in K7 multiplied by 12.

B13) B14: =IF(A14=A13. we click on Paste and we observe the values in A15:F30.14. Numerical Analysis Using MATLAB® and Excel®. we highlight A14:F14.00000 0.00000 1.The Bisection Method for Root Approximation A 1 2 3 4 5 6 7 8 9 10 11 12 B C D E F G H Spreadsheet for finding approximations of the real roots of polynomials using the Bisection method Equation: y = f(x) = x − 5 = 0 Powers of x and corresponding coefficients of given polynomial f(x) Enter coefficients of f(x) in Row 9 2 x 7 x 6 x 5 x 4 x 3 x 2 x 0 Constant -5 0. C13. from the Edit menu.00000 x1 x2 xm (x1+x2)/2 f(x1) f(xm) f(x1)f(xm) Figure 2. we highlight the range A15:A30. Partial spreadsheet for Example 2.13.14. Third Edition Copyright © Orchard Publications 2−25 . Finally. Then. we place the cursor on A15. C13. We verify that the numbers on D14:F14 are as shown on the spreadsheet of Figure 2. We place the cursor on D14 and from the Edit menu we click on Paste. we highlight D13:F13 and on the Edit menu we click on Copy.00000 0. B13) We copy C13 into C14 and we verify that C14: =(A14+B14)/2 Next.00000 0.8 D13: =$A$9*A13^7+$B$9*A13^6+$C$9*A13^5+$D$9*A13^4 +$E$9*A13^3+$F$9*A13^2+$G$9*A13^1+$H$9*A13^0 E13: =$A$9*C13^7+$B$9*C13^6+$C$9*C13^5+$D$9*C13^4 +$E$9*C13^3+$F$9*C13^2+$G$9*C13^1+$H$9*C13^0 F13: =D13*E13 A14: =IF(A14=A13.00000 0. and holding the mouse left button. The square root of 5 accurate to six decimal places is shown on C30 in the spreadsheet of Figure 2. from the Edit menu we click on Copy.

00000 0.00320 -0.25000 2.50000 2.21484 -0.00000 0.00000 0.00006 0.00006 -0.23602 2.00000 -0.00000 0.23608 2.00757 0.00000 x1 2.23607 xm (x1+x2)/2 2.18750 2.23608 2.14.23608 2.23828 2.00000 0.00757 -0.00020 -0.50000 2.23438 2.23438 2.21875 2.00047 -0.00007 -0.00020 -0.00116 -0.23633 2.23607 2.21484 -0.23605 2.23596 2.48438 0.00757 -0.23606 x2 3.00102 -0.23584 2.02740 0.00000 0.23438 2.23608 2.00003 f(xm) 1.12500 2.24219 2.00000 -1.25000 2.00000 Figure 2.23606 f(x1) -1.00757 -0.25000 2.06250 0.18750 2. Third Edition Copyright © Orchard Publications .06250 -0.00006 -0.23535 2.23828 2.00000 2.00000 -0.25000 2.00000 1.25000 2.8 2−26 Numerical Analysis Using MATLAB® and Excel®.10406 0.00990 0.00000 0.00000 2.00102 0.01657 0.23633 2.23584 2.00007 -0.23605 2.00047 -0.00320 -0.07715 -0.00757 -0.00000 -1.00058 -0.23584 2.00000 2.23438 2.12500 2.48438 -0.00000 0.48438 -0.00000 0.00021 -0.00102 -0.23596 2.23605 2.25000 0.00000 0.23606 2.25000 2.00001 f(x1)f(xm) -1.24219 2.23633 2.00000 2.07715 -0.00002 0.00003 -0.23633 2.23438 2.23535 2. Entire spreadsheet for Example 2.25000 -0.21875 2.23607 2.00001 0.00000 0.23602 2.Chapter 2 Root Approximations A 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 B C D E F G H Spreadsheet for finding approximations of the real roots of polynomials using the Bisection method 2 Equation: y = f(x) = x − 5 = 0 Powers of x and corresponding coefficients of given polynomial f(x) Enter coefficients of f(x) in Row 9 x 7 x 6 x 5 x 4 x 3 x 2 x 0 Constant -5 0.23608 2.

• The Bisection (or interval halving) method is an algorithm for locating the real roots of a function. either f ( x 1 ) > 0 and f ( x 2 ) < 0 . This requires repeated use of the IF function which has the =IF(logical_test.4 Summary • Newton’s (or Newton−Raphson) method can be used to approximate the roots of any linear or non−linear equation of any degree. we must begin with a reasonable approximation of the root value. • The MATLAB the while end loop evaluates a group of statements an indefinite number of times and thus can be effectively used for root approximation. • For approximating real roots we can use Excel’s Goal Seek feature. The objective is to find two values of x.value_if_true. In all cases. if we have the function y = f ( x ) . • We can use a spreadsheet to approximate the real roots of linear and non−linear equations but to approximate all roots (real and complex conjugates) it is advisable to use MATLAB. We use Goal Seek when we know the desired result of a single formula. it is prudent to construct a template (model spreadsheet) with the appropriate formulas and then enter the coefficients of the polynomial to find its real roots. we can compute the midpoint xm of the interval x 1 ≤ x ≤ x 2 with x1 + x2 x m = ---------------2 • We can use the Bisection Method with MATLAB to approximate one of the roots by specifying a number of iterations using a for end or by specifying a tolerance using a while end loop program. • For repetitive tasks. • We can use an Excel spreadsheet to construct a template that approximates a real root of a function with the bisection method. say x 1 and x 2 . such as finding the roots of polynomials. that is. but we do not know the input value which satisfies that result. find the value of the independent variable x which satisfies that goal. we can use Goal Seek to set the dependent variable y to the desired value (goal) and from it. this can best be done by plotting f ( x ) versus x . so that f ( x 1 ) and f ( x 2 ) have opposite signs.value_if_false) Numerical Analysis Using MATLAB® and Excel®. Thus.Summary 2. It uses the formula f ( xn ) x n + 1 = x n – -------------f ' ( xn ) To apply Newton’s method. If any of these two conditions is satisfied. Third Edition Copyright © Orchard Publications 2−27 . or f ( x 1 ) < 0 and f ( x 2 ) > 0 .

Repeat Exercise 1 above using the Bisection method. f 1 ( x ) = x + x – 3 b. Third Edition Copyright © Orchard Publications . 3. have opposite signs. f 2 ( x ) = 4 a = 1 b = 2 a = 2 b = 4 2x + 1 – x + 4 Hint: Start with x 0 = ( a + b ) ⁄ 2 2. and verify from the graph that f ( a ) and f ( b ) .2 2−28 Numerical Analysis Using MATLAB® and Excel®. Repeat Example 2.5 using MATLAB. use Newton’s method to estimate the root of f ( x ) = 0 that lies between a and b . where a and b defined below. Then. Use MATLAB to sketch the graph y = f ( x ) for each of the following functions. a. Hint: Use the procedure of Example 2.Chapter 2 Root Approximations 2.5 Exercises 1.

169 ) + 1.164 .4526 0. plot(x.1443 . f1x=x.169 – -----------------------------------------------x 2 = x 1 – -------------3 7.= 1.436 ) . x=−2:0.= 1.1 – -----------------------------------x 1 = x 0 – -------------.169 – --------------.169 ) + 1 Check with MATLAB: pa=[1 0 0 1 −3].169 – 3 .39 f ' ( x1 ) 4 ( 1.324 4 ( 1.1.5 1 1.25 so we take x 0 = 1.3241i 0. roots(pa) ans = -1. a.25 so we choose a = 1 and b = 1.1 – -------------------.f1x).5 2 From the plot above we see that the positive root lies between x = 1 and x = 1.^4+x−3. Third Edition Copyright © Orchard Publications 2−29 .= 1.Solutions to End-of-Chapter Exercises 2.1 – 3 ( – 0.= 1.1 as our first approximation. We compute the next value x 1 as 4 f ( x0 ) ( 1.1640 Numerical Analysis Using MATLAB® and Excel®.0365 ( 1.1 ) + 1.= 1.5 -1 -0.1 ) + 1 The second approximation yields 4 f( x1 ) 0. grid 16 14 12 10 8 6 4 2 0 -2 -4 -2 -1.5 0 0.1443 + 1.05:2.6 Solutions to End-of-Chapter Exercises 1.= 1.3241i 1.169 3 f ' ( x0 ) 6.

solve(f2x) 7 – 7 = 0. To compute the next value x 1 we first need to find the first derivative of f 2 ( x ) .*x+1)−sqrt(x+4).f2x). x=−5:0.⋅ ( 2x + 1 ) – 1 ⁄ 2 ⋅ 2 – -.05:5.*x+1)−sqrt(x+4). plot(x.⋅ ( x + 4 ) – 1 ⁄ 2 ⋅ 1 = ----------------------.= 3 – ----------------------------------------------. We also observe that since f ( x 0 ) = there was no need to find the first derivative f ' ( x0 ) . f2x=sqrt(2. ans = 3 2−30 Numerical Analysis Using MATLAB® and Excel®. grid Warning: Imaginary parts of complex X and/or Y arguments ignored.Chapter 2 Root Approximations b. the real root is exactly x = 3 .⋅ f ( x ) = -– ------------------2 2 dx 2 2x + 1 2 x + 4 f( x0 ) 2×3+1– 3+4 0 .5 -2 -5 -4 -3 -2 -1 0 1 2 3 4 5 From the plot above we see that the positive root is very close to x = 3 and so we take x 0 = 3 as our first approximation. We rewrite it as f2 ( x ) = 2x + 1 – x + 4 = ( 2x + 1 ) 1⁄2 1⁄2 – (x + 4) Then. 0. Check with MATLAB: syms x.5 0 -0. 1 1 d 1 1 .5 -1 -1.= 3 . Third Edition Copyright © Orchard Publications .= 3 – --------------------x 1 = x 0 – -------------f ' ( x0 ) 1 ⁄ 7 – 1 ⁄ (2 7) 1 ⁄ (2 7) and Thus. f2x=sqrt(2.

After saving this file as exercise2. if (f1*fm<0) x2=xm.163574 -0.160156 -0.057803 1. Before we write the program script. fm is f(xm) disp('-------------------------') % insert line under xm and fm for k=1:12.500000 3. Third Edition Copyright © Orchard Publications 2−31 . We will use the while end loop MATLAB program and specify a tolerance of 0.m function y= exercise2(x). f1=exercise2(x1).171875 0.6f \n'.Solutions to End-of-Chapter Exercises 2. else x1=xm.691406 1. a.187500 0.014045 1.125000 -0.162109 -0.m. Numerical Analysis Using MATLAB® and Excel®.^ 4 +x − 3. x2=2.156250 -0. We will use the for end loop MATLAB program and specify 12 iterations. We need to redefine the function m−file because the function in part (b) is not the same as in part a.056411 1. We will define this function as exercise2 and will save it as exercise2. y = x .028229 1.000200 1.001584 b.250000 0. % x1=a and x2=b disp(' xm fm') % xm is the average of x1 and x2. fm=exercise2(xm). we execute the following program: x1=1.176041 1.003367 1. end end MATLAB displays the following: xm fm ------------------------1.163086 -0.164063 0. f2=exercise2(x2).fm)% Prints xm and fm on same line.273193 1.6f %13. xm=(x1+x2) / 2. xm.00001.562500 1. fprintf('%9. we must define a function assigned to the given polynomial and save it as a function m − file.006930 1.163818 -0.

the program % will not display any values because xm=(x1+x2)/2 = 3 = answer disp(' xm fm'). x2=4. tol=0.000065 2.002065 3. y = sqrt(2.000002 2−32 Numerical Analysis Using MATLAB® and Excel®.000001 3. xm=(x1+x2)/2.010938 0. % If we specify x1=a=2 and x2=b=4. else x1=xm.014289 3.999121 -0.000012 2.000014 3.000005 3.000443 2.062500 0. xm.000061 0.037013 2.068779 2.999658 -0.000195 0. we execute the following program: x1=2.000166 2.1. After saving this file as exercise2.999994 -0.*x+1)−sqrt(x+4).002344 0. f1=exercise2(x1).000011 0. f2=exercise2(x2).993750 -0.650000 -0. disp('-------------------------') while (abs(x1-x2)>2*tol).Chapter 2 Root Approximations function y= exercise2(x).fm).m.925000 -0.000037 2.00001.005299 3. MATLAB displays the following: xm fm ------------------------3. fprintf('%9.6f \n'. end end When this program is executed.000369 3.011733 2.998047 -0.6f %13. x2=xm.028125 0. if (f1*fm<0).001182 3.999927 -0.200000 0.3. Third Edition Copyright © Orchard Publications . fm=exercise2(xm).000027 0.

The second value (2) is not exactly correct as we can see from the plot.*x)+x−1. From Example 2. To find a good approximation of the second root that lies between x = 2 and x = 3 .Solutions to End-of-Chapter Exercises 3. Third Edition Copyright © Orchard Publications 2−33 .*x)+sin(2. This is because when solving equations of periodic functions.y).5. y = f ( x ) = cos 2x + sin 2x + x – 1 We use the following script to plot this function. grid 6 4 2 0 -2 -4 -6 -8 -5 -4 -3 -2 -1 0 1 2 3 4 5 Let us find out what a symbolic solution gives. y=cos(2. y=cos(2*x)+sin(2*x)+x−1. x=−5:0.*x)+sin(2. function y=exercise3(x) % Finding roots by Newton's method using MATLAB y=cos(2. there are an infinite number of solutions and MATLAB restricts its search for solutions to a limited range near zero and returns a non−unique subset of solutions. syms x. we write and save the function files exercise3 and exercise3der as defined below. solve(y) ans = [0] [2] The first value (0) is correct as it can be seen from the plot above and also verified by substitution of x = 0 into the given function. function y=exercise3der(x) Numerical Analysis Using MATLAB® and Excel®.02:5. plot(x.*x)+x−1.

x)) Enter starting value: 3 First approximation is x = 2. fprimex = exercise3der(x). Third Edition Copyright © Orchard Publications .*x)+1. fx=exercise3(x). x)) while input('Next approximation? (<enter>=no.6f \n'. fprimex=exercise3der(x). x = xnext.6f \n'. x=xnext. xnext = x−fx/fprimex.*sin(2. xnext=x−fx/fprimex.*x)+2. Now. we write and execute the following program and we find that the second root is x = 2. x)) end. x = input('Enter starting value: '). disp(sprintf('%9. fx = exercise3(x).1=yes)'). disp(sprintf('Next approximation is x = %9. disp(sprintf('First approximation is x = %9. fprimex = exercise3der(x).Chapter 2 Root Approximations % Finding roots by Newton's method % The following is the first derivative of % the function defined as exercise3 y=−2. fx = exercise3(x).6f \n'.*cos(2.229485 2−34 Numerical Analysis Using MATLAB® and Excel®.2295 and this is consistent with the value shown on the plot.

Also. phase angle. Numerical Analysis Using MATLAB® and Excel®.5 1 0 Time 0 Time -0 . 3. Phasors which are rotating vectors in terms of complex numbers. and period are defined. Stated differently. an alternating current alternates between positive and negative values at regularly recurring intervals of time.Chapter 3 Sinusoids and Phasors T his chapter is an introduction to alternating current waveforms.1 Alternating Voltages and Currents The waveforms shown in Figure 3. Voltage or Current Voltage or Current 1 2 0 . the average of the positive and negative values over a period is zero. Voltage and current relationships are expressed in sinusoidal terms. The characteristics of sinusoids are discussed and the frequency. Third Edition Copyright © Orchard Publications 3−1 . are also introduced and their relationships to sinusoids are derived.1 may represent alternating currents or voltages.5 -1 -1 0 2 T 4 6 8 1 0 1 2 -2 0 2 T 4 6 8 1 0 1 2 Voltage or Current Time Voltage or Current Time T T Figure 3. Examples of alternating voltages and currents Thus an alternating current (AC) is defined as a periodic current whose average value over a period is zero.1.

Generation of a sinusoid by rotation of a phasor At the completion of one cycle. and since ω is the angular velocity. and then repeats itself to form another cycle. can be represented by a sum of sinusoids. the cos ω t and sin ω t functions are generated from the projections of the phasor on the horizontal and vertical axis respectively. Thus.1.2 Characteristics of Sinusoids Consider the sine waveform shown in Figure 3. is used to express the number of cycles per second. that is. then ωT = 2π or π ----T = 2 ω (3.2. A = 1 as shown. where f ( t ) may represent either a voltage or a current function. denoted as Hz . 3. Thus. f (t) Voltage or Current Sine Wavef orm π ⁄ 2 ( 90 ° ) Phasor Direction of rotation 0 (0°) 2 π ( 360 ° ) A π ( 180 ° ) ω A = 1 0 π π /2 3π /2 2π −A 3 π ⁄ 2 ( 270 ° ) Time Figure 3. Third Edition Copyright © Orchard Publications . we say 3−2 Numerical Analysis Using MATLAB® and Excel®. which is a circle with radius of one unit. or any other unit. Unless otherwise stated. and let f ( t ) = A sin t where A is the amplitude of this function. if it takes one second to complete one cycle (one revolution around the unit circle). Two main reasons for studying sinusoids are: (1) many physical phenomena such as electric machinery produce (nearly) sinusoidal voltages and currents and (2) by Fourier analysis. t = T (one period). any periodic waveform which is not a sinusoid.2. A sinusoid (sine or cosine function) can be constructed graphically from the unit circle. it has traveled 2 π radians or 360 ° degrees.1) The term frequency in Hertz.Chapter 3 Sinusoids and Phasors As shown in Figure 3. commonly known as angular or radian frequency. We observe that when the phasor has completed a cycle (one revolution). if we let the phasor (rotating vector) travel around the unit circle with an angular velocity ω . the period T of an alternating current or voltage is the smallest value of time which separates recurring values of the alternating waveform. such as the square and sawtooth waveforms on the previous page. our subsequent discussion will be restricted to sine or cosine waveforms and these are referred to as sinusoids.

these are of the same frequency. We must remember that when we say that one sinusoid leads or lags another sinusoid. Numerical Analysis Using MATLAB® and Excel®. V max1 0. For example. of course. It is possible. we see that the cosine function will be the same as the sine function if the latter is shifted to the left by π ⁄ 2 radians.2) The frequency f is often referred to as the cyclic frequency to distinguish it from the radian frequency ω . if we shift the cosine function to the right by π ⁄ 2 radians or 90 ° .1) we have 1 f = -T or ω = 2πf (3.3.5 cos ω t sin ω t π⁄4 π⁄2 π 3π ⁄ 2 2π ωt 0 -0. we obtain the sine waveform. or 90 ° . If the phase angle between them is 0 ° degrees. it is convenient to plot sinusoids versus ω t (radians) rather that time t . the two sinusoids are said to be in-phase. v ( t ) = V max cos ω t . we say that the cosine function leads (is ahead of) the sine function by π ⁄ 2 radians or 90 ° . and i ( t ) = I max sin ω t are plotted as shown in Figure 3.5 -1 – V max 0 1 2 3 T 4 5 6 7 Figure 3. Obviously.4 shows three sinusoids which are outof-phase. Third Edition Copyright © Orchard Publications 3−3 . Thus. Likewise. or there is a phase angle of 90 ° between the cosine and sine functions. in this case. Plot of the cosine and sine functions By comparing the sinusoidal waveforms of Figure 3. we say that the sine function lags (is behind) the cosine function by π ⁄ 2 radians or 90 ° . Since the cosine and sine functions are usually known in terms of degrees or radians.3. two sinusoids of different frequencies can never be in phase. that two sinusoids are out-of-phase by a phase angle other than 90 ° .3. Figure 3.Characteristics of Sinusoids that the frequency is 1 Hz or one cycle per second. Another common expression is that the cosine and sine functions are out-of-phase by 90 ° . The frequency is denoted by the letter f and in terms of the period T and (3.

these must first be written either both as cosine functions. and – sin x = sin ( x ± 180 ° ) and sin x = cos ( x – 90 ° ) 3−4 Numerical Analysis Using MATLAB® and Excel®. When two sinusoids are to be compared in terms of their phase difference. Out-of-phase sinusoids It is convenient to express the phase angle in degrees rather than in radians in a sinusoidal function.5 1 0. For example. Then.25 sin ( ω t – ϕ ) 1. Example 3.5 0 -0. We should remember also that a negative amplitude implies 180 ° phase shift.5 -4 sin ω t θ 0. it is acceptable to express as v ( t ) = 100 sin ( 2000 π t – π ⁄ 6 ) v ( t ) = 100 sin ( 2000 π t – 30 ° ) since the subtraction inside the parentheses needs not to be performed. and should also be written with positive amplitudes.5 -1 -1. Third Edition Copyright © Orchard Publications .Chapter 3 Sinusoids and Phasors 1.1 Find the phase difference between the sinusoids i 1 = 120 cos ( 100 π t – 30 ° ) and i 2 = – 6 sin ( 100 π t – 30 ° ) Solution: We recall that the minus (−) sign indicates a ± 180 ° phase shift. and that the sine function lags the cosine by 90 ° .4. or both as sine functions.75 sin ( ω t + θ ) 0 ωt ϕ -2 0 2 4 6 8 Figure 3.

we will be using several trigonometric identities.= 0. Third Edition Copyright © Orchard Publications 3−5 . provide the following relations and formulas for quick reference.283 … radians in 360 ° degrees. therefore. has an angular velocity ω = 2 π n radians per second. or i 1 lags i 2 by 90 ° . Then. Definition of radian As shown in Figure 3. a rotating vector that completes n revolutions per second. the radian is a circular angle subtended by an arc equal in length to the radius of the circle. cos 0 ° = cos 360 ° = cos 2 π = 1 π 3 .≈ 57.5. In our subsequent discussion. derivatives and integrals involving trigonometric functions.= -.5.707 4 2 π 1 cos 60 ° = cos -.= -----. Some useful trigonometric relations are given below for quick reference. we see that i 2 leads i 1 by 90 ° .5) (3. We. The circumference of a circle is 2 π r units.866 cos 30 ° = cos -6 2 π 2 cos 45 ° = cos -. there are 2 π or 6. whose radius is r units in length.Characteristics of Sinusoids i 2 = 6 sin ( 100 π t – 210 ° ) = 6 sin ( 100 π t + 150 ° ) = 6 cos ( 100 π t + 150 ° – 90 ° ) = 6 cos ( 100 π t + 60 ° ) and comparing i 2 with i 1 .= 0.= 0. Then.3) The angular velocity is expressed in radians per second.3 ° 1 radian = ---------2π (3. 1 radian = 57. therefore.7) Numerical Analysis Using MATLAB® and Excel®.3 deg r r π radians Figure 3.5.6) (3.5 3 2 (3. and it is denoted by the symbol ω .4) (3.= -----. Let us also review the definition of a radian and its relationship to degrees with the aid of Figure 3. 360 ° .

8) (3.= 0 2 2π –1 cos 120 ° = cos ----.= --------. Third Edition Copyright © Orchard Publications .19) (3.866 3 2 (3.= 0 cos 270 ° = cos 3 2 5π cos 300 ° = cos ----.20) (3.5 3 11 π cos 330 ° = cos -------.5 3 2 3 π --------.23) 3−6 Numerical Analysis Using MATLAB® and Excel®.= 0.707 ----.= – cos 240 ° = cos 4 2 3 π ----.14) (3.= -----.13) (3.= 1 sin 30 ° = sin π 2 6 π 2 sin 45 ° = sin -.866 6 2 2 π --------.16) (3.= – cos 225 ° = cos 5 2 4 1 π ----.866 6 sin 0 ° = sin 360 ° = sin 2 π = 0 -.= – 0.= 0.18) (3.= – 0.= ----.= 1 2 2π 3 sin 120 ° = sin ----.9) (3.22) (3.866 ----.= 0.= – 0.= – 0.= -----.17) (3.10) (3.= – 0.707 4 2 π 1 sin 60 ° = sin -.12) (3.= – cos 150 ° = cos 5 2 6 cos 180 ° = cos π = – 1 7π – 3 cos 210 ° = cos ----.= -.15) (3.5 ----.11) (3.21) (3.= 0.866 3 2 π sin 90 ° = sin -.5 -.= 0.Chapter 3 Sinusoids and Phasors π cos 90 ° = cos -.= 0.

= ----.25) (3.= – 0.33) (3.32) (3.38) (3.= --------.= --------.42) Numerical Analysis Using MATLAB® and Excel®.34) (3.707 sin 225 ° = sin ----4 2 3 π --------.= – 0.24) (3.= – 0.= – sin 240 ° = sin 4 2 3 π ----.= 0.39) (3.= – 0.31) (3. Third Edition Copyright © Orchard Publications 3−7 .28) (3.30) (3.26) (3.= ---------cot θ = ----------sin θ tan θ 1 sec θ = ----------cos θ 1 csc θ = ---------sin θ tan ( 90 ° + θ ) = – cot θ (3.40) (3.35) (3.Characteristics of Sinusoids 5π 1 sin 150 ° = sin ----.= –1 sin 270 ° = sin 3 2 5π – 3 sin 300 ° = sin ----.= – 0.27) (3.866 ----.= -.37) (3.41) (3.5 6 2 sin 180 ° = sin π = 0 7π –1 sin 210 ° = sin ----.36) (3.5 6 2 5π – 2 .5 6 2 cos ( – θ ) = cos θ cos ( 90 ° + θ ) = – sin θ cos ( 180 ° – θ ) = – cos θ sin ( – θ ) = – sin θ sin ( 90 ° + θ ) = cos θ sin ( 180 ° – θ ) = sin θ sin θ tan θ = ----------cos θ cos θ 1 .29) (3.866 3 2 11 π –1 sin 330 ° = sin -------.= ----.

53) (3.( 1 + cos 2 θ ) cos θ = -2 1 2 .48) (3.sin ( θ – φ ) cos θ sin φ = -2 2 1 -.49) (3.45) (3.cos ( θ – φ ) – 1 -.43) (3.cos ( θ + φ ) sin θ sin φ = -2 2 2 2 2 2 (3.sin ( θ – φ ) .59) 3−8 Numerical Analysis Using MATLAB® and Excel®.51) (3.55) (3.58) (3.54) (3.47) (3.56) (3.( 1 – cos 2 θ ) sin θ = -2 1 1 .52) (3.cos ( θ + φ ) + -cos θ cos φ = -2 2 1 .sin ( θ + φ ) + 1 sin θ cos φ = -2 2 1 .sin ( θ + φ ) – 1 -.50) (3.57) (3.44) (3.46) (3.Chapter 3 Sinusoids and Phasors tan ( 180 ° – θ ) = – tan θ cos ( θ + φ ) = cos θ cos φ – sin θ sin φ cos ( θ – φ ) = cos θ cos φ + sin θ sin φ sin ( θ + φ ) = sin θ cos φ + cos θ sin φ sin ( θ – φ ) = sin θ cos φ – cos θ sin φ tan θ + tan φ tan ( θ + φ ) = ------------------------------1 – tan θ tan φ tan θ – tan φtan ( θ – φ ) = ------------------------------1 + tan θ tan φ cos θ + sin θ = 1 cos 2 θ = cos θ – sin θ sin 2 θ = 2 sin θ cos θ 2 tan θ tan 2 θ = --------------------2 1 – tan θ 1 2 .cos ( θ – φ ) . Third Edition Copyright © Orchard Publications .

cos ax + c a 1 . dv d ( sin v ) = cos v ----dx dx dv d ( cos v ) = – sin v ----dx dx d ( e v ) = e v dv ----dx dx (3.= ---------------------------1 b+c -(β + γ) tan -2 2 c–a ---------. Third Edition Copyright © Orchard Publications 3−9 .= ---------------------------1 c+a (3.= ----------------------------1 a+b -(α + β) tan -2 c = a + b – 2ab cos γ 1 -(β – γ) tan -- 2 2 2 2 b–c ---------. by the law of cosines.68) (3.= --------sin α sin β sin γ a = b + c – 2bc cos α b = a + c – 2ac cos β 2 2 2 2 2 2 (3.70) ∫ sin ax dx ∫ cos a x dx ∫e ax = –1 -. by the law of sines.6 be any triangle. General triangle Then. 1 -(α – β) tan -2 a – b ----------. a β c γ b α Figure 3.64) -(γ + α) tan -2 The following differential and integral trigonometric and exponential functions.65) (3.= ---------. are used extensively in engineering.61) (3.66) (3.6.62) (3.69) (3.67) (3.sin ax + c = -a 1 ax +c a -e dx = -- Numerical Analysis Using MATLAB® and Excel®.63) 1 -(γ – α) tan -- and by the law of tangents. a b c ---------.60) (3.Characteristics of Sinusoids Let Figure 3.

In other words.2 Find the angle θ if cos –10. Third Edition Copyright © Orchard Publications . we will designate the x -axis (abscissa) as the real axis. and the rotating vector A is referred to as a phasor. When this vector is rotated by another 90 ° for a total of 270 ° . if w = sin v . but it has been rotated counterclockwise by 90 ° .Chapter 3 Sinusoids and Phasors 3. r is a real number and jr is an imaginary number. Therefore. In the electrical engineering field.3 Inverse Trigonometric Functions The notation cos –1y or arc cos y is used to denote an angle whose cosine is y . if it is given that a vector A has the direction along the right side of the x -axis as shown in Figure 3. Example 3. j = 1 . Thus. if y = cos x . 2 3 4 3−10 Numerical Analysis Using MATLAB® and Excel®. * A more appropriate nomenclature for the real and imaginary axes would be the axis of the cosines and the axis of the sines respectively. we conclude that j = – 1 . Similarly. Essentially.7. Note: In our subsequent discussion. Therefore. These are called Inverse Trigonometric Functions. then x = cos –1y . Thus.5 = θ Solution: Here. the imaginary axis is just as important as the real axis. its value becomes j ( – A ) = – j A . then u = tan –1z . the square root of minus one is denoted as i . j = – j .5 . In this case. cos 60 ° = 0. by the operator j . Also. say r . From (3. another multiplication of the new vector j A by j will produce another 90 ° counterclockwise direction.4 Phasors In the language of mathematics. Thus. A fourth 90 ° rotation returns the vector to its original position. and thus its value is again A .* An imaginary number is the product of a real number. we denote i as j to avoid confusion with current i . i = – 1 .5. and the y -axis (ordinate) as the imaginary axis with the understanding that the “imaginary” axis is just as “real” as the real axis. the vector A has rotated 180 ° and its new value now is – A . multiplication of this vector by the operator j will result in a new vector j A whose magnitude remains the same. that is. j is an operator that produces a 90 ° counterclockwise rotation to any vector to which it is applied as a multiplying factor. θ = 60 ° 3. then v = sin –1w . we want to find the angle θ given that its cosine is 0. and if z = tan u .7).

Addition and Subtraction of Phasors jA y j ( jA ) = j 2 A = – A A x 2 j ( – jA ) = – j A = A j ( – A ) = j 3 A = – jA Figure 3. When written as A = a + jb . a = Re { A } and b = Im { A } where Re { A } denotes real part of A . if A = a + jb and B = c + jd . and an imaginary component equal to the sum of the imaginary components. Then. 3.5 Addition and Subtraction of Phasors The sum of two phasors has a real component equal to the sum of the real components. are equal if and only if their real parts are equal and also their imaginary parts are equal. we change the signs of the components of the subtrahend and we perform addition. For example. The j operator A complex number is the sum (or difference) of a real number and an imaginary number.3 It is given that A = 3 + j4 . henceforth any complex number will be referred to as a phasor.7. and B = 4 – j2 . then and A + B = (a + c) + j(b + d) A – B = (a – c) + j(b – d ) Example 3. is a complex number. Thus. and b = Im { A } the imaginary part of A . Find A + B and A – B Solution: Numerical Analysis Using MATLAB® and Excel®. Since in engineering we use complex quantities as phasors. By definition. A = B if and only if a = c and b = d . two phasors A and B where A = a + jb and B = c + jd . Third Edition Copyright © Orchard Publications 3−11 . Thus. For subtraction. it is said to be expressed in rectangular form. the number A = a + jb where a and b are both real numbers.

then A ⋅ B = ( a + jb ) ⋅ ( c + jd ) = ac + jad + jbc + j 2 bd and since j 2 = – 1 .Chapter 3 Sinusoids and Phasors A + B = (3 + j4) + (4 – j2) = (3 + 4) + j(4 – 2) = 7 + j2 A – B = (3 + j4) – (4 – j2) = (3 – 4) + j(4 + 2) = – 1 + j6 3. Then. Find A ⋅ B Solution: A ⋅ B = ( 3 + j4 ) ⋅ ( 4 – j2 ) = 12 – j6 + j16 – j 2 8 = 20 + j10 The conjugate of a phasor. Thus. is another phasor with the same real component. then A ⋅ A∗ = ( a + jb ) ( a – jb ) = a 2 – jab + jab – j 2 b 2 = a + b 2 2 3−12 Numerical Analysis Using MATLAB® and Excel®.71) Example 3. it follows that A ⋅ B = ac + jad + jbc – b d = ( ac – bd ) + j ( ad + bc ) (3. the result is a real number. if A = a + jb . and making use of the fact that j 2 = – 1 . and with an imaginary component of opposite sign. denoted as A∗ . Third Edition Copyright © Orchard Publications . Thus.4 It is given that A = 3 + j4 and B = 4 – j2 . if A = a + jb and B = c + jd . if A = a + jb . then A∗ = a – j b . but the imaginary component has opposite sign. Thus. Find A∗ Solution: The conjugate of the phasor A has the same real component. A∗ = 3 – j 5 If a phasor A is multiplied by its conjugate.6 Multiplication of Phasors Phasors are multiplied using the rules of elementary algebra. Example 3.5 It is given that A = 3 + j5 .

This procedure is called rationalization of the quotient. it is desirable to obtain the quotient separated into a real part and an imaginary part. Using this procedure. and it is done by multiplying the denominator by its conjugate.73) 3−13 . and B = 4 + j3 . B∗ a + jb a + jb ) ( c – jd ) = A ac + bd ) + j ( bc – ad ) A --.72) In (3. Example 3.96 + j 0.7 It is given that A = 3 + j4 .= ( --.28 .= ---------------2 2 25 25 B 25 4 + j3 (4 + j3)(4 – j3) 4 +3 3.7 Division of Phasors When performing division of phasors.72).+ j ---------------------2 2 2 2 c +d c +d (3. Find A ⋅ A∗ Solution: A ⋅ A∗ = ( 3 + j5 ) ( 3 – j5 ) = 3 + 5 = 9 + 25 = 34 2 2 3.8 Exponential and Polar Forms of Phasors The relations e and Numerical Analysis Using MATLAB® and Excel®. we get 73 + j4 (3 + j4)(4 – j3) 12 – j 9 + j 16 + 12 24 + j 7 A ----.⋅ -----.= ------------= 0.= -------------------------------------.Division of Phasors Example 3.+ j ------. Find A ⁄ B Solution: Using the procedure of (3. if A = a + jb and B = c + jd .= 24 .6 It is given that A = 3 + j5 . then.= ( ----------------------------------------------------------------------------------------2 2 B B∗ B c + jd ( c + jd ) ( c – jd ) c +d bc – ad ) ( ac + bd ) ( = ---------------------. we see that the quotient is easily separated into a real and an imaginary part. Third Edition Copyright © Orchard Publications jθ = cos θ + j sin θ (3.= ------------------------------------------. Thus.= ------------.72). we multiplied both the numerator and denominator by the conjugate of the denominator to eliminate the j operator from the denominator of the quotient.

80) To convert a phasor from exponential to rectangular form. Multiplying (3.79) Therefore. to convert a phasor from rectangular to exponential form. and thus Ce jθ = a + jb (3.73) by the real positive constant C we get: Ce jθ – jθ = cos θ – j sin θ (3. C = a +b 2 2 2 2 2 2 2 2 2 2 2 (3.77) or C = a +b 2 2 (3. we get a + b = ( C cos θ ) + ( C sin θ ) = C ( cos θ + sin θ ) = C Then.Chapter 3 Sinusoids and Phasors e are known as the Euler’s identities.77) b = C sin θ = tan θ -.74) = C cos θ + j C sin θ (3.81) 3−14 Numerical Analysis Using MATLAB® and Excel®. we get a = C cos θ and b = C sin θ Squaring and adding the expressions in (3.75) and (3.77). we use the expression a + jb = a +b e 2 2 –1 (3.76) Equating real and imaginary parts in (3. we use the expressions Ce Ce jθ = C cos θ + j C sin θ = C cos θ – j C sin θ – jθ (3. say a + jb . from (3.78) Also.--------------a C cos θ –1 b ⎞ θ = tan ⎛ -- or ⎝a⎠ j ⎛ tan ⎝ b⎞ -a⎠ (3.75) This expression represents a phasor. Third Edition Copyright © Orchard Publications .76).

The real and imaginary components of this phasor are shown in Figure 3.1° 3 Re Figure 3.8 Convert the following phasors to exponential and polar forms: a. thetax=angle(x)*180/pi.8 and 3. – 1 + j2 c. magx=abs(x). that is. In Examples 3.1301 or Numerical Analysis Using MATLAB® and Excel®. [x. 3 + j4 = 3 +4 ⋅e 2 2 j tan ( 4 ⁄ 3 ) – = 5e j53. Ce jθ = C ∠θ (3.9 below. we will verify the results with the following MATLAB co-ordinate transformation functions: [theta.82) where the left side of (3.1 ° = 5 ∠53. 4 Im 5 53.8. and the right side is the polar form.r] = cart2pol(x.1 ° Check with MATLAB: x=3+j*4. Third Edition Copyright © Orchard Publications 3−15 .r) − transforms from polar to Cartesian co−ordinates Example 3. 4 – j3 Solution: a. disp(magx).y) − transforms from Cartesian to polar co−ordinates.Exponential and Polar Forms of Phasors The polar form is essentially the same as the exponential form but the notation is different. 3 + j4 b. We must remember that the phase angle θ is always measured with respect to the positive real axis.82) is the exponential form. – 2 – j d.y] = pol2cart(theta. and rotates in the counterclockwise direction. disp(thetax) 5 53. The components of 3 + j4 Then.8.

6° Re 2 Then.9. please refer to “Introduction to Simulink with Engineering Applications”. Im 5 63.Chapter 3 Sinusoids and Phasors x = 3. Third Edition Copyright © Orchard Publications .6 ° = 2. disp(magy). magy=abs(y). The components of – 1 + j2 1 +2 e 2 j tan ( 2 ⁄ – 1 ) – = 5e j116. For an introduction to Simulink.1301 We can also verify the result with Simulink®* as shown in the model of Figure 3. The real and imaginary components of this phasor are shown in Figure 3.236 ∠116. Figure 3. Simulink model for Example 3.9. A brief introduction to Simulink is provided in Appendix B. thetay=angle(y)*180/pi. – 1 + j2 = 2 Figure 3. 3−16 Numerical Analysis Using MATLAB® and Excel®. deg = theta*180/pi theta = 0.10.6 ° = 5 ∠116. [theta. disp(thetay) * The reader who is not familiar with Simulink may skip this model and all others without loss of continuity.8 (a) b.10.y). The K value for the Gain block has been specified as 180 ⁄ π to convert radians into degrees.9273 r = 5 deg = 53.6 ° Check with MATLAB: y=−1+j*2. ISBN 09744239-7-1.4° −1 116. y = 4.r] = cart2pol(x.

r] = cart2pol(x. Third Edition Copyright © Orchard Publications 3−17 .2361 116.11: Figure 3. Then.Exponential and Polar Forms of Phasors 2.5651 Check with the Simulink model of Figure 3.0344 r = 2.6° 5 Re −153.11.12. deg = theta*180/pi theta = 2.4 ° 206.5651 or x = −1.4 ) ° = 2. The components of – 2 – j Check with MATLAB: v=−2−j*1.4°(Measured Clockwise) −1 Figure 3. y = 2. magv=abs(v). disp(thetav) Numerical Analysis Using MATLAB® and Excel®.6 ° = Im 5 ∠206. disp(magv).12.236 ∠– 153. –2 –j 1 = 2 +1 ⋅e 2 2 j tan ( – 1 ⁄ – 2 ) – = 5e j206. [theta.6° −2 26.6 ° = 5e j ( – 153.8 (b) c.y).2361 deg = 116. Simulink model for Example 3. The real and imaginary components of this phasor are shown in Figure 3. thetav=angle(v)*180/pi.

4 –j 3 = 4 +3 ⋅e 2 2 Figure 3.13: Figure 3. The components of 4 – j3 j tan ( – 3 ⁄ 4 ) – = 5e j323. y = −1.Chapter 3 Sinusoids and Phasors 2. disp(thetaw) 3−18 Numerical Analysis Using MATLAB® and Excel®.8 (c) d.4349 or x = −2.1 ° = 5 ∠323.1 ° = 5e – j36. Im 323.6779 r = 2.y). thetaw=angle(w)*180/pi.14. deg = theta*180/pi theta = -2.13. magw=abs(w).4349 Check with the Simulink model of Figure 3. Simulink model for Example 3.9 ° Check with MATLAB: w=4−j*3.1° −36.9° −3 5 4 Re Then. The real and imaginary components of this phasor are shown in Figure 3.14.2361 -153. Third Edition Copyright © Orchard Publications . [theta.r] = cart2pol(x. disp(magw).9 ° = 5 ∠– 36.2361 deg = -153.

– 2 ∠30 ° = 2 ∠( 30 ° + 180 ° ) = 2 ∠210 ° = 2 ∠– 150 ° The components of this phasor are shown in Figure 3.8699 Check with the Simulink model of Figure 3.r] = cart2pol(x.6435 r = 5 deg = -36.Exponential and Polar Forms of Phasors 5 -36.9 Express the phasor – 2 ∠30 ° in exponential and in rectangular forms. Simulink model for Example 3. Since each j rotates a vector by 90 ° counterclockwise. Solution: We recall that – 1 = j 2 . Numerical Analysis Using MATLAB® and Excel®. Therefore.16.y).15. [theta.8699 or x = 4. y = −3. then – 2 ∠30 ° is the same as 2 ∠30 ° rotated counterclockwise by 180 ° . Third Edition Copyright © Orchard Publications 3−19 . deg = theta*180/pi theta = -0.8 (d) Example 3.15: Figure 3.

9541 Check with the Simulink model of Figure 3.16. 2 ∠– 150 ° = 2e – j 150 ° Figure 3.83) 3−20 Numerical Analysis Using MATLAB® and Excel®. that is. The components of 2 ∠– 150 ° = 2 ( cos 150 ° – j sin 150 ° ) = 2 ( – 0.73 Re 30° 2 −150°(Measured Clockwise) −1 Then. however.r) x = -1.73 – j Check with MATLAB: r = −2.17: Figure 3. Third Edition Copyright © Orchard Publications . if A = M ∠θ and B = N ∠φ then. To multiply two phasors in exponential (or polar) form. [x.Chapter 3 Sinusoids and Phasors Im 210° −1. theta = 30/pi. AB = MN ∠( θ + φ ) = M e jθ Ne jφ = MN e j(θ + φ ) (3.7578 y = -0.17.9 Note: The rectangular form is most useful when we add or subtract phasors.5 ) = – 1. the exponential and polar forms are most convenient when we multiply or divide phasors. Simulink model for Example 3. we multiply the magnitudes and we add the phase angles.866 – j0.y] = pol2cart(theta*180/pi.

58e j74.58 × 7.7.58.22e – j 118. It can be done faster with just MATLAB.18.7 ° ) = 90.8276 deg = -44.18*: Figure 3.10 Multiply A = 12.22 ∠– 118.7 ° ) ] = 90.4 ° and multiplication in exponential form yields AB = ( 12. Numerical Analysis Using MATLAB® and Excel®. r=r1*r2.7 ° Solution: Multiplication in polar form yields AB = ( 12.3 ° by B = 7. deg1=74.3. The intent here is to introduce relevant Simulink blocks for more complicated models. deg2=−118.4000 Check with the Simulink model of Figure 3.3 ° + ( – 118.83e – j 44.3 ° ) ( 7.83 ∠– 44.10 * It would certainly be a waste of time to use Simulink for such an application.22.Exponential and Polar Forms of Phasors Example 3. Simulink model for Example 3.3 ° – 118. r2=7. Third Edition Copyright © Orchard Publications 3−21 .4 ° Check with MATLAB: r1=12.22 ) ∠[ 74.83e j ( 74.7 ° ) = 90. deg=deg1+deg2 r = 90.58 ∠74.

7 ° Solution: Division in polar form yields A 12.22e j74.7424 deg = 193 Check with the Simulink model of Figure 3. that is.7 ° ) ] = 1.3.= ---------------------------.19*: * Same comment as on the footnote of the previous page.= 1.7 ° = 1.74e – j 167 ° Check with MATLAB: r1=12.3 ° – ( – 118.= --------------N jφ N B Ne jθ (3.74 ∠[ 74. deg=deg1−deg2 r = 1.e .58 ∠74.= M A ---.7 ° Division in exponential form yields 12. Third Edition Copyright © Orchard Publications .= 1.74 ∠193 ° = 1.11 Divide A = 12.58e A --.22 ∠– 118.7. r=r1/r2. deg2=−118. deg1=74.84) Example 3. 3−22 Numerical Analysis Using MATLAB® and Excel®.58 ∠74.= ---------------------------------.22 ∠– 118.3 ° by B = 7.∠( θ – φ ) = M --.74e j193 ° = 1.74e – j118.7 ° B 7.74 ∠– 167 ° B 7.22. A = M ∠θ and B = N ∠φ j(θ – φ ) M e .3 ° j74. r2=7.3 ° --.58.3 ° e j118. if then.Chapter 3 Sinusoids and Phasors To divide one phasor by another when both are expressed in exponential or polar form. we divide the magnitude of the dividend by the magnitude of the divisor. and we subtract the phase angle of the divisor from the phase angle of the dividend.

11 Numerical Analysis Using MATLAB® and Excel®.19. Third Edition Copyright © Orchard Publications 3−23 . Simulink model for Example 3.Exponential and Polar Forms of Phasors Figure 3.

For example. 3−24 Numerical Analysis Using MATLAB® and Excel®. • The notation cos –1y or arc cos y is used to denote an angle whose cosine is y . these must first be written either both as cosine functions. we say that the cosine and sine functions are out-of-phase by 90 ° . these are of the same frequency since two sinusoids of different frequencies can never be in phase. we write v ( t ) = 100 sin ( 2000 π t – π ⁄ 6 ) as v ( t ) = 100 sin ( 2000 π t – 30 ° ) • When two sinusoids are to be compared in terms of their phase difference. Alternately. or both as sine functions. • It is customary to express the phase angle in degrees rather than in radians in a sinusoidal function. • Two (or more) sinusoids can be out-of-phase by a phase angle other than 90 ° . denoted as Hz . f = 1 ⁄ T . The frequency f is often referred to as the cyclic frequency to distinguish it from the radian frequency ω. • A negative amplitude implies 180 ° phase shift. • The period T of an alternating current or voltage is the smallest value of time which separates recurring values of the alternating waveform. and should also be written with positive amplitudes. The circumference of a circle is 2 π r .9 Summary • An alternating current (or voltage) alternates between positive and negative values at regularly recurring intervals of time. • The radian is a circular angle subtended by an arc equal in length to the radius of the circle.Chapter 3 Sinusoids and Phasors 3. or there is a phase angle of 90 ° between the cosine and sine functions. if y = cos x . Thus. • Sine and cosine waveforms and these are referred to as sinusoids. • The cosine function leads (is ahead of) the sine function by π ⁄ 2 radians or 90 ° . then x = cos – 1y . is used to express the number of cycles per second. and the sine function lags (is behind) the cosine function by π ⁄ 2 radians or 90 ° . whose radius is r units in length. • A phasor is a rotating vector expressed as a complex number where j is an operator that rotates a vector by 90 ° in a counterclockwise direction. Third Edition Copyright © Orchard Publications . These are called Inverse Trigonometric Functions. • It is important to remember that when we say that one sinusoid leads or lags another sinusoid. • The angular velocity ω is commonly known as angular or radian frequency and ω T = 2 π • The term frequency in Hertz. The frequency is denoted by the letter f and in terms of the period T .

jθ = cos θ + j sin θ and e – jθ = cos θ – j sin θ are known as the Euler’s identi- • To convert a phasor from rectangular to exponential form. that is. Thus. Third Edition Copyright © Orchard Publications 3−25 . Thus. however. Ce jθ = C ∠θ and it is important to remember that the phase angle θ is always measured with respect to the positive real axis. then. For subtraction. are equal if and only if their real parts are equal and also their imaginary parts are equal. and with an imaginary component of opposite sign. we use the expressions Ce Ce jθ = C cos θ + j C sin θ = C cos θ – j C sin θ – jθ • The polar form is essentially the same as the exponential form but the notation is different. it is desirable to obtain the quotient separated into a real part and an imaginary part. we use the expression a + jb = a +b e 2 2 j ⎛ tan ⎝ –1 b⎞ -a⎠ • To convert a phasor from exponential to rectangular form.= ------------. Numerical Analysis Using MATLAB® and Excel®. denoted as A∗ . A = B if and only if a = c and b = d. • When performing division of phasors. Thus. • The rectangular form is most useful when we add or subtract phasors. the exponential and polar forms are most convenient when we multiply or divide phasors. bc – ad ) a + jb a + jb ) ( c – jd ) ( ac + bd ) + j ( bc – ad ) ( ac + bd ) j( A --. then A + B = ( a + c ) + j ( b + d ) and A – B = ( a – c ) + j ( b – d ) • Phasors are multiplied using the rules of elementary algebra. and an imaginary component equal to the sum of the imaginary components. is another phasor with the same real component. if A = a + jb and B = c + jd .= ( ------------------------------------= ----------------------------------------------------= ---------------------+ ---------------------2 2 2 2 2 2 B c + jd ( c + jd ) ( c – jd ) c +d c +d c +d • The relations e ties. • The sum of two phasors has a real component equal to the sum of the real components. If A = a + jb and B = c + jd . then A ⋅ B = ac + jad + jbc – b d = ( ac – bd ) + j ( ad + bc ) • The conjugate of a phasor.Summary • Two phasors A and B where A = a + jb and B = c + jd . then A∗ = a – j b . and rotates in the counterclockwise direction. Thus. if A = a + jb and B = c + jd . This is achieved by multiplying the denominator by its conjugate. we change the signs of the components of the subtrahend and we perform addition. if A = a + jb .

if A = M ∠θ and B = N ∠φ then.e N jφ N B Ne jθ 3−26 Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications .= --------------.= ---. that is. that is. if then. A = M ∠θ and B = N ∠φ M j(θ – φ) M e A . we divide the magnitude of the dividend by the magnitude of the divisor.∠( θ – φ ) = M --. AB = MN ∠( θ + φ ) = M e jθ Ne jφ = MN e j(θ + φ ) • To divide one phasor by another when both are expressed in exponential or polar form. we multiply the magnitudes and we add the phase angles.Chapter 3 Sinusoids and Phasors • To multiply two phasors in exponential (or polar) form. and we subtract the phase angle of the divisor from the phase angle of the dividend.

Perform the following operations.b. 4 100 2 ( 1 – j ) Check your answers with MATLAB 4.10 Exercises 1. -------------. ( 3 – j2 ) ⋅ ( – 2 – j3 ) g. ---------------. ---------------3 + j2 –3–j 4 – j10 ( 3 – j2 )∗ 3. ( 2 – j3 ) – ( 2 – j3 )∗ d. Third Edition Copyright © Orchard Publications 3−27 . ---------------------. ( 2 – j4 ) ⋅ ( 3 + j5 ) ⋅ ( 3 – j2 ) ⋅ ( – 2 – j3 ) 2. ( 2 – j4 ) ⋅ ( 3 + j5 ) f. and check your answers with MATLAB. 5. compute: 12 + j5 b.b. and check your answers with MATLAB. ------------------– j30 ° 5 ∠– 150 ° – 2e j60 ° Check your answers with MATLAB 9 – j4 6. ----------------b. Compute the rectangular form of 4 ∠30 ° e a. Find the real and imaginary components of -----------------– 5 + jx Numerical Analysis Using MATLAB® and Excel®. ( 3 – j2 ) ⋅ ( 3 – j2 )∗ e. Any phasor A can be expressed as A = a + jb = r ( cos θ + j sin θ ) = re jθ Using the identities ( re ) = r e a.d. ( – 3 + j5 ) – ( 1 + j6 ) c. a.Exercises 3. Perform the following operations. 6 jθ n jn θ or n re jθ = n re jθ ⁄ n . --------------------a. Compute the exponential and polar forms of – 8 + j3 9 + j5a. 22 + j6 8 + j6 120 ( 3 – j2 ) . ( 2 – j4 ) + ( 3 + j4 ) b.c. -----------------– 4 – j2 – 2 + j4 Check your answers with MATLAB.

(2−4j)*(3+5j)*(3−2j)*(−2−3j) ( 3 – j2 ) ⋅ ( – 2 – j3 ) = – 6 – j9 + j4 – 6 = – 12 – j5 ( 2 – j4 ) ⋅ ( 3 + j5 ) ⋅ ( 3 – j2 ) ⋅ ( – 2 – j3 ) = ( 6 + j10 – j12 + 20 ) ⋅ ( – 6 – j9 + j4 – 6 ) = ( 26 – j2 ) ⋅ ( – 12 – j5 ) = – 312 – j130 + j24 – 10 = – 322 – j106 ans = 5 ans = -4.5.0000 .0000i ans = -3. ( 3 – j2 ) ⋅ ( 3 – j2 )∗ = ( 3 – j2 ) ⋅ ( 3 + j2 ) = 9 + j6 – j6 + 4 = 13 e.1. (2−3j)−(2+3j).. ( – 3 + j5 ) – ( 1 + j6 ) = – 4 – j c..0000 . g.Chapter 3 Sinusoids and Phasors 3..0000i ans = 0 .0000 .6. a. ( 2 – j4 ) + ( 3 + j4 ) = 5 + 0 = 5 b.11 Solutions to End−of−Chapter Exercises 1.2200e+002 . ( 2 – j3 ) – ( 2 – j3 )∗ = ( 2 – j3 ) – ( 2 + j3 ) = 0 – j6 d.0000i ans = -12. (3−2j)*(−2−3j). ( 2 – j4 ) ⋅ ( 3 + j5 ) = 6 + j10 – j12 + 20 = 26 – j2 f.0000i ans = 13 ans = 26. Check with MATLAB: (2−4j)+(3+4j). (3−2j)*(3+2j).0600e+002i 3−28 Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications .2. (−3+5j)−(1+6j).1. (2−4j)*(3+5j).

= –3–j 2 2 –3–j –3–j –3+j 3 +1 10 + j10 300 120 . ------------------------------------------. (3−2j)/(3+2j) ans = 6 .= -------------b.= ---------------= ------------------. (100*sqrt(2)*(1−j))^(1/4) Numerical Analysis Using MATLAB® and Excel®.4 1200 + j1200 120 480 .0905 ) ( cos ( π ⁄ 16 ) – j sin ( π ⁄ 16 ) ) = 3.Solutions to End−of−Chapter Exercises 2.1i ans = 120/29 + 300/29i ans = 5/13 .= -------.+ j ⋅ ----------2 2 4 – j10 4 – j10 4 + j10 4 + 10 116 116 29 29 3 – j2 ) ( 3 – j2 ) – j6 – j6 – 4 ( 3 – j2 ) = -----------------5. 22 2 2 3 + j2 3 + j2 3 – j2 3 +2 13 8 + j6 – 3 + j – 24 + j8 – j18 – 6 8 + j6 – 30 – j10 . --------------------2 2 ( 3 – j2 )∗ ( 3 + j2 ) ( 3 – j2 ) 3 +2 13 13 13 Check with MATLAB: 22+6j)/(3+2j). 4 100 2 ( 1 – j ) = 4 100 2 ⋅ 2e –j π ⁄ 4 = ( 100 2 ⋅ 2e –j π ⁄ 4 1 ⁄ 4 ) = ( 100 2 ) 1⁄4 ⋅ 2 1 ⁄ 4 – j π ⁄ 16 e = ( 3.3948 ⁄ 6 = 13 1⁄6 ⋅e j0.53 + j0.= 6 – j2 a. Third Edition Copyright © Orchard Publications 3−29 .5334 ( cos 0. a.0658 = 1.395 = 6 13 ⋅ e j0.= 120 ⋅ ----------------------.⋅ -------------.= ------------------------------------.12/13i 3.10 b.= 480 .⋅ ------------.6883 – j0. (8+6j)/(−3−j).= ------------------------------------------------------------.7337 Check with MATLAB: (12+5j)^(1/6). 120/(4−10j).– j ⋅ 12 – j12 ------------------⋅( ----= 9 ---------------.0658 ) = 1.0658 + j sin 0. 22 + j6 3 – j2 66 – j44 + j18 + 12 + j6 78 – j26 .= ---------------c. 6 12 + j5 = 6 13e j0. -------------= ----------------------.4485 × 1.2i ans = -3 .= -------------------------------------------.= 5 d.+ j ⋅ -------.

.3588 2 2 j tan ( 3 ⁄ – 8 ) 73 e – 8 + j3 8 +3 ⋅e .= ---------------------------------–1 j3. 4 ∠30 ° = ( 4 ⁄ 5 ) ∠180 ° = – 0. abs(x). – j 3.3022e .= ------------------------------------------------------.Chapter 3 Sinusoids and Phasors ans = 1.6883 .= -------------------------.= 1.0.5e b.3022 ∠– 177.= 2. abs(y).5 ( 0 + j ) = – j0.9105 ans = 42. angle(x)*180/pi.9105e j42.5 3−30 Numerical Analysis Using MATLAB® and Excel®..3022e – j177.8744 ° = 1. – 2 + j4 2 +4 ⋅e = 1.5081 ° –1 = 2.⋅ ----------------.0981 9 + j5 9 +5 ⋅e 106 ⋅ e ----------------.5104 ans = 1.9105 ∠42.5081 ° – j0. Third Edition Copyright © Orchard Publications .8744 ° Check with MATLAB: x=(9+5j)/(−4−2j). a..7483 .6052 – 4 – j2 2 2 j tan ( – 2 ⁄ – 4 ) 20 ⋅ e 4 +2 ⋅e 2 2 j tan ( 5 ⁄ 9 ) –1 j0.7337i 4.5 ( cos 90 ° + j sin 90 ° ) = – 0.1008i ans = 3.1071 2 2 j tan ( 4 ⁄ – 2 ) 20 e b.9105e j0. a.5071 = 2.8789 5.8 ---------------------5 ∠– 150 ° e = – 0.5301 + 0. ------------------– j30 ° – 2e j60 ° j90 ° = – 0.3022 ans = -177. y=(−8+3j)/(−2+4j). angle(y)*180/pi ans = 2.= ----------------------------------------------------–1 – j1.

8000 . exp(pi*j/3)/(−2*exp(−pi*j/6)) ans = -0. Third Edition Copyright © Orchard Publications 3−31 .0.0000 ..Solutions to End−of−Chapter Exercises Check with MATLAB: 4*(cos(pi/6)+sin(pi/6)*j)/(5*(cos(−5*pi/6)+sin(−5*pi/6)*j)).5000i 6.= ---------------------.+ j ---------------------2 2 2 2 – 5 + jx – 5 – j x – 5 + jx x + 25 5 +x x + 25 Numerical Analysis Using MATLAB® and Excel®..0.= -----------------.0000i ans = -0.. 9 – j4 – 5 – j x – 45 – j9x + j20 – 4x 9 – j4 – 4x – 45 – 9x + 20 -----------------.= -------------------------------------------------.⋅ -------------.

Some definitions and examples are not applicable to subsequent material presented in this text. These are denoted with a dagger ( † ) and may be skipped. but are included for subject continuity.1) The numbers a ij are the elements of the matrix where the index i indicates the row. Thus. a 43 indicates the element positioned in the fourth row and third column. a22. a 33. and j indicates the column in which each element is positioned. …. In a square matrix. if a matrix has five rows and five columns. Determinants. and Gauss’s elimination method are introduced. Thus. a 33. it is said to be a square matrix of order 5. A matrix of m rows and n columns is said to be of m × n order matrix. Cramer’s rule. Third Edition Copyright © Orchard Publications 4−1 . a nn . Numerical Analysis Using MATLAB® and Excel®.Chapter 4 Matrices and Determinants T his chapter is an introduction to matrices and matrix operations. …. the elements a 11. Alternately. 2 3 7 1 –1 5 or 1 3 1 –2 1 –5 4 –7 6 In general form. are located on the main diagonal.1 Matrix Definition A matrix is a rectangular array of numbers such as those shown below. and reference to more advance topics in matrix theory. the matrix is said to be a square matrix of order m (or n ). we say that the matrix elements a 11. 4. a matrix A is denoted as a 11 a 12 a 13 … a 1n a 21 a 22 a 23 … a 2n A = a 31 a 32 a 33 … a 3n … … … … … a m1 a m2 a m3 … a mn (4. a nn are called the main diagonal elements. If m = n . a 22.

where each element of C is the sum (difference) of the corresponding elements of A and B . 3. A+B % Define matrices A and B % Add A and B * Henceforth. Third Edition Copyright © Orchard Publications . −1 2 5].2 Matrix Operations Two matrices A = a ij and B = b ij are equal. If A = a ij and B = b ij are conformable for addition (subtraction). These are discussed in matrix theory textbooks.3) Example 4. 3 . 4−2 Numerical Analysis Using MATLAB® and Excel®.1 Compute A + B and A – B given that A = 1 2 3 and B = 2 3 0 0 1 4 –1 2 5 Solution: A+B = 1+2 0–1 2+3 1+2 3+0 = 3 5 4+5 –1 3 3 9 and A–B = 1–2 0+1 2 – 3 3 – 0 = –1 –1 3 1–2 4–5 1 –1 –1 Check with MATLAB: A=[1 2 3. 2 . n (4. that is. all paragraphs and topics preceded by a dagger ( † ) may be skipped.Chapter 4 Matrices and Determinants † The sum of the diagonal elements of a square matrix A is called the trace* of A . their sum (difference) will be another matrix C with the same order as A and B . C = A ± B = [ a ij ± b ij ] (4. …. † A matrix in which every element is zero. 2. 0 1 4]. if and only if a ij = b ij i = 1. is called a zero matrix.2) Two matrices are said to be conformable for addition (subtraction). m j = 1 . 4. A = B . that is. …. if they are of the same order m × n. B=[2 3 0.

2 Multiply the matrix A = 1 –2 2 3 by (a) k 1 = 5 and (b) k 2 = – 3 + j2 Solution: a.Matrix Operations ans = 3 -1 A−B 5 3 3 9 % Subtract B from A ans = -1 1 -1 -1 3 -1 If k is any scalar (a positive or negative number). then multiplication of a matrix A by the scalar k . k2=(−3 + 2*j). and not [ k ] which is a 1 × 1 matrix. k1*A % Define scalars k1 and k2 % Define matrix A % Multiply matrix A by constant k1 ans = 5 10 k2*A -10 15 %Multiply matrix A by constant k2 Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications 4−3 . k 1 ⋅ A = 5 × 1 – 2 = 5 × 1 5 × ( – 2 ) = 5 – 10 2 3 5×2 5×3 10 15 b. Example 4. 2 3]. is the multiplication of every element of A by k . A=[1 −2. k 2 ⋅ A = ( – 3 + j2 ) × 1 – 2 = ( – 3 + j2 ) × 1 ( – 3 + j2 ) × ( – 2 ) = – 3 + j2 2 3 ( – 3 + j2 ) × 2 ( – 3 + j2 ) × 3 – 6 + j4 6 – j4 – 9 + j6 Check with MATLAB: k1=5.

then. to obtain the product A ⋅ B .0000+ 2.0000i -9. The product A ⋅ B will then be an m × n matrix.0000i -6. the operation is row by column.4. therefore the product C ⋅ D is 4−4 Numerical Analysis Using MATLAB® and Excel®.3 Given that 1 C = 2 3 4 and D = – 1 2 compute the products C ⋅ D and D ⋅ C Solution: The dimensions of matrices C and D are respectively 1 × 3 3 × 1 .0000+ 4.Chapter 4 Matrices and Determinants ans = -3. Thus. B and A are not conformable for multiplication B p×n A m×p For matrix multiplication. the product A ⋅ B (but not B ⋅ A ) is conformable for multiplication only if A is an m × p and matrix B is an p × n matrix. Example 4. Shows that A and B are conformable for multiplication A m×p B p×n Indicates the dimension of the product A ⋅ B For the product B ⋅ A we have: Here.0000. only when the number of columns of matrix A is equal to the number of rows of matrix B .0000+ 6. we multiply each element of a row of A by the corresponding element of a column of B . A convenient way to determine if two matrices are conformable for multiplication is to write the dimensions of the two matrices side−by−side as shown below. Third Edition Copyright © Orchard Publications . we add these products.0000i 6.0000i Two matrices A and B are said to be conformable for multiplication A ⋅ B in that order. That is.

4. and it will become apparent later in this chapter. 1 D ⋅ C = –1 2 3 4 = 2 (1) ⋅ (2) (1) ⋅ (3) (1) ⋅ (4) ( –1 ) ⋅ ( 2 ) ( –1 ) ⋅ ( 3 ) ( –1 ) ⋅ ( 4 ) (2) ⋅ (2) (2) ⋅ (3) (2) ⋅ (4) 2 3 4 = –2 –3 –4 4 6 8 Check with MATLAB: C=[2 3 4]. the product D ⋅ C is also feasible. when we discuss the inverse of a matrix. Third Edition Copyright © Orchard Publications 4−5 . and will result in a 1 × 1 .4) Numerical Analysis Using MATLAB® and Excel®. However. an equivalent operation exists. 2]. −1. that is. Multiplication of these will produce a 3 × 3 matrix as follows.3 Special Forms of Matrices † A square matrix is said to be upper triangular when all the elements below the diagonal are zero. The matrix A below is an upper triangular matrix.Special Forms of Matrices feasible. is not defined. a 11 a 12 a 13 … a 1n A = 0 a 22 a 23 … a 2n 0 0 … … … … … 0 … … 0 0 0 … a mn (4. C*D % Define matrices C and D % Multiply C by D ans = 7 D*C % Multiply D by C ans = 2 -2 4 3 -3 6 4 -4 8 Division of one matrix by another. 1 C ⋅ D = 2 3 4 –1 = ( 2 ) ⋅ ( 1 ) + ( 3 ) ⋅ ( –1 ) + ( 4 ) ⋅ ( 2 ) = 7 2 The dimensions for D and C are respectively 3 × 1 1 × 3 and therefore. D=[1.

1 0 0 0 1 0 0 0 1 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 1 0 0 1 (4. For applications. For example. The matrix B below is a lower triangular matrix.7) A scalar matrix with k = 1 .6) † A diagonal matrix is called a scalar matrix. is called an identity matrix I . and 4 × 4 identity matrices. 4−6 Numerical Analysis Using MATLAB® and Excel®. refer to Chapter 14. Third Edition Copyright © Orchard Publications . except those in the diagonal.Chapter 4 Matrices and Determinants In an upper triangular matrix. The matrix D below is a scalar matrix with k = 4 . † A square matrix is said to be diagonal. a 11 0 C = 0 0 0 0 … 0 0 a 22 0 … 0 0 … 0 0 0 0 … 0 0 0 … a mn (4. 3 × 3 . The matrix C below is a diagonal matrix.8) The MATLAB eye(n) function displays an n × n identity matrix. † A square matrix is said to be lower triangular. not all elements above the diagonal need to be non−zero. not all elements below the diagonal need to be non−zero.5) In a lower triangular matrix. Shown below are 2 × 2 . refer to Chapter 14. For applications. when all the elements above the diagonal are zero. a 11 B = 0 0 … 0 0 … 0 a 21 a 22 … … … 0 0 … … … … 0 a m1 a m2 a m3 … a mn (4. if all elements are zero. 4 D = 0 0 0 0 4 0 0 0 0 4 0 0 0 0 4 (4. if a 11 = a 22 = a 33 = … = a nn = k where k is a scalar.

Special Forms of Matrices eye(4)% Display a 4 by 4 identity matrix ans = 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 Likewise. 4 5 6] % Define matrix A A = 1 4 2 5 3 6 Numerical Analysis Using MATLAB® and Excel®. For example. For example. 4 −7 6] % Define matrix A A = 1 -2 4 then. let A be defined as A=[1 3 1. if 1 1 2 3 then A T = 2 4 5 6 3 4 5 6 A= (4. eye(size(A)) 3 1 -7 1 -5 6 displays ans = 1 0 0 0 1 0 0 0 1 T † The transpose of a matrix A . Thus. for the above example.9) In MATLAB we use the apostrophe (′) symbol to denote and obtain the transpose of a matrix. is the matrix that is obtained when the rows and columns of matrix A are interchanged. produces an identity matrix whose size is the same as matrix A . A=[1 2 3. the eye(size(A)) function. Third Edition Copyright © Orchard Publications 4−7 . denoted as A . −2 1 −5.

that is.0000i 3. 0 2 –3 A = –2 0 –4 3 4 0 0 –2 A = 2 0 –3 –4 T 3 4 = –A 0 4−8 Numerical Analysis Using MATLAB® and Excel®. and it is denoted as A∗ . A = 1 + j2 3 j 2 – j3 A∗ = 1 – j2 3 –j 2 + j3 † MATLAB has two built−in functions which compute the complex conjugate of a number.0000 .Chapter 4 Matrices and Determinants A'% Display the transpose of A ans = 1 2 3 4 5 6 T † A symmetric matrix A . Using MATLAB with the matrix A defined as above.0000 + 2. For example. An example of a symmetric matrix is shown below.0000i T † A square matrix A such that A = – A . conj(x).0000 + 3. is one such that A = A . computes the complex conjugate of any complex number.2. conj(A). Third Edition Copyright © Orchard Publications . is called the conjugate of A .1.0000 0 . and the second. 3 2−3j] % Define and display matrix A A = 1. is called skew−symmetric.0000 . 1 2 3 2 4 –5 3 –5 6 1 2 3 2 4 –5 = A 3 –5 6 A = A = T (4. the transpose of a matrix A is the same as A . computes the conjugate of a matrix A .0000i 3. the matrix obtained from A by replacing each element by its conjugate.0000 conj_A=conj(A) 0 + 1.0000i 2. The first. we obtain A = [1+2j j.0000i 2. An example is shown below.10) † If a matrix A has complex numbers as elements.3.0000i % Compute and display the conjugate of A conj_A = 1.

Determinants Therefore. the determinant of A . † A square matrix A such that A 1 A = 1+j 2 1–j 3 –j T∗ = A . that is.11) then. Let A be a determinant of order 2 .4 Determinants Let matrix A be defined as the square matrix a 11 a 12 a 13 … a 1n a 21 a 22 a 23 … a 2n A = a a a …a 31 32 33 3n … … … … … a n1 a n2 a n3 … a nn (4. is defined as detA = a 11 a 22 a 33 … a nn + a 12 a 23 a 34 … a n 1 + a 13 a 24 a 35 … a n 2 + … – a n 1 … a 22 a 13 … – a n 2 … a 23 a 14 – a n 3 … a 24 a 15 – … The determinant of a square matrix of order n is referred to as determinant of order n .13) Numerical Analysis Using MATLAB® and Excel®. For example. matrix A above is skew−Hermitian.12) (4. † A square matrix A such that A j A = –1–j –2 1–j 3j j T∗ = – A . matrix A above is skew symmetric. matrix A above is Hermitian. 1+j 3 j 2 1 T* A = –j 1–j 0 2 1+j 3 j 2 –j = A 0 2 1 T A = j 1–j 0 2 Therefore. denoted as detA . For example. is called skew−Hermitian. is called Hermitian. 4. –1–j 3j j –2 –j T* j A = 1+j 0 2 –1+j – 3j –j –2 –j = –A 0 j 2 T j A = 1–j 2 0 Therefore. A = a 11 a 12 a 21 a 22 (4. Third Edition Copyright © Orchard Publications 4−9 .

% This file computes the determinant of a 2x2 matrix % It must be saved as function (user defined) file % det2x2. B=[2 −1. % function y=det2x2(A). It is highly recommended that this % function file is created in MATLAB's Editor Window.4 Given that A = 1 2 and B = 2 – 1 3 4 2 0 compute detA and detB . 4−10 Numerical Analysis Using MATLAB® and Excel®. please refer to Introduction to Simulink with Engineering Applications. Make sure % that his directory is added to MATLAB's search % path accessed from the Editor Window as File>Set Path> % Add Folder.m in the current Work Directory.14) Example 4.Chapter 4 Matrices and Determinants Then. detA = a 11 a 22 – a 21 a 12 (4. ISBN 0−9744239−7−1.* The MATLAB user−defined function file below can be used to compute the determinant of a 2 × 2 matrix. this function is not included in the MATLAB Run−Time Function Library List that is used with the Simulink Embedded MATLAB Function block. Page 16−3. * For an example using this block. Solution: detA = 1 ⋅ 4 – 3 ⋅ 2 = 4 – 6 = – 2 detA = 2 ⋅ 0 – 2 ⋅ ( – 1 ) = 0 – ( – 2 ) = 2 % Define matrices A and B % Compute the determinant of A Check with MATLAB: A=[1 2. 3 4]. 2 0]. det(A) ans = -2 det(B) % Compute the determinant of B ans = 2 While MATLAB has the built−in function det(A) for computing the determinant of a matrix A. Third Edition Copyright © Orchard Publications .

Let A be a matrix of order 3..Determinants y=A(1.] and then % type det2x2(A) at the command prompt. we must first compute the cofactors. that is.2)−A(1. Third Edition Copyright © Orchard Publications 4−11 . To evaluate higher order determinants. % % To run this program. detA is found from detA = a 11 a 22 a 33 + a 12 a 23 a 31 + a 11 a 22 a 33 – a 11 a 22 a 33 – a 11 a 22 a 33 – a 11 a 22 a 33 (4. When this is done properly. a 11 a 12 a 13 a 11 a 12 a 21 a 22 a 23 a 21 a 22 a 31 a 32 a 33 a 31 a 32 − + This method works only with second and third order determinants. we obtain (4.5 Compute detA and detB given that 2 A = 1 2 3 5 0 1 1 0 2 –3 –4 0 –2 0 –5 –6 and B = 1 Solution: Numerical Analysis Using MATLAB® and Excel®. these will be defined shortly. then subtract the products formed by the diagonals from lower left to upper right as shown on the diagram of the next page. is to write the first two columns to the right of the 3 × 3 matrix.16) above.2)*A(2. and add the products formed by the diagonals from upper left to lower right.1)*A(2.. Example 4.1). a 11 a 12 a 13 A = a 21 a 22 a 23 a 31 a 32 a 33 (4. define the 2x2 matrix in % MATLAB's Command Window as A=[.15) then.16) A convenient method to evaluate the determinant of order 3..

det(A) ans = 9 B=[2 −3 −4. −A(3.. Make sure % that his directory is added to MATLAB's search % path accessed from the Editor Window as File>Set Path> % Add Folder.3)−A(3.2)*A(1. 1 0 1. Third Edition Copyright © Orchard Publications . 2 1 0].2)*A(2.3)*A(1. define the 3x3 matrix in % MATLAB's Command Window as A=[.1)*A(3. det(B) % Define matrix B and compute detB % Define matrix A and compute detA ans = -18 The MATLAB user−defined function file below can be used to compute the determinant of a 3 × 3 matrix.2)*A(3.Chapter 4 Matrices and Determinants 2 detA = 1 2 3 5 2 3 0 1 1 0 1 0 2 1 or Likewise.2)*A(2.1)*A(2.3)*A(2.1)*A(2.1)*A(1. % % To run this program. 1 0 −2..2). 0 −5 −6]. detA = ( 2 × 0 × 0 ) + ( 3 × 1 × 1 ) + ( 5 × 1 × 1 ) – ( 2 × 0 × 5 ) – ( 1 × 1 × 2 ) – ( 0 × 1 × 3 ) = 11 – 2 = 9 2 –3 –4 2 –3 detB = 1 0 – 2 1 – 2 0 –5 –6 2 –6 detB = [ 2 × 0 × ( – 6 ) ] + [ ( – 3 ) × ( – 2 ) × 0 ] + [ ( – 4 ) × 1 × ( – 5 ) ] – [ 0 × 0 × ( – 4 ) ] – [ ( – 5 ) × ( – 2 ) × 2 ] – [ ( – 6 ) × 1 × ( – 3 ) ] = 20 – 38 = – 18 or Check with MATLAB: A=[2 3 5. % This file computes the determinant of a 3x3 matrix % It must be saved as function (user defined) file % det3x3. y=A(1.2). 4−12 Numerical Analysis Using MATLAB® and Excel®.1)−A(3. % function y=det3x3(A).1)+A(1..3)+A(1...3)*A(2. It is highly recommended that this % function file is created in MATLAB's Editor Window.] and then % type det3x3(A) at the command prompt.m in the current Work Directory.3)*A(3.

M 22 . M 32 . Solution: M 11 = a 22 a 23 a 32 a 33 M 12 . the determinant of the remaining n – 1 square matrix is called the minor of determinant A . M 13 and the cofactors α 11 . α 12 and α 13 . M 23 . M 12 = a 21 a 23 a 31 a 33 M 11 = a 21 a 22 a 31 a 32 and α 11 = ( – 1 ) 1+1 M 11 = M 11 α 12 = ( – 1 ) 1+2 M 12 = – M 12 α 13 = M 13 = ( – 1 ) 1+3 M 13 The remaining minors M 21 . and it is denoted as M ij . The signed minor ( – 1 ) i+j M ij is called the cofactor of a ij and it is denoted as α ij . a 11 a 12 a 13 … a 1n a 21 a 22 a 23 … a 2n A = a a a … a 31 32 33 3n … … … … … a n1 a n2 a n3 … a nn (4.5 Minors and Cofactors Let matrix A be defined as the square matrix of order n as shown below. M 33 and cofactors Numerical Analysis Using MATLAB® and Excel®. and jth column.Minors and Cofactors 4. M 31 .18) compute the minors M 11 .17) If we remove the elements of its ith row. Third Edition Copyright © Orchard Publications 4−13 . Example 4.6 Given that a 11 a 12 a 13 A = a 21 a 22 a 23 a 31 a 32 a 33 (4.

the cofactors on the diagonals have the same sign as their minors. the value of the determinant of A is the sum of the products obtained by multiplying each element of any row or any column by its cofactor. α 31.19) compute its cofactors.22) α 31 = ( – 1 ) 3+1 2 – 3 = – 8.Chapter 4 Matrices and Determinants α 21. α 23.20) α 13 = ( – 1 ) 2 –4 = 0 –1 2 1 –3 = –9 –1 –6 α 21 = ( – 1 ) 2 –3 = 6 2 –6 1 2 = –4 –1 2 (4. Solution: α 11 = ( – 1 ) 1+1 – 4 2 = 20 2 –6 1+3 α 12 = ( – 1 ) 1+2 2 2 = 10 –1 –6 2+1 (4.23) α 33 = ( – 1 ) 3+3 1 2 = –8 2 –4 (4. and α 33 are defined similarly. α 22.24) It is useful to remember that the signs of the cofactors follow the pattern + − − + + − − + + − + − − + + − − + + − + − + − + that is.7 Given that 1 2 –3 A = 2 –4 2 –1 2 –6 (4. 4−14 Numerical Analysis Using MATLAB® and Excel®. Let A be a square matrix of any size. Example 4. –4 2 α 32 = ( – 1 ) 3+2 1 – 3 = –8 2 2 (4. Third Edition Copyright © Orchard Publications . α 32.21) α 22 = ( – 1 ) 2+2 α 23 = ( – 1 ) 2+3 (4.

Thus. We must use the above procedure to find the determinant of a matrix A of order 4 or higher. A = 1 2 –3 2 –4 2 –1 2 –6 (4. Example 4. det(A) % Define matrix A and compute detA ans = 40 The MATLAB user−defined function file below can be used to compute the determinant of a 4 × 4 matrix.26) +a 31 a 22 a 23 a 24 – a 41 a 22 a 23 a 24 Determinants of order five or higher can be evaluated similarly.9 Compute the value of the determinant Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications 4−15 .Minors and Cofactors Example 4. a fourth−order determinant can first be expressed as the sum of the products of the elements of its first row by its cofactor as shown below. 2 −4 2. a 11 a 12 a 13 a 14 A = a 21 a 22 a 23 a 24 a 31 a 32 a 33 a 34 a 41 a 42 a 43 a 44 a 22 a 23 a 24 a 42 a 43 a 44 a 12 a 13 a 14 a 42 a 43 a 44 a 12 a 13 a 14 = a 11 a 32 a 33 a 34 – a 21 a 32 a 33 a 34 a 42 a 43 a 44 a 12 a 13 a 14 a 32 a 33 a 34 (4.8 Compute the determinant of A using the elements of the first row. −1 2 −6].25) Solution: detA = 1 – 4 2 – 2 2 2 – 3 2 – 4 = 1 × 20 – 2 × ( – 10 ) – 3 × 0 = 40 2 –6 –1 –6 –1 2 Check with MATLAB: A=[1 2 −3.

27) Solution: Using the above procedure. % The following statement initializes y y=0.5 or Example 4. 4−16 Numerical Analysis Using MATLAB® and Excel®.m in the current Work Directory. % This file computes the determinant of a nxn matrix % It must be saved as function (user defined) file % detnxn. using the procedure of Example 4.Chapter 4 Matrices and Determinants 2 –1 0 A = –1 1 0 4 0 3 –3 0 0 –3 –1 –2 1 (4.8. −3 0 0 1].n]=size(A). Third Edition Copyright © Orchard Publications ⎧ ⎪ ⎪ ⎨ ⎪ ⎪ ⎩ ⎧ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎩ . % and det3x3(A) for a 3x3 matrix. we find and thus [ a ] = 6 . we will multiply each element of the first column by its cofactor. % MATLAB allows us to use the user-defined functions to be recursively % called on themselves so we can call det2x2(A) for a 2x2 matrix. delta = det(A) delta = -33 The MATLAB user−defined function file below can be used to compute the determinant of a n × n matrix. 4 0 3 −2. % function y=detnxn(A). Make sure % that his directory is added to MATLAB's search % path accessed from the Editor Window as File>Set Path> % Add Folder. −1 1 0 −1. Then. [ c ] = 0 . It is highly recommended that this % function file is created in MATLAB's Editor Window. % The following statement defines the size of the matrix A [n. 1 0 –1 A= 2 0 3 – 2 0 0 1 [a] –1 0 –3 –( –1 ) 0 3 –2 0 0 1 [b] –1 0 – 3 +4 1 0 – 1 0 0 1 [c] –1 0 – 3 –( –3 ) 1 0 – 1 0 3 –2 [d] ⎧ ⎪ ⎪ ⎨ ⎪ ⎪ ⎩ ⎧ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎩ Next. [ d ] = – 36 detA = [ a ] + [ b ] + [ c ] + [ d ] = 6 – 3 + 0 – 36 = – 33 We can verify our answer with MATLAB as follows: A=[ 2 −1 0 −3. [ b ] = – 3 .

Minors and Cofactors if n==2 y=det2x2(A). Property 1: If all elements of one row or one column are zero. detA is zero because the second column in A is 2 times the first column. Check with MATLAB: A=[2 4 1..29) Here. For example.28) then.. [1:(i−1) (i+1):n])). An example of this is the determinant of the cofactor [ c ] above. 1 2 1]. the determinant is zero.] and then % type detnxn(A) at the command prompt. detA = 2 3 1 4 6 2 1 1 1 2 3 1 4 6 = 12 + 4 + 6 – 6 – 4 – 12 = 0 2 (4.. det(A) ans = 0 Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications 4−17 . return end % For 4x4 or higher order matrices we use the following: % (We can define n and matrix A in Command Window for i=1:n y=y+(−1)^(i+1)*A(1. Some useful properties of determinants are given below. 3 6 1. the determinant is zero. return end % if n==3 y=det3x3(A). if 2 A = 3 1 4 6 2 1 1 1 (4. end % % To run this program. Property 2: If all the elements of one row or column are m times the corresponding elements of another row or column. define the nxn matrix in % MATLAB's Command Window as A=[.i)*detnxn(A(2:n.

6 Cramer’s Rule Let us consider the systems of the three equations below a 11 x + a 12 y + a 13 z = A a 21 x + a 22 y + a 23 z = B a 31 x + a 32 y + a 33 z = C (4. B. Third Edition Copyright © Orchard Publications . and C. This follows from Property 2 with m = 1 .31) are determinants that are formed from Δ by the substitution of the known values A.30) is a homogeneous set of equations. We observe that the numerators of (4. y . and v 3 if 2v 1 – 5 – v 2 + 3v 3 = 0 – 2v 3 – 3v 2 – 4v 1 = 8 v 2 + 3v 1 – 4 – v 3 = 0 (4. Example 4.31) provided that the determinant Δ (delta) is not zero.10 Use Cramer’s rule to find v 1. Then. x = y = z = 0 also. v 2. and D 3 are all zero as we found in Property 1 above. 4−18 Numerical Analysis Using MATLAB® and Excel®. D 1. then. If (4. for the coefficients of the desired unknown. 4. that is. Cramer’s rule applies to systems of two or more equations. if A = B = C = 0 .32) and verify your answers with MATLAB. the determinant is zero.30) and let Δ a 11 a 12 a 13 = a 21 a 22 a 23 a 31 a 32 a 33 D1 = A a 11 a 13 B a 21 a 23 C a 31 a 33 D2 = a 11 A a 13 a 21 B a 23 a 31 C a 33 D3 = a 11 a 12 A a 21 a 22 B a 31 a 32 C Cramer’s rule states that the unknowns x . D 2.Chapter 4 Matrices and Determinants Property 3: If two rows or two columns of a matrix are identical. and z can be found from the relations D1 x = ----- Δ D2 y = ----- Δ D3 z = ----- Δ (4.

3 4 −1]. and transferring known values to the right side. % Compute the determinant of D2 d3=[2 −1 5.= ----35 7 Δ D2 170 34 x 2 = ----. 8 −3 −2. v2 and v3.= – ----35 7 Δ D3 55 11 x 3 = ----. % The elements of D3 detd3=det(d3).Cramer’s Rule Solution: Rearranging the unknowns v . % The elements of D1 detd1=det(d1). 3 1 −1]. we obtain 2v 1 – v 2 + 3v 3 = 5 – 4v 1 – 3v 2 – 2v 3 = 8 3v 1 + v 2 – v 3 = 4 (4. format rat % Express answers in ratio form B=[2 −1 3. using (4.31) we obtain D1 85 17 x 1 = ----.33) Now. % Compute he determinant of D3 Numerical Analysis Using MATLAB® and Excel®. 4 1 −1]. −4 −3 8. −4 8 −2. % The following script will compute and display the values of v1. % Compute the determinant of D1 d2=[2 5 3.= – ----35 7 Δ (4. Δ = 2 –1 3 –4 –3 –2 3 1 –1 5 –1 3 8 –3 –2 4 1 –1 2 –4 3 5 3 8 –2 4 –1 5 8 4 2 –1 – 4 – 3 = 6 + 6 – 12 + 27 + 4 + 4 = 35 3 1 5 –1 8 – 3 = 15 + 8 + 24 + 36 + 10 – 8 = 85 4 1 2 –4 3 5 8 = – 16 – 30 – 48 – 72 + 16 – 20 = – 170 4 D1 = D2 = D3 = 2 –1 –4 –3 3 1 2 –1 – 4 – 3 = – 24 – 24 – 20 + 45 – 16 – 16 = – 55 3 1 Therefore.= – -------. by Cramer’s rule. −4 −3 −2. % The elements of D2 detd2=det(d2). 3 1 4].= ----. % Compute the determinant D of B d1=[5 −1 3. Third Edition Copyright © Orchard Publications 4−19 .34) We will verify with MATLAB as follows. % The elements of the determinant D delta=det(B).= – ----.

This method is best illustrated with the following example which consists of the same equations as the previous example. % Compute the value of v1 % Compute the value of v2 % Compute the value of v3 % Display the value of v1 % Display the value of v2 % Display the value of v3 v1= 17/7 v2= -34/7 v3= -11/7 These are the same values as in (4. Third Edition Copyright © Orchard Publications . disp('v2=').disp(v3).disp(v2). Example 4.35) with the third to eliminate the unknown v 2 and we obtain the following equation. Then.disp(v1). we add the first equation of (4. Subsequently.34) 4. substitution of the two values found can be made into an equation with three unknowns from which we can find the value of the third unknown. % disp('v1=').7 Gaussian Elimination Method We can find the unknowns in a system of two or more equations also by the Gaussian elimination method. 4−20 Numerical Analysis Using MATLAB® and Excel®. we can find the second unknown. and v 3 of 2v 1 – v 2 + 3v 3 = 5 – 4v 1 – 3v 2 – 2v 3 = 8 3v 1 + v 2 – v 3 = 4 (4.11 Use the Gaussian elimination method to find v 1. This procedure is repeated until all unknowns are found. v2=detd2/delta. by substitution to another equation with two unknowns.Chapter 4 Matrices and Determinants v1=detd1/delta. This can be done by multiplying the terms of any of the equations of the system by a number such that we can add (or subtract) this equation to another equation in the system so that one of the unknowns will be eliminated. v 2. v3=detd3/delta. disp('v3='). With this method. the objective is to eliminate one unknown at a time.35) Solution: As a first step.

10.37).37) from (4.39) Finally.36) Next.– ----. Then. is defined as the n square matrix shown on the next page.– ----. we obtain the following equation.= – ----7 7 7 7 (4.38) into (4.36) or (4.36) yields 11 7v 3 = – 11 or v 3 = – ----7 (4.37) Subtraction of (4. The Gaussian elimination is further discussed in Chapter 14 in conjunction with the LU factorization method. The Gaussian elimination method works well if the coefficients of the unknowns are small integers. Numerical Analysis Using MATLAB® and Excel®. However. 4.38) Now.8 The Adjoint of a Matrix Let us assume that A is an n square matrix and α ij is the cofactor of a ij . By substitution of (4. Third Edition Copyright © Orchard Publications 4−21 .35) by 3.⎞ = 9 or v 1 = 17 ----5v 1 + 2 ⋅ ⎛ – 11 ⎝ 7 ⎠ 7 (4. as in Example 4. Then the adjoint of A . and we add it with the second to eliminate v 2 .The Adjoint of a Matrix 5v 1 + 2v 3 = 9 (4. 5v 1 – 5v 3 = 20 (4. we can find the unknown v 1 from either (4.36) we obtain ----. we multiply the third equation of (4. we can find the last unknown v 2 from any of the three equations of (4.11. denoted as adjA .35). By substitution into the first equation we obtain 34 33 35 34 v 2 = 2v 1 + 3v 3 – 5 = ----. it becomes impractical if the coefficients are large or fractional numbers.40) These are the same values as those we found in Example 4.

Example 4. This topic is discussed in Appendix C. Example 4.13 Given that 4−22 Numerical Analysis Using MATLAB® and Excel®.Chapter 4 Matrices and Determinants α 11 α 21 α 31 … α n1 α 12 α 22 α 32 … α n2 adjA = α α α … α 13 23 33 n3 … … … … … α 1n α 2n α 3n … α nn (4. the matrix is said to be ill−conditioned.12 Compute adjA given that 1 2 3 A = 1 3 4 1 4 3 (4. if the determinant of that matrix is very small. Third Edition Copyright © Orchard Publications . are the elements of the ith column (row) of adjA .9 Singular and Non−Singular Matrices An n square matrix A is called singular if detA = 0 . If an n square matrix A is nearly singular.42) Solution: 3 4 adjA = – 1 1 1 1 4 3 4 3 3 4 – 2 4 1 1 3 3 3 3 2 3 3 4 – 2 3 3 4 = –7 6 –1 1 0 –1 1 –2 1 – 1 2 1 4 1 2 1 3 4. A is called non−singular. if detA ≠ 0 .41) We observe that the cofactors of the elements of the ith row (column) of A . that is.

A is called the inverse of B . we can compute its inverse from the relation A –1 1 = ----------adjA detA (4. Numerical Analysis Using MATLAB® and Excel®. and likewise. Third Edition Copyright © Orchard Publications 4−23 . that is.The Inverse of a Matrix 1 A = 2 3 2 3 3 4 5 7 (4. detA = 9 + 8 + 12 – 9 – 16 – 6 = – 2 . and since this is a non−zero value. 4.43) determine whether this matrix is singular or non−singular. where I is the identity matrix. it is possible to compute the inverse of A using (4.44). matrix A is singular.10 The Inverse of a Matrix If A and B are n square matrices such that AB = BA = I . Solution: detA = 1 2 3 2 3 3 4 5 7 1 2 2 3 = 21 + 24 + 30 – 27 – 20 – 28 = 0 3 5 Therefore. From Example 4. find A–1 Solution: Here. A = B –1 If a matrix A is non−singular.14 Given that 1 2 3 A = 1 3 4 1 4 3 (4. that is.45) compute its inverse.44) Example 4. denoted as B = A–1 . B is called the inverse of A .12.

0000 0.5000 0. 1 4 3].5000 -0.5000 2 3 4 3 4 3 -3.15 Prove the validity of (4.1 0 – 1 = – 0.adjA = ----= ----------–2 detA 1 –2 1 – 0.47) Example 4. A –1 1 1 1 –3 ⁄ 2 . A –1 –7 6 –1 3.2 –3 = = ----------adjA = -2 –2 4 detA –1 2 and 4−24 Numerical Analysis Using MATLAB® and Excel®.5 0 0.0000 0 1.5 1 – 0.5000 Multiplication of a matrix A by its inverse A–1 produces the identity matrix I .5 (4. AA –1 = I or A A = I –1 (4.5 . Third Edition Copyright © Orchard Publications . 1 3 4.5000 -0.47) for A = 4 2 3 2 2 –3 –2 4 Proof: detA = 8 – 6 = 2 and adjA = Then. that is. invA=inv(A) % Define matrix A and compute its inverse A = 1 1 1 invA = 3.5 – 3 0.46) Check with MATLAB: A=[1 2 3.5 1 1 .Chapter 4 Matrices and Determinants adjA = –7 6 –1 1 0 –1 1 –2 1 Then.5000 -0.

Multiplication of both sides of (4. Third Edition Copyright © Orchard Publications 4−25 . = x2 2 3 6 1 2 8 x3 (4.Solution of Simultaneous Equations with Matrices AA –1 = 4 2 3 1 –3 ⁄ 2 = 4 – 3 2 –1 2 2–2 –6+6 = 1 –3+4 0 0 = I 1 4. . and x 3 using the inverse matrix method. We assume that A and X are conformable for multiplication.48) where A and B are matrices whose elements are known. the given set of equations is AX = B where 2 A= 1 3 x1 3 1 9 = B X .11 Solution of Simultaneous Equations with Matrices Consider the relation AX = B (4.50) or X=A B –1 Therefore.16 Given the system of equations ⎧ 2x 1 + 3x 2 + x 3 = 9 ⎫ ⎪ ⎪ ⎨ x 1 + 2x 2 + 3x 3 = 6 ⎬ ⎪ ⎪ ⎩ 3x 1 + x 2 + 2x 3 = 8 ⎭ (4. x 2.51) compute the unknowns x 1.52) Then. Numerical Analysis Using MATLAB® and Excel®. we can use (4. Solution: In matrix form.50) to solve any set of simultaneous equations that have solutions. and X is a matrix (a column vector) whose elements are the unknowns. Example 4.48) by A–1 yields: A AX = A B = IX = A B –1 –1 –1 (4. We will refer to this method as the inverse matrix method of solution of simultaneous equations.49) (4.

7 1 – 5 6 = ----X = x 2 = ----18 18 5 ⁄ 18 0. A –1 1 –5 7 1 1 . A B for the matrix equation A ⋅ X = B . and the adjoint adjA . where matrix X is the same size as matrix B .53) we obtain the solution as follows.6111 0.29 = 29 ⁄ 18 = 1. this is MATLAB’s solution of example. we find the determinant detA . this function is not included in the MATLAB Run−Time Function Library List that is used with the Simulink Embedded MATLAB Function block. The MATLAB user− defined function file below can be used to compute the determinant of a 2 × 2 matrix. while MATLAB has the built−in function det(A) for computing the determinant of a matrix A. it is easier to use the matrix left division operation X = A \ B .2778 As stated earlier.9444 1.61 .adjA = ----= ----------18 detA –5 7 1 and by (4. and multiply A–1 by B .53) –1 or x1 x2 x3 2 = 1 3 3 1 2 3 1 2 9 6 8 (4. 1. 1 2 3. However. For this % Observe that B is a column vector –1 A=[2 3 1.54) Next. B=[9 6 8]'. X=A \ B 4−26 Numerical Analysis Using MATLAB® and Excel®.7 1 –5 .55) To verify our results.Chapter 4 Matrices and Determinants X = A B –1 (4. X = 1.94 1 –5 7 9 35 35 ⁄ 18 1 1 . A userdefined function to compute the inverse of an n × n is presented in Chapter 14. 3 1 2]. we could use the MATLAB inv(A) function. 1 –5 7 detA = 18 and adjA = 7 1 – 5 –5 7 1 Therefore.28 –5 7 1 8 5 x3 x1 (4. Third Edition Copyright © Orchard Publications .

I 2. R –1 219 135 81 1 1 . where 10 – 9 0 100 R = – 9 20 – 9 .17 10I 1 – 9I 2 = 100 0 0 – 9I 1 + 20I 2 – 9I 3 = – 9I 2 + 15I 3 = (4.135 150 90 .56) Use the inverse matrix method to compute the values of the currents I 1. Solution: For this example.17 For the electric circuit of Figure 4. This is found from the relation R –1 1 = ----------adjR detR (4. Circuit for Example 4.58) Then. For this example.57) Therefore.1.1.Solution of Simultaneous Equations with Matrices Example 4. and I 3 . the matrix equation is RI = V or I = R–1 V . we find that 219 135 81 detR = 975. adjR = 135 150 90 81 90 119 (4. the mesh equations are 1Ω 2Ω 9Ω 2Ω 9Ω 4Ω − I 1 V = 100 v + I2 I3 Figure 4.adjR = -------= ----------975 detR 81 90 119 and Numerical Analysis Using MATLAB® and Excel®. we find the determinant and the adjoint of R . Third Edition Copyright © Orchard Publications 4−27 . V = 0 0 – 9 15 0 I1 and I = I 2 I3 The next step is to find R–1 .

122 8.3077 9 10 Figure 4. R(2. −9 20 −9.092 I= 13.154 0. we can use Microsoft Excel’s MINVERSE (Matrix Inversion) and MMULT (Matrix Multiplication) functions. V=[100 0 0]'.138 0.846 8 0.8462 8. Then. I=R\V I = 22. we enter the elements of matrix R as shown in Figure 4.85 I = I 2 = -------= -------0 975 975 81 90 119 0 81 8.135 = 13.225 0. R(1.083 22.3077 Spreadsheets also have the capability of solving simultaneous equations using the inverse matrix method.1)=−9. 0 −9 15]. % No need to make entry for A(1.8462 8. For instance. V=[100 0 0]'. the above script could also have been written as: R(1.2.17 with a spreadsheet 4−28 Numerical Analysis Using MATLAB® and Excel®.3077 We can also use subscripts to address the individual elements of the matrix.3)=15. say B3:D5. R(2.46 1 100 .2)=20. R(2.135 150 90 .31 I3 I1 Check with MATLAB: R=[10 −9 0.3)=−9. We start with a blank spreadsheet and in a block of cells.2)=−9. we enter the elements of matrix V in G3:G5.4615 13.462 -1 R = 0. Accordingly. Third Edition Copyright © Orchard Publications . to obtain the values of the three currents in Example 4.092 0.138 0. I=R\V I = 22.4615 13.1)=10.2)=−9.2. The procedure is as follows: 1.083 0. Solution of Example 4.3) since it is zero.Chapter 4 Matrices and Determinants 219 135 81 100 219 22. A B C D E F G H 1 Spreadsheet for Matrix Inversion and Matrix Multiplication 2 3 10 -9 0 100 4 R= -9 20 -9 V= 0 5 0 -9 15 0 6 7 0. R(3.17. R(3.

We format this block for number display with three decimal places. The values of I then appear in G7:G9.61) Compute. Next. We observe that R–1 appears in these cells.+ -------------. With this range highlighted and making sure that the cell marker is in B7. we compute and display the inverse of R.= 0 – j100 100 50 (4.Solution of Simultaneous Equations with Matrices 2.59) and the voltages V 1 and V 2 can be computed from the nodal equations V 1 – 170 ∠0 ° V 1 – V 2 V 1 – 0 ------------------------------.+ -------------. and express the current I x in both rectangular and polar forms by first simplifying like Numerical Analysis Using MATLAB® and Excel®. we type the formula =MININVERSE(B3:D5) and we press the Crtl−Shift−Enter keys simultaneously. that is. we choose the block of cells G7:G9 for the values of the current I . and with the cell marker positioned in G7. we type the formula =MMULT(B7:D9. we highlight them.18 V1 – V2 I X = -----------------R3 (4.3. the current I X can be found from the relation VS + 85 Ω V1 j200 Ω R1 C IX R2 −j100 Ω V2 50 Ω − 170∠0° R3 = 100 Ω L Figure 4.+ -----------------.18 For the phasor circuit of Figure 4. Example 4. We choose B7:D9 for the elements of this inverted matrix. Circuit for Example 4. Third Edition Copyright © Orchard Publications 4−29 .60) (4. R–1 .3. As before. 3. Now.= 0 85 100 j200 V 2 – 170 ∠0 ° V 2 – V 1 V 2 – 0 ------------------------------.G3:G5) and we press the Crtl−Shift−Enter keys simultaneously.+ -----------------.

V . disp(' ------------------'). IX=(V(1)−V(2))/R3 % Compute the value of IX IX = 0. where Y = admit tan ce . and I = current . % Define Y. we find I X from R3=100.01 V 1 + ( 0. For the polar form we use magIX=abs(IX) magIX = % Compute the magnitude of IX 4−30 ⎧ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎪ ⎩ ⎧ ⎨ ⎩ Numerical Analysis Using MATLAB® and Excel®. and then writing the above relations in matrix form as YV = I .V(1).01V 2 = 2 – 0. −0. Y=[0. I=[2. The script is shown below.01j]. % Display V1 and V2 with dash line underneath fprintf('%9.62) in matrix form as 0.0218 – j0.0590i and this is the rectangular form of I X .01.01 Y V1 V2 V = 2 j1.0218 – j0.005 – 0.01 0.005 ) V 1 – 0.63). Simplifying and rearranging the nodal equations of (4. and find V fprintf('\n').0. % Insert a line disp(' V1 V2').V(2)) % Display values of V1 and V2 in tabular form fprintf('\n')% Insert another line V1 V2 -----------------104. we obtain ( 0.0218−0.3f %9.7j].7 ⎧ ⎨ ⎩ I (4.60) and (4.7 Next.62) where the matrices Y . and I are as indicatedin (4. 1.005j −0. V=Y\I.01 0. we write (4.5149 .01 ) V 2 = j1. and to do all other computations. Solution: The elements of the Y matrix are the coefficients of V 1 and V 2 . V = voltage .01 – 0.905 53. collecting.Chapter 4 Matrices and Determinants terms.61).03 + j0. We will use MATLAB to compute the voltages V 1 and V 2 .3f\n'.03+0.03 + j0.416 Next. I. Third Edition Copyright © Orchard Publications .

Numerical Analysis Using MATLAB® and Excel®. and thus we cannot use them to compute matrices that include complex numbers in their elements. Third Edition Copyright © Orchard Publications 4−31 . in polar form I X = 0.5183 thetaIX=angle(IX)*180/pi % Compute angle theta in degrees thetaIX = -6.5326 Therefore.53 ° Spreadsheets have limited capabilities with complex numbers.Solution of Simultaneous Equations with Matrices 0.518 ∠– 6.

then multiplication of a matrix A by the scalar k .. 4−32 Numerical Analysis Using MATLAB® and Excel®. The product A ⋅ B will then be an m × n matrix. to obtain the product A ⋅ B . Thus. that is. • Division of one matrix by another. is not defined. their sum (dif- ference) will be another matrix C with the same order as A and B . we add these products. If A = a ij and B = b ij are conformable for addition (subtraction). m j = 1 . • Two matrices A = a ij and B = b ij are equal. i. • Two matrices A and B are said to be conformable for multiplication A ⋅ B in that order. the product A ⋅ B (but not B ⋅ A ) is conformable for multiplication only if A is an m × p and matrix B is an p × n matrix. A matrix of m rows and n columns is said to be of m × n order matrix. 3. If m = n . only when the number of columns of matrix A is equal to the number of rows of matrix B . if they are of the same order m × n . and not [ k ] which is a 1 × 1 matrix. then. • For matrix multiplication. n • Two matrices are said to be conformable for addition (subtraction). That is.e.12 Summary • A matrix is a rectangular array of numbers whose general form is a 11 a 12 a 13 … a 1n a 21 a 22 a 23 … a 2n A = a 31 a 32 a 33 … a 3n … … … … … a m1 a m2 a m3 … a mn The numbers a ij are the elements of the matrix where the index i indicates the row. if and only if a ij = b ij i = 1. …. we multiply each element of a row of A by the corresponding element of a column of B . is the multiplication of every element of A by k . the matrix is said to be a square matrix of order m. 2 . ….Chapter 4 Matrices and Determinants 4. and j indicates the column in which each element is positioned. the operation is row by column. C = A ± B = [ a ij ± b ij ] • If k is any scalar (a positive or negative number). Third Edition Copyright © Orchard Publications . 2. A = B . where each element of C is the sum (difference) of the corresponding elements of A and B . 3 .

Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications 4−33 . • The MATLAB eye(n) function displays an n × n identity matrix and the eye(size(A)) function displays an identity matrix whose size is the same as matrix A . We must use this procedure to find the determinant of a matrix A of order 4 or higher. • The transpose of a matrix A . the value of the determinant of A is the sum of the products obtained by multiplying each element of any row or any column by its cofactor. is called an identity matrix I . A scalar matrix with k = 1 . • Let A be a square matrix of any size. is the matrix that is obtained when the rows and columns of matrix A are interchanged. and jth column. • The determinant of a square matrix A where a 11 a 12 a 13 … a 1n a 21 a 22 a 23 … a 2n A = a a a …a 31 32 33 3n … … … … … a n1 a n2 a n3 … a nn T is denoted as detA and it is defined as detA = a 11 a 22 a 33 … a nn + a 12 a 23 a 34 … a n 1 + a 13 a 24 a 35 … a n 2 + … – a n 1 … a 22 a 13 … – a n 2 … a 23 a 14 – a n 3 … a 24 a 15 – … • If from a matrix A be defined as a 11 a 12 a 13 … a 1n a 21 a 22 a 23 … a 2n A = a a a …a 31 32 33 3n … … … … … a n1 a n2 a n3 … a nn we remove the elements of its ith row. the determinant of the remaining n – 1 square matrix is called the minor of determinant A . • The signed minor ( – 1 ) i+j M ij is called the cofactor of a ij and it is denoted as α ij .Summary • A scalar matrix is a square matrix where a 11 = a 22 = a 33 = … = a nn = k and k is a scalar. denoted as A . and it is denoted as M ij .

4−34 Numerical Analysis Using MATLAB® and Excel®. by substitution to another equation with two unknowns. denoted as adjA . This procedure is repeated until all unknowns are found. If all the elements of one row or column are m times the corresponding elements of another row or column. and z can be found from the relations D1 x = ----- Δ D2 y = ----- Δ D3 z = ----- Δ provided that the determinant Δ (delta) is not zero.Chapter 4 Matrices and Determinants • Some useful properties of determinants are: a. the determinant is zero. • We can find the unknowns in a system of two or more equations also by the Gaussian elimination method. If all elements of one row or one column are zero. • Cramer’s rule states that if a system of equations is defined as a 11 x + a 12 y + a 13 z = A a 21 x + a 22 y + a 23 z = B a 31 x + a 32 y + a 33 z = C and we let Δ a 11 a 12 a 13 = a 21 a 22 a 23 a 31 a 32 a 33 D1 = A a 11 a 13 B a 21 a 23 C a 31 a 33 D2 = a 11 A a 13 a 21 B a 23 a 31 C a 33 D3 = a 11 a 12 A a 21 a 22 B a 31 a 32 C the unknowns x . Subsequently. the objective is to eliminate one unknown at a time. we can find the second unknown. substitution of the two values found can be made into an equation with three unknowns from which we can find the value of the third unknown. • If A is an n square matrix and α ij is the cofactor of a ij . If two rows or two columns of a matrix are identical. With this method. Third Edition Copyright © Orchard Publications . the determinant is zero. the determinant is zero. c. Then. y . This can be done by multiplying the terms of any of the equations of the system by a number such that we can add (or subtract) this equation to another equation in the system so that one of the unknowns will be eliminated. the adjoint of A . is defined as the n square matrix below. b.

A is called the inverse of B . we can use the relation X= A –1 B to solve any set of simultaneous equations that have solutions. • The matrix left division operation is defined as X = A \ B . that is. where matrix X is the same size as matrix B .Summary α 11 α 21 α 31 … α n1 α 12 α 22 α 32 … α n2 adjA = α α α … α 13 23 33 n3 … … … … … α 1n α 2n α 3n … α nn • An n square matrix A is called singular if detA = 0 . denoted as B = A –1 . we can compute its inverse from the relation A –1 1 = ----------adjA detA • Multiplication of a matrix A by its inverse A –1 produces the identity matrix I . We refer to this method as the inverse matrix method of solution of simultaneous equations. –1 • We can use Microsoft Excel’s MINVERSE (Matrix Inversion) and MMULT (Matrix Multiplication) functions. Third Edition Copyright © Orchard Publications 4−35 . X is a matrix (a column vector) whose elements are the unknowns and A and X are conformable for multiplication. we cannot use them to compute matrices that include complex numbers in their elements. A = B –1 • If a matrix A is non-singular. where I is the identity matrix. B is called the inverse of A . AA –1 = I or A A = I –1 • If A and B are matrices whose elements are known. • If A and B are n square matrices such that AB = BA = I . this is MATLAB’s solution of A B for the matrix equation A ⋅ X = B . and likewise. A is called non-singular. if detA ≠ 0 . However. Numerical Analysis Using MATLAB® and Excel®. that is. to solve any set of simultaneous equations that have solutions.

det ( A ⋅ B ) f. 6. a. Solve the following system of equations using the inverse matrix method. C ⋅ A g. x 1 – 2x 2 + x 3 = – 4 – 2x 1 + 3x 2 + x 3 = 9 3x 1 + 4x 2 – 5x 3 = 0 5. Verify your answers with Excel or MATLAB. A ⋅ C c. detA b. Perform the following computations. B – D h. C – D 2.13 Exercises For Exercises 1 through 3 below. – x 1 + 2x 2 – 3x 3 + 5x 4 = 14 x 1 + 3x 2 + 2x 3 – x 4 = 9 3x 1 – 3 x 2 + 2x 3 + 4x 4 = 19 4x 1 + 2x 2 + 5x 3 + x 4 = 27 7. C + D e. A + C c. a. if possible. A ⋅ B b. Verify your answers with Excel or MATLAB. A – C g. Verify your answers with Excel or MATLAB. B ⋅ D d. if possible. detC d. detD e. Perform the following computations. Solve the following system of equations using Cramer’s rule. D ⋅ A h. A + B b. C ⋅ D e. Perform the following computations. B . A – B f. detB c. the matrices A .Chapter 4 Matrices and Determinants 4. Third Edition Copyright © Orchard Publications . D 3. B + D d. · ⋅C a. Repeat Exercise 4 using the Gaussian elimination method. Verify your answers with Excel or MATLAB. det ( A ⋅ C ) 4. Verify your answers with Excel or MATLAB. if possible. C and D are defined as: 1 –1 –4 A = 5 7 –2 3 –5 6 5 9 –3 B = –2 8 2 7 –4 6 4 6 C= – 3 8 5 –2 D = 1 –2 3 –3 6 –4 1. B ⋅ A f. x1 –3 1 3 4 3 1 –2 ⋅ x2 = –2 0 2 3 5 x3 4−36 Numerical Analysis Using MATLAB® and Excel®. Use the MATLAB det(A) function to find the unknowns of the system of equations below.

Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications 4−37 .Exercises 8. Use Excel to find the unknowns for the system 2 2 –1 2 4 3 –4 1 3 –4 –2 2 x1 1 –2 x 10 3 ⋅ 2 = x3 – 14 2 7 1 x4 Verify your answers with the MATLAB left division operation.

A + B = 5 – 2 6 8 –7 1+5 –1+9 –4–3 = 3 15 0 7+8 –2+2 10 – 9 12 3+7 –5–4 6+6 b. A – B = 5 + 2 3–7 1–5 –1–9 7–8 –5+4 – 4 – 10 – 1 –4+3 = 7 –1 –4 –2–2 –4 –1 0 6–6 f. B=[5 9 −3. C + D not conformable for addition e. A⋅B = 1 × 5 + ( –1 ) × ( –2 ) + ( –4 ) × 7 5 × 5 + 7 × ( –2 ) + ( –2 ) × 7 3 × 5 + ( –5 ) × ( –2 ) + 6 × 7 – 21 17 – 29 – 3 109 – 13 67 – 37 17 1 × 9 + ( –1 ) × 8 + ( –4 ) × ( –4 ) 5 × 9 + 7 × 8 + ( –2 ) × ( –4 ) 3 × 9 + ( –5 ) × 8 + 6 × ( –4 ) 1 × ( –3 ) + ( –1 ) × 2 + ( –4 ) × 6 5 × ( –3 ) + 7 × 2 + ( –2 ) × 6 3 × ( –3 ) + ( –5 ) × 2 + 6 × 6 a. A – C not conformable for subtraction g. Third Edition Copyright © Orchard Publications . = Check with MATLAB: A=[1 −1 −4. B – D not conformable for subtraction h. C – D not conformable for subtraction 2. A + C not conformable for addition c. a. A ⋅ C = 17 109 -37 -29 -13 17 1 × 6 + ( –1 ) × 8 + ( –4 ) × ( –2 ) 5 × 6 + 7 × 8 + ( –2 ) × ( –2 ) 3 × 6 + ( –5 ) × 8 + 6 × ( –2 ) 1 × 4 + ( –1 ) × ( –3 ) + ( –4 ) × 5 5 × 4 + 7 × ( –3 ) + ( –2 ) × 5 3 × 4 + ( –5 ) × ( –3 ) + 6 × 5 – 13 6 = – 11 90 57 – 34 c. 7 −4 6]. −2 8 2.Chapter 4 Matrices and Determinants 4. 5 7 −2. 3 −5 6]. B + D not conformable for addition d. A*B ans = -21 -3 67 b.14 Solutions to End−of−Chapter Exercises 1. ⋅ D = ( – 3 ) × 1 + 8 × ( – 3 ) 4 × ( –2 ) + 6 × 6 4 × 3 + 6 × ( –4 ) ( –3 ) × ( –2 ) + 8 × 6 ( –3 ) × 3 + 8 × ( –4 ) 5 × 1 + ( –2 ) × ( –3 ) 5 × ( –2 ) + ( –2 ) × 6 5 × 3 + ( –2 ) × ( –4 ) 4 × 1 + 6 × ( –3 ) – 14 28 – 12 = – 27 54 – 41 11 – 22 23 4−38 Numerical Analysis Using MATLAB® and Excel®. B ⋅ D not conformable for multiplication d.

matrix must be square d. et ( A ⋅ B ) = detA ⋅ detBand from parts (a) and (b). = h. 1 –1 –4 1 –1 detA = 5 7 – 2 5 7 3 –5 6 3 –5 = 1 × 7 × 6 + ( –1 ) × ( –2 ) × 3 + ( –4 ) × 5 × ( –5 ) – [ 3 × 7 × ( –4 ) + ( –5 ) × ( –2 ) × 1 + 6 × 5 × ( –1 ) ] = 42 + 6 + 100 – ( – 84 ) – 10 – ( – 30 ) = 252 5 9 –3 5 9 detB = – 2 8 2 – 2 8 7 –4 6 7 –4 = 5 × 8 × 6 + 9 × 2 × 7 + ( –3 ) × ( –2 ) × ( –4 ) – [ 7 × 8 × ( –3 ) + ( –4 ) × 2 × 5 + 6 × ( –2 ) × 9 ] = 240 + 126 – 24 – ( – 168 ) + 40 – ( – 108 ) = 658 b. C ⋅ A not conformable for multiplication D⋅A = 1 × 1 + ( –2 ) × 5 + 3 × 3 1 × ( –1 ) + ( –2 ) × 7 + 3 × ( –5 ) 1 × ( –4 ) + ( –2 ) × ( –2 ) + 3 × 6 ( –3 ) × 1 + 6 × 5 + ( –4 ) × 3 ( –3 ) × ( –1 ) + 6 × 7 + ( –4 ) × ( –5 ) ( –3 ) × ( –4 ) + 6 × ( –2 ) + ( –4 ) × 6 0 – 30 18 15 65 – 24 1 × 4 + ( –2 ) × ( –3 ) + 3 × 5 ( –3 ) × 4 + 6 × ( –3 ) + ( –4 ) × 5 1 × 6 + ( –2 ) × 8 + 3 × ( –2 ) = 25 – 16 ( –3 ) × 6 + 6 × 8 + ( –4 ) × ( –2 ) – 50 38 g. detC does not exist. det ( A ⋅ B ) = 252 × 658 = 165816 f. c. Third Edition Copyright © Orchard Publications 4−39 . D ⋅ C = 3. detD does not exist. f. a. matrix must be square e. det ( A ⋅ C ) does not exist because detC does not exist Numerical Analysis Using MATLAB® and Excel®.Solutions to End−of−Chapter Exercises 5 × 1 + 9 × 5 + ( –3 ) × 3 B ⋅ A = 5 × ( –1 ) + 9 × 7 + ( –3 ) × ( –5 ) 5 × ( –4 ) + 9 × ( –2 ) + ( –3 ) × 6 41 73 – 56 = 44 48 4 5 – 65 16 ( –2 ) × 1 + 8 × 5 + 2 × 3 ( –2 ) × ( –1 ) + 8 × 7 + 2 × ( –5 ) ( –2 ) × ( –4 ) + 8 × ( –2 ) + 2 × 6 7 × 1 + ( –4 ) × 5 + 6 × 3 7 × ( –1 ) + ( –4 ) × 7 + 6 × ( –5 ) 7 × ( –4 ) + ( –4 ) × ( –2 ) + 6 × 6 e.

= -------.= –1 Δ – 22 D2 – 44 x 2 = -----. x 1 – 2x 2 + x 3 = – 4 (1) – 2x 1 + 3x 2 + x 3 = 9 (2) 3x 1 + 4x 2 – 5x 3 = 0 (3) Multiplication of (1) by 2 yields 2x 1 – 4x 2 + 2x 3 = – 8 (4) Addition of (2) and (4) yields – x 2 + 3x 3 = 1 (5) 4−40 Numerical Analysis Using MATLAB® and Excel®.= -------.= 1 Δ – 22 5. Third Edition Copyright © Orchard Publications .Chapter 4 Matrices and Determinants 4.= -------. 1 –2 1 1 –2 Δ = –2 3 1 –2 3 3 4 –5 3 4 = 1 × 3 × ( –5 ) + ( –2 ) × 1 × 3 + 1 × ( –2 ) × 4 – [ 3 × 3 × 1 + 4 × 1 × 1 + ( –5 ) × ( –2 ) × ( –2 ) ] = – 15 – 6 – 8 – 9 – 4 + 20 = – 22 –4 –2 1 4 –2 9 3 1 9 3 0 4 –5 0 4 D1 = = –4 × 3 × ( –5 ) + ( – 2 ) × 1 × 0 + 1 × 9 × 4 – [ 0 × 3 × 1 + 4 × 1 × 4 + ( –5 ) × 9 × ( –2 ) ] = 60 + 0 + 36 – 0 + 16 – 90 = 22 1 –4 1 1 –4 D2 = –2 9 1 –2 9 3 0 –5 3 0 = 1 × 9 × ( –5 ) + ( –4 ) × 1 × 3 + 1 × ( –2 ) × 0 – [ 3 × 9 × 1 + 0 × 1 × 1 + ( –5 ) × ( –2 ) × ( –4 ) ] = – 45 – 12 – 0 – 27 – 0 + 40 = – 44 1 –2 –4 1 –2 D3 = –2 3 9 –2 3 3 4 0 3 4 = 1 × 3 × 0 + ( –2 ) × 9 × 3 + ( –4 ) × ( –2 ) × 4 – [ 3 × 3 × ( –4 ) + 4 × 9 × 1 + 0 × ( –2 ) × ( –2 ) ] = 0 – 54 + 32 + 36 – 36 – 0 = – 22 D1 22 x 1 = -----.= 2 Δ – 22 D3 – 22 x 3 = -----.

Solutions to End−of−Chapter Exercises Multiplication of (1) by – 3 yields – 3 x 1 + 6x 2 – 3x 3 = 12 (6) Addition of (3) and (6) yields 10x 2 – 8x 3 = 12 (7) Multiplication of (5) by 10 yields – 10x 2 + 30x 3 = 10 (8) Addition of (7) and (8) yields 22x 3 = 22 (9) or x 3 = 1 (10) Substitution of (10) into (7) yields 10x 2 – 8 = 12 (11) or x 2 = 2 (12) and substitution of (10) and (12) into (1) yields x 1 – 4 + 1 = – 4 (13) or x 1 = – 1 (14) 6. 9 3 2 −1. D2=[−1 14 −3 5. 27 2 5 1]. 4 27 5 1]. 4 2 5 1]. 3 −3 2 4. 1 3 2 9. 3 19 2 4. 1 9 2 −1. x4=det(D4)/det(Delta) x1=1 x2=2 x3=3 x4=4 Numerical Analysis Using MATLAB® and Excel®. 19 −3 2 4. 1 3 9 −1. x2=det(D2)/det(Delta). Third Edition Copyright © Orchard Publications 4−41 . Delta=[−1 2 −3 5. x1=det(D1)/det(Delta).. 4 2 5 27]. 4 2 27 1].. D1=[14 2 −3 5. D3=[−1 2 14 5. 3 −3 2 19. D4=[−1 2 −3 14. x3=det(D3)/det(Delta). 3 −3 19 4. 1 3 2 −1..

50 1.Chapter 4 Matrices and Determinants 7.⋅ adjA = -------.50 – 3.50 4−42 Numerical Analysis Using MATLAB® and Excel®. 1 3 4 1 3 detA = 3 1 – 2 3 1 2 3 5 2 3 = 1 × 1 × 5 + 3 × ( –2 ) × 2 + 4 × 3 × 3 – [ 2 × 1 × 4 + 3 × ( – 2 ) × 1 + 5 × 3 × 3 ] = 5 – 12 + 36 – 8 + 6 – 45 = – 18 11 – 3 – 10 adjA = – 19 – 3 14 7 3 –8 –1 11 – 3 – 10 – 11 ⁄ 18 3 ⁄ 18 10 ⁄ 18 1 1 .⋅ – 19 – 3 14 = = ----------19 ⁄ 18 3 ⁄ 18 – 14 ⁄ 18 detA – 18 – 7 ⁄ 18 – 3 ⁄ 18 8 ⁄ 18 7 3 –8 – 11 ⁄ 18 3 ⁄ 18 10 ⁄ 18 – 3 19 ⁄ 18 3 ⁄ 18 – 14 ⁄ 18 – 2 = – 7 ⁄ 18 – 3 ⁄ 18 8 ⁄ 18 0 33 ⁄ 18 – 6 ⁄ 18 + 0 – 57 ⁄ 18 – 6 ⁄ 18 + 0 = 21 ⁄ 18 + 6 ⁄ 18 + 0 27 ⁄ 18 – 63 ⁄ 18 = 27 ⁄ 18 A x1 X = x2 = x3 1. Third Edition Copyright © Orchard Publications .

A\B ans = -11. A=[2 4 3 −2. 2 −2 2 1].0000 9. −1 3 −4 2. 2 −4 1 3. B=[1 10 −14 7]'. Third Edition Copyright © Orchard Publications 4−43 .0000 Numerical Analysis Using MATLAB® and Excel®.5000 1.5000 12.Solutions to End−of−Chapter Exercises 8.

3) we obtain v C ( t ) = It + k where k represents the constants of integration of both sides. Numerical Analysis Using MATLAB® and Excel®.4) Integrating both sides of (5. Solutions of differential equations with numerical methods is discussed in Chapter 9. dv C = Idt (5.Chapter 5 Differential Equations. is constant. 5.= I dt and by separation of the variables. and State Equations T his chapter is a review of ordinary differential equations and an introduction to state variables and state equations. i C ( t ) = I = cons tan t (5. Example 5. v C ( t ) is the voltage across the capacitor.1) where i C ( t ) is the current through the capacitor. as a function of time.1 Simple Differential Equations In this section we present two simple examples to show the importance of differential equations in engineering applications.3) (5. State Variables. Solution: It is given that the current.1) we obtain dv C -------. Find the voltage v C across this capacitor as a function of time given that the voltage at some reference time t = 0 is V 0 . that is.2) By substitution of (5. and the constant C is the capacitance in farads (F).2) into (5.1 The current and voltage in a capacitor are related by dv C i C ( t ) = C -------dt (5. For this example C = 1 F and the capacitor is being charged by a constant current I . Third Edition Copyright © Orchard Publications 5−1 .

**Chapter 5 Differential Equations, State Variables, and State Equations
**

We can find the value of the constant k by making use of the initial condition, i.e., at t = 0 , v C = V 0 and (5.4) then becomes

V0 = 0 + k

(5.5) (5.6)

**or k = V 0 , and by substitution into (5.4),
**

v C ( t ) = It + V 0

This example shows that when a capacitor is charged with a constant current, a linear voltage is produced across the terminals of the capacitor. Example 5.2 Find the current i L ( t ) through an inductor whose slope at the coordinate ( t, i L ) is cos t and the current i L passes through the point ( π ⁄ 2 ,1 ) . Solution: We are given that

di L ------- = cos t dt

(5.7) (5.8) (5.9)

**By separating the variables we obtain
**

di L = cos tdt

**and integrating both sides we obtain
**

i L ( t ) = sin t + k

where k represents the constants of integration of both sides. We find the value of the constant k by making use of the initial condition. For this example, ω = 1 and thus at ω t = t = π ⁄ 2 , i L = 1 . With these values (5.9) becomes

--+k 1 = sin π 2

(5.10)

**or k = 0 , and by substitution into (5.9),
**

i L ( t ) = sin t

(5.11)

5.2 Classification

Differential equations are classified by: 1. Type - Ordinary or Partial

5−2

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

Classification

2. Order - The highest order derivative which is included in the differential equation 3. Degree - The exponent of the highest power of the highest order derivative after the differential equation has been cleared of any fractions or radicals in the dependent variable and its derivatives For example, the differential equation

2 ⎛ d 4 y⎞ ⎛ d 3 y⎞ ⎛ d 2 y⎞ dy⎞ 8 – 2x y - + --------------------------⎟ + 3 ⎛ ----= ye ⎜ 4⎟ + 5 ⎜ 3⎟ + 6 ⎜ ------2 ⎝ ⎠ 3 dx ⎝ dx ⎠ ⎝ dx ⎠ ⎝ dx ⎠ x +1 2 4 6

is an ordinary differential equation of order 4 and degree 2 . If the dependent variable y is a function of only a single variable x , that is, if y = f ( x) , the differential equation which relates y and x is said to be an ordinary differential equation and it is abbreviated as ODE. The differential equation

2 dy d y - + 2 = 5 cos 4t ------- + 3 ----2 dt dt

**is an ODE with constant coefficients. The differential equation
**

d y ----- + ( x2 – n2 ) = 0 - + x dy x 2 ------2 dt dt

2

is an ODE with variable coefficients. If the dependent variable y is a function of two or more variables such as y = f ( x, t ) , where x and t are independent variables, the differential equation that relates y , x , and t is said to be a partial differential equation and it is abbreviated as PDE. An example of a partial differential equation is the well-known one-dimensional wave equation shown below.

2 2∂ y ∂ y ------- = a ------2 2 ∂x ∂t 2

Most engineering problems are solved with ordinary differential equations with constant coefficients; however, partial differential equations provide often quick solutions to some practical applications as illustrated with the following three examples.

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

5−3

**Chapter 5 Differential Equations, State Variables, and State Equations
**

Example 5.3 The equivalent resistance R T of three resistors R 1 , R 2 , and R 3 in parallel is obtained from

1 1 1 1 ------ = ----- + ----- + ----RT R1 R2 R3

Given that initially R 1 = 5 Ω , R 2 = 20 Ω , and R 3 = 4 Ω , compute the change in R T if R 2 is increased by 10 % and R 3 is decreased by 5 % while R 1 does not change. Solution: The initial value of the equivalent resistance is R T = 5 || 20 || 4 = 2 Ω . Now, we treat R 2 and R 3 as constants and differentiating R T with respect to R 1 we obtain

∂ RT ⎛ RT ⎞ 2 1 1 ∂ RT - = – ----– ------ --------- or --------- = -----2 2 ∂ R1 ⎝ R1 ⎠ R1 RT ∂ R1

Similarly,

∂ RT ⎛ RT ⎞ 2 ∂ RT ⎛ RT ⎞ 2 --------- = -----and --------- = -----∂ R2 ⎝ R2 ⎠ ∂ R3 ⎝ R3 ⎠

**and the total differential dRT is
**

RT ⎞ 2 ∂ RT ∂ RT ∂ RT RT 2 RT 2 - dR 1 + --------- dR 2 + --------- dR 3 = ⎛ -----R T = --------dR 1 + ⎛ ------ ⎞ dR 2 + ⎛ ------ ⎞ dR ⎝ R1 ⎠ ⎝ R2 ⎠ ⎝ R3 ⎠ ∂ R1 ∂ R2 ∂ R3

**By substitution of the given numerical values we obtain
**

2 2 2- ⎞ 2 2 2 - ⎞ ( – 0.2 ) = 0.02 – 0.05 = – 0.03 - ⎞ ( 0 ) + ⎛ ----dR T = ⎛ -( 2 ) + ⎛ -⎝ 20 ⎠ ⎝4⎠ ⎝5 ⎠

Therefore, the eequivalent resistance decreases by 3 % . Example 5.4 In a series RC electric circuit that is excited by a sinusoidal voltage, the magnitude of the impedance Z is computed from Z =

R + X C . Initially, R = 4 Ω and X C = 3 Ω . Find the change

2 2

in the impedance Z if the resistance R is increased by 0.25 Ω ( 6.25 % ) and the capacitive reactance X C is decreased by 0.125 Ω ( – 4.167% ).

5−4

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

Classification

Solution:

∂Z ∂Z - and --------- ; then we compute the change in impedance We will first find the partial derivatives -----∂R ∂ XC

**from the total differential dZ . Thus,
**

XC ∂Z R ∂Z ------ = --------------------------- and --------- = --------------------------∂R ∂ XC 2 2 2 2 R + XC R + XC

and

R dR + X C dX C ∂Z ∂Z - dR + ---------- dX C = --------------------------------------dZ = -----∂ XC ∂R 2 2 R + XC

**and by substitution of the given values
**

1 – 0.375 4 ( 0.25 ) + 3 ( – 0.125 ) - = ------------------------- = 0.125 dZ = ---------------------------------------------------5 2 2 4 +3

**Therefore, if R increases by 6.25 % and X C decreases by 4.167% , the impedance Z increases by
**

4.167% .

Example 5.5 A light bulb is rated at 120 volts and 75 watts. If the voltage decreases by 5 volts and the resistance of the bulb is increased by 8 Ω , by how much will the power change? Solution: At V = 120 volts and P = 75 watts, the bulb resistance is

----- = 120 ----------- = 192 Ω R = V P 75

2 2

and since

V∂P V - then -----∂P - = – ---------- and ------ = 2V P = ----2 ∂R R ∂V R R

2 2

and the total differential is

2 2 ( 120 ) ∂P ∂P 2V 120 - ( 8 ) = – 9.375 V - dR = 2 ---------------- ( – 5 ) – ---------- dV + ------ dR = ------ dV – ----dP = -----2 2 192 ∂V ∂R R R 192

That is, the power will decrease by 9.375 watts.

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

5−5

**Chapter 5 Differential Equations, State Variables, and State Equations 5.3 Solutions of Ordinary Differential Equations (ODE)
**

A function y = f ( x) is a solution of a differential equation if the latter is satisfied when y and its derivatives are replaced throughout by f ( x) and its corresponding derivatives. Also, the initial conditions must be satisfied. For example a solution of the differential equation

d y -------+y = 0 2 dx

2

is

y = k 1 sin x + k 2 cos x

since y and its second derivative satisfy the given differential equation. Any linear, time-invariant system can be described by an ODE which has the form

d y d y dy - + a0 y - + a n – 1 --------------- + … + a 1 ----a n -------n n–1 dt dt dt d x d x dx - + b0 x - + b m – 1 ---------------- + … + b 1 ----b m --------m n–1 dt = dt dt Excitation ( Forcing ) Function x ( t ) NON – HOMOGENEOUS DIFFERENTIAL EQUATION

m m–1 n n–1

(5.12)

If the excitation in (B12) is not zero, that is, if x ( t ) ≠ 0 , the ODE is called a non-homogeneous ODE. If x ( t ) = 0 , it reduces to:

d y y ----- + a0 y = 0 - + an – 1 d --------------- + … + a 1 dy a n -------n n–1 dt dt dt HOMOGENEOUS DIFFERENTIAL EQUATION

n n–1

⎧ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎩

(5.13)

The differential equation of (5.13) above is called a homogeneous ODE and has n different linearly independent solutions denoted as y 1 ( t ), y 2 ( t ), y 3 ( t ), …, y n ( t ) . We will now prove that the most general solution of (5.13) is:

yH ( t ) = k1 y1 ( t ) + k2 y2 ( t ) + k3 y3 ( t ) + … + kn yn ( t )

(5.14)

where the subscript H on the left side is used to emphasize that this is the form of the solution of the homogeneous ODE and k 1, k 2, k 3, …, k n are arbitrary constants.

5−6

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Solutions of Ordinary Differential Equations (ODE)
**

Proof: Let us assume that y 1 ( t ) is a solution of (5.13); then by substitution,

d y1 d y1 dy 1 - + a0 y1 = 0 ----------------- + … + a 1 ------a + a n ----------n – 1 n n–1 dt dt dt

n n–1

(5.15)

**A solution of the form k 1 y 1 ( t ) will also satisfy (5.13) since
**

d d d - ( k y ) + a0 ( k1 y1 ) - ( k y ) + a n – 1 ------------- ( k 1 y 1 ) + … + a 1 ---a n -----n 1 1 n–1 dt 1 1 dt dt d y1 dy 1 ⎛ d y1 ⎞ - + a 0 y 1⎟ = 0 - + … + a 1 ------= k 1 ⎜ a n ----------+ a n – 1 ----------------n n–1 dt ⎝ dt ⎠ dt

n n–1 n n–1

(5.16)

**If y = y 1 ( t ) and y = y 2 ( t ) are any two solutions, then y = y 1 ( t ) + y 2 ( t ) will also be a solution since
**

d y1 d y1 dy 1 - + a0 y1 = 0 - + … + a 1 ------+ a n – 1 ----------------a n ----------n n–1 dt dt dt

n n–1

and

d y2 d y2 d y2 - + a0 y2 = 0 ----------------- + … + a 1 --------a a n ----------+ n – 1 n n–1 dt dt dt

n n–1

Therefore,

dd da n -----( y + y 2 ) + a n – 1 ------------( y + y 2 ) + … + a 1 ---( y1 + y2 ) + a0 ( y1 + y2 ) n 1 n–1 1 dt dt dt n n–1 d d d - y1 + a0 y1 - y 1 + a n – 1 ------------- y 1 + … + a 1 ---= a n -----n n–1 dt dt dt n n–1 d d d - y2 + a0 y2 = 0 - y 2 + a n – 1 ------------- y 2 + … + a 1 ---+ a n -----n n–1 dt dt dt

n n–1

(5.17)

In general, if

y = k 1 y 1 ( t ), k 2 y 1 ( t ), k 3 y 3 ( t ), …, k n y n ( t )

**are the n solutions of the homogeneous ODE of (5.13), the linear combination
**

y = k1 y1 ( t ) + k2 y1 ( t ) + k3 y3 ( t ) + … + kn yn ( t )

is also a solution. In our subsequent discussion, the solution of the homogeneous ODE, i.e., the complementary

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

5−7

**Chapter 5 Differential Equations, State Variables, and State Equations
**

solution, will be referred to as the natural response, and will be denoted as y N ( t ) or simply y N . The particular solution of a non-homogeneous ODE will be referred to as the forced response, and will be denoted as y F ( t ) or simply y F . Accordingly, we express the total solution of the non-homogeneous ODE of (5.12) as:

y(t) = y

Natural Response

+y

Forced Response

= yN + yF

(5.18)

The natural response y N contains arbitrary constants and these can be evaluated from the given initial conditions. The forced response y F , however, contains no arbitrary constants. It is imperative to remember that the arbitrary constants of the natural response must be evaluated from the total response.

**5.4 Solution of the Homogeneous ODE
**

Let the solutions of the homogeneous ODE

d y d y dy - + a0 y = 0 - + a n – 1 --------------- + … + a 1 ----a n -------n n–1 dt dt dt

n n–1

(5.19)

**be of the form
**

y = ke

st

(5.20)

**Then, by substitution of (5.20) into (5.19) we obtain
**

a n ks e + a n – 1 ks

n st n – 1 st

e + … + a 1 kse + a 0 ke

st

st

st

= 0

or

( an s + an – 1 s

n n–1

+ … + a 1 s + a 0 ) ke

= 0

(5.21)

**We observe that (5.21) can be satisfied when
**

( an s + an – 1 s

n n–1

+ … + a 1 s + a 0 ) = 0 or k = 0

or s = – ∞

(5.22)

but the only meaningful solution is the quantity enclosed in parentheses since the latter two yield trivial (meaningless) solutions. We, therefore, accept the expression inside the parentheses as the only meaningful solution and this is referred to as the characteristic (auxiliary) equation, that is,

⎧ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎩ an s + an – 1 s

n n–1

+ … + a1 s + a0 = 0

(5.23)

Characteristic Equation

Since the characteristic equation is an algebraic equation of an nth-power polynomial, its solutions are s 1, s 2, s 3, …, s n , and thus the solutions of the homogeneous ODE are:

5−8

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Solution of the Homogeneous ODE
**

y 1 = k 1 e , y 2 = k 2 e , y 3 = k 3 e , …, y n = k n e

s1 t s2 t s3 t sn t

(5.24)

Case I − Distinct Roots If the roots of the characteristic equation are distinct (different from each another), the n solutions of (5.23) are independent and the most general solution is:

yN = k1 e

s1 t

+ k2 e

s2 t

+ … + kn e

sn t

(5.25)

FOR DISTINCT ROOTS

Case II − Repeated Roots If two or more roots of the characteristic equation are repeated (same roots), then some of the terms of (5.24) are not independent and therefore (5.25) does not represent the most general solution. If, for example, s 1 = s 2 , then, k1 e

s1 t s t

s1 t

+ k2 e

s2 t

= k1 e

s1 t

+ k2 e

s1 t

= ( k1 + k2 ) e

s1 t

= k3 e

s1 t

and we see that one term of (5.25) is lost. In this case, we express one of the terms of (5.25), say as k 2 te 1 . These two represent two independent solutions and therefore the most general solution has the form:

k2 e yN = ( k1 + k2 t ) e

s1 t

+ k3 e

s3 t

+ … + kn e

sn t

(5.26)

**If there are m equal roots the most general solution has the form:
**

yN = ( k1 + k2 t + … + km t

m–1

)e

s1 t

+ kn – i e

s2 t

+ … + kn e

sn t

(5.27)

FOR M EQUAL ROOTS

Case III − Complex Roots If the characteristic equation contains complex roots, these occur as complex conjugate pairs. Thus, if one root is s 1 = – α + j β where α and β are real numbers, then another root is

s 1 = – α – j β . Then, k1 e

s1 t

+ k2 e

s2 t

= k1 e = e = e = e

– αt + jβt

+ k2 e

–α t –j β t

= e

–α t

( k1 e

jβt

+ k2 e

–j β t

)

–α t –α t –α t

**( k 1 cos β t + jk 1 sin β t + k 2 cos β t – jk 2 sin β t ) [ ( k 1 + k 2 ) cos β t + j ( k 1 – k 2 ) sin β t ] ( k 3 cos β t + k 4 sin β t ) = e
**

–α t

k 5 cos ( β t + ϕ )

FOR TWO COMPLEX CONJUGATE ROOTS

(5.28)

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

5−9

**Chapter 5 Differential Equations, State Variables, and State Equations
**

If (5.28) is to be a real function of time, the constants k 1 and k 2 must be complex conjugates. The other constants k 3 , k 4 , k 5 , and the phase angle ϕ are real constants. The forced response can be found by a. The Method of Undetermined Coefficients or b. The Method of Variation of Parameters We will study the Method of Undetermined Coefficients first.

**5.5 Using the Method of Undetermined Coefficients for the Forced Response
**

For simplicity, we will only consider ODEs of order 2 . Higher order ODEs are discussed in differential equations textbooks. Consider the non-homogeneous ODE

a dy

2 2

d - y + cy = f ( x ) + b ---dt dt

(5.29)

where a , b , and c are real constants. We have learned that the total (complete) solution consists of the summation of the natural and forced responses. For the natural response, if y 1 and y 2 are any two solutions of (5.29), the linear combination

y 3 = k 1 y 1 + k 2 y 2 , where k 1 and k 2 are arbitrary constants, is also a solution, that is, if we know

the two solutions, we can obtain the most general solution by forming the linear combination of y 1 and y 2 . To be certain that there exist no other solutions, we examine the Wronskian Determinant defined below.

y1 y2 d d - y 2 – y 2 ----- y1 ≠ 0 W ( y 1, y 2 ) ≡ d = y1 ----d dx dx ----- y1 ----- y2 dx dx WRONSKIAN DETERMINANT

(5.30)

If (5.30) is true, we can be assured that all solutions of (5.29) are indeed the linear combination of y 1 and y 2 . The forced response is obtained by observation of the right side of the given ODE as it is illustrated by the examples that follow.

5−10

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Using the Method of Undetermined Coefficients for the Forced Response
**

Example 5.6 Find the total solution of the ODE

dy - + 3y = 0 + 4 ----dt dt d y

2 2

(5.31)

subject to the initial conditions y ( 0 ) = 3 and y' ( 0 ) = 4 where y' = dy ⁄ dt Solution: This is a homogeneous ODE and its total solution is just the natural response found from the characteristic equation s + 4s + 3 = 0 whose roots are s 1 = – 1 and s 2 = – 3 . The total response is:

y ( t ) = yN ( t ) = k1 e + k2 e

–t – 3t 2

(5.32)

**The constants k 1 and k 2 are evaluated from the given initial conditions. For this example,
**

y ( 0 ) = 3 = k1 e + k2 e

0 0

or

k1 + k2 = 3

(5.33)

–t – 3t t=0

Also,

dy y' ( 0 ) = 4 = ----dt = – k 1 e – 3k 2 e

t=0

or

– k 1 – 3k 2 = 4

(5.34)

**Simultaneous solution of (5.33) and (5.34) yields k 1 = 6.5 and k 2 = – 3.5 . By substitution into (5.32), we obtain
**

y ( t ) = y N ( t ) = 6.5e – 3.5e

–t – 3t

(5.35)

**Check with MATLAB:
**

y=dsolve('D2y+4*Dy+3*y=0', 'y(0)=3', 'Dy(0)=4')

**y = (-7/2*exp(-3*t)*exp(t)+13/2)/exp(t)
**

pretty(y)

- 7/2 exp(-3 t) exp(t) + 13/2 ----------------------------exp(t) The function y = f ( t ) , of relation (5.35), shown in Figure 5.1, was plotted with the use of the MATLAB script Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

5−11

**Chapter 5 Differential Equations, State Variables, and State Equations
**

y=dsolve('D2y+4*Dy+3*y=0', 'y(0)=3', 'Dy(0)=4'); ezplot(y,[0 5])

3.5 3 2.5 2 1.5 1 0.5 0 13/2 exp(-t)-7/2 exp(-3 t)

0

1

2 t

3

4

5

Figure 5.1. Plot for the function y = f ( t ) of Example 5.6.

**Example 5.7 Find the total solution of the ODE
**

dy - + 3y = 3e –2t + 4 ----dt dt dy

2 2

(5.36)

subject to the initial conditions y ( 0 ) = 1 and y' ( 0 ) = – 1 Solution: The left side of (5.36) is the same as that of Example 5.6.Therefore,

yN ( t ) = k1 e + k2 e

–t – 3t

(5.37)

(We must remember that the constants k 1 and k 2 must be evaluated from the total response). To find the forced response, we assume a solution of the form

y F = Ae

– 2t

(5.38)

We can find out whether our assumption is correct by substituting (5.38) into the given ODE of (5.36). Then,

4Ae

– 2t

– 8Ae

– 2t

+ 3Ae

– 2t

= 3e

– 2t

(5.39)

from which A = – 3 and the total solution is

5−12

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Using the Method of Undetermined Coefficients for the Forced Response
**

y ( t ) = yN + yF = k1 e + k2 e

–t – 3t

–3 e

– 2t

(5.40)

**The constants k 1 and k 2 are evaluated from the given initial conditions. For this example,
**

y ( 0 ) = 1 = k 1 e + k 2 e – 3e

0 0 0

or

k1 + k2 = 4

(5.41)

–t – 3t

Also,

dy y' ( 0 ) = – 1 = ----dt = – k 1 e – 3k 2 e

t=0

+ 6e

– 2t t=0

or

– k 1 – 3k 2 = – 7

(5.42)

**Simultaneous solution of (5.41) and (5.42) yields k 1 = 2.5 and k 2 = 1.5 . By substitution into (5.40), we obtain
**

y ( t ) = y N + y F = 2.5e + 1.5e

–t – 3t

–3 e

– 2t

(5.43)

**Check with MATLAB:
**

y=dsolve('D2y+4*Dy+3*y=3*exp(−2*t)', 'y(0)=1', 'Dy(0)=−1')

**y = (-3*exp(-2*t)*exp(t)+3/2*exp(-3*t)*exp(t)+5/2)/exp(t)
**

pretty(y)

-3 exp(-2 t) exp(t) + 3/2 exp(-3 t) exp(t) + 5/2 -----------------------------------------------exp(t) The plot is shown in Figure 5.2 was produced with the MATLAB script

y=dsolve('D2y+4*Dy+3*y=3*exp(−2*t)', 'y(0)=1', 'Dy(0)=−1'); ezplot(y,[0 8])

**Example 5.8 Find the total solution of the ODE
**

dy - + 9y = 0 + 6 ----dt dt dy

2 2

(5.44)

subject to the initial conditions y ( 0 ) = – 1 and y' ( 0 ) = 1

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

5−13

46) and (5.8 0.47) From (5.47) we obtain k 1 = – 1 and k 2 = – 2 .48) y=dsolve('D2y+6*Dy+9*y=0'.45) Next.1 0 0 1 2 t 3 4 5 5/2 exp(-t)+3/2 exp(-3 t)-3 exp(-2 t) Figure 5.2. Check with MATLAB: y ( t ) = –e – 3t – 2te – 3t (5. Solution: This is a homogeneous ODE and therefore its total solution is just the natural response found from the characteristic equation s + 6s + 9 = 0 whose roots are s 1 = s 2 = – 3 (repeated roots). ----y' ( 0 ) = 1 = dy dt = – 3k 1 e t=0 + k2 e – 3t – 3k 2 te – 3t t=0 or – 3k 1 + k 2 = 1 (5.4 0.45). and State Equations 1 0.46) – 3t Also. y ( 0 ) = –1 = k1 e + k2 ( 0 ) e 0 0 or k1 = –1 (5.5 0. Thus. Plot for the function y = f ( t ) of Example 5. we evaluate the constants k 1 and k 2 from the given initial conditions.7.2 0.9 0.Chapter 5 Differential Equations. the total response is y ( t ) = yN = k1 e – 3t 2 + k 2 te – 3t (5. For this example. Third Edition Copyright © Orchard Publications .7 0. By substitution into (5. State Variables.6 0.3 0. 'Dy(0)=1') 5−14 Numerical Analysis Using MATLAB® and Excel®. 'y(0)=−1'.

[0 3]) 0 -0. Third Edition Copyright © Orchard Publications 5−15 . we find the forced response by assuming a solution of the form y F = Ae – 2t (5.1 -0. 'y(0)=−1'.51) We can find out whether our assumption is correct by substitution of (5. 'Dy(0)=1').3 was produced with the MATLAB script y=dsolve('D2y+6*Dy+9*y=0'. Then. the roots of the characteristic equation of (5.2 -0. ezplot(y.5 1 1.+ 6y = 3e – 2t + 5 ----dt dt dy 2 2 (5. Example 5.49).50) Next.5 3 -exp(-3 t)-2 exp(-3 t) t Figure 5. Plot for the function y = f ( t ) of Example 5.49) are s 1 = – 2 and s 2 = – 3 and thus the natural response has the form yN = k1 e – 2t + k2 e – 3t (5. By inspection.3.49) Solution: No initial conditions are given.7 -0.6 -0.51) into the given ODE of (5.9 Find the total solution of the ODE dy .8.4 -0. we will express the solution in terms of the constants k 1 and k 2 .5 t 2 2.8 -0.3 -0. Numerical Analysis Using MATLAB® and Excel®.5 -0.9 -1 0 0. therefore.Using the Method of Undetermined Coefficients for the Forced Response y = -exp(-3*t)-2*exp(-3*t)*t The plot shown in Figure 5.

the total response is y ( t ) = yN + yF = k1 e – 2t + k2 e – 3t + 3te – 2t (5.52) but the sum of the three terms on the left side of (5. therefore.50).51) by t in order to eliminate the duplication of terms in the total response. The problem here is that the right side of the given ODE of (5.54) Check with MATLAB: y=dsolve('D2y+5*Dy+6*y=3*exp(−2*t)') y = -3*exp(-2*t)+3*t*exp(-2*t)+C1*exp(-3*t)+C2*exp(-2*t) We observe that the first and last terms of the displayed expression above have the same form and thus they can be combined to form a single term C3*exp(-2*t). Third Edition Copyright © Orchard Publications . Then.Chapter 5 Differential Equations.55) is the same of that of Example 5. and State Equations 4Ae – 2t – 10Ae – 2t + 6Ae – 2t = 3e – 2t (5.53) into (5.53) that is. we multiply (5.49) has the same form as one of the terms of the natural response of (5. we will express solution in terms of the constants k 1 and k 2 .55) Solution: No initial conditions are given.56) Next. the natural response is the same.10 Find the total solution of the ODE d 2y + 5 dy ----. We observe that the left side of (5. To work around this problem. Therefore. we find that A = 3 . to find the forced response and we assume a solution of the form 5−16 Numerical Analysis Using MATLAB® and Excel®. Therefore.49) and equating like terms. by substitution of (5. that is. Example 5.9.+ 6y = 4 cos 5t 2 dt dt (5.52) is zero whereas the right side can never be zero unless we let t → ∞ and this produces a meaningless result. it has the form yN = k1 e – 2t + k2 e – 3t (5. we assume that the forced response has the form y F = Ate – 2t (5. State Variables. namely the term k 1 e – 2t .

57) into the given ODE of (5. we let y F = k 3 sin 5 t + k 4 cos 5t (5. sin kx .Using the Method of Undetermined Coefficients for the Forced Response y F = A cos 5t (5. This inconsistency is a result of our failure to recognize that the derivatives of A cos 5t produce new terms of the form B sin 5t and these terms must be included in the forced response.sin 5t + -------+ -------493 493 (5.1 summarizes the forms of the forced response for a second order Numerical Analysis Using MATLAB® and Excel®. [k3 k4]=solve(19*x+25*y. x where n is a positive integer.55) we obtain – 25 k 3 sin 5t – 25k 4 cos 5 t + 25k 3 cos 5 t – 25k 4 sin 5 t + 6k 3 sin 5t + 6k 4 cos 5 t = 4 cos 5 t Collecting like terms and equating sine and cosine terms.58) and by substitution into (5. the total solution is y ( t ) = yN + yF ( t ) = k1 e – 2t + k2 e – 3t – 38 50 . and linear combinations of these. y=simple(y) y = -38/493*cos(5*t)+50/493*sin(5*t)+C1*exp(-3*t)+C2*exp(-2*t) In most engineering problems the right side of the non−homogeneous ODE consists of elementary functions such as k (constant). – 25A cos 5t – 25A sin 5 t + 6A cos 5t = – 19A cos 5t – 25A sin 5 t = 4 cos 5t but this relation is invalid since by equating cosine and sine terms. Accordingly. Then. cos kx .56).60) Check with MATLAB: y=dsolve('D2y+5*Dy+6*y=4*cos(5*t)').57) We can find out whether our assumption is correct by substitution of the assumed solution of (5. we obtain the following set of equations 19k 3 + 25k 4 = 0 25k 3 – 19 k 4 = 4 (5. 25*x−19*y−4) k3 = 50/493 k4 = -38/493 Therefore. we find that A = – 4 ⁄ 19 and also A = 0 . e .59) We use MATLAB to solve (5. Third Edition Copyright © Orchard Publications n kx 5−17 .59) format rat.cos 5t . Table 5.

61) Solution: No initial conditions are given. that is. This term with n = 1 . and thus the natural response has the form yN = k1 e –2 t + k 2 te –2 t (5.+ 4y = te – 2t – e – 2t + 4 ----dt dt (5. State Variables.11 Find the total solution of the ODE d 2y 2 dy . TABLE 5. the most general form of the forced response y F ( t ) is the linear combination of these terms. if a term in y F ( t ) is a duplicate of a term in the natural response y N ( t ) . Third Edition Copyright © Orchard Publications .+ cy = f ( t ) Forced Response of the ODE a ------2 dt dt f (t) k (constant) k t ( n = positive integer) ke rt n 2 Form of Forced Response y F ( t ) K (constant) K0 t + K1 t Ke rt n n–1 + … + Kn – 1 t + Kn ( r =real or complex) k cos α t or k sin α t ( α =constant) K 1 cos α t + K 2 sin α t k t e cos α t or k t e sin α t n rt n rt ( K0 t + K1 t n n n–1 + … + K n – 1 t + K n ) e cos α t + … + K n – 1 t + K n ) e sin α t rt rt + ( K0 t + K1 t n–1 We must remember that if f ( t ) is the sum of several terms.+ b ----. r = – 2 . we must multiply y F ( t ) by the lowest power of t that will eliminate the duplication. The roots of the characteristic equation are equal. we refer to the table of the previous page and from the last row we choose the term k t e cos α t . n rt 5−18 Numerical Analysis Using MATLAB® and Excel®. s 1 = s 2 = – 2 . and State Equations ODE with constant coefficients. Also. therefore we will express solution in terms of the constants k 1 and k 2 . and α = 0 .Chapter 5 Differential Equations. Example 5.1 Form of the forced response for 2nd order differential equations d y dy .62) To find the forced response (particular solution).

61) and equating with the right side. we need to evaluate the constants k 3 and k 4 . f1=diff(f0).64) Now.61).62) by t to obtain a suitable form for the forced response which now is yF = ( k3 t + k4 t ) e 3 2 –2 t – 2t – 2t (5.64) and combining the forced response with the natural response. we equate f above with the right side of the given ODE. Therefore the forced response will have the form yF = ( k3 t + k4 ) e –2 t (5. f1=simple(f1) % Define symbolic variables % Forced response (5. This is done by substituting (5. f2=simple(f2) % Compute and simplify second derivative f2 = 2*exp(-2*t)*(3*k3*t+k4-6*k3*t^2-4*k4*t+2*k3*t^3+2*k4*t^2) f=f2+4*f1+4*f0.63) 2 But the terms e and te are also present in (5. f=simple(f) % Form and simplify the left side of the given ODE f = 2*(3*k3*t+k4)*exp(-2*t) Finally. syms t k3 k4 f0=(k3*t^3+k4*t^2)*exp(−2*t).t e – --t e + -6 2 (5.64) % Compute and simplify first derivative f1 = -t*exp(-2*t)*(-3*k3*t-2*k4+2*k3*t^2+2*k4*t) f2=diff(f0. Third Edition Copyright © Orchard Publications 5−19 .65) and we find k 3 = 1 ⁄ 6 and k 4 = – 1 ⁄ 2 . We use MATLAB do the computations as shown below.Using the Method of Undetermined Coefficients for the Forced Response reduces to kte –2 t . therefore.2).64) into the given ODE of (5. that is 2 ( 3k 3 t + k 4 ) e – 2t = te – 2t –e – 2t (5. we obtain the total solution y ( t ) = k1 e –2 t + k 2 te –2 t 1 3 –2 t 1 2 –2 t . we multiply (5.66) We verify this solution with MATLAB as follows: z=dsolve('D2y+4*Dy+4*y=t*exp(−2*t)−exp(−2*t)') z = 1/6*exp(-2*t)*t^3-1/2*exp(-2*t)*t^2 +C1*exp(-2*t)+C2*t*exp(-2*t) Numerical Analysis Using MATLAB® and Excel®. By substitution of these values into (5.

12 Find the total solution of 2y dy d .+ 3y = 12 ------.6 Using the Method of Variation of Parameters for the Forced Response In certain non−homogeneous ODEs. method of variation of parameters Solution: With either method. State Variables. The characteristic equation yields 5−20 Numerical Analysis Using MATLAB® and Excel®.69) will yield the values of du1 ⁄ dt and du 2 ⁄ dt . Third Edition Copyright © Orchard Publications .68) (5. which when substituted into (5.Chapter 5 Differential Equations.= f (t) dt dt dt dt (5. For these ODEs we must use the method of variation of parameters.+ 4 ----2 dt dt (5. integration of these will produce u 1 and u 2 . then.⋅ ------. the right side f ( t ) cannot be determined by the method of undetermined coefficients.71) in terms of the constants k 1 and k 2 by the a.+ ------. and State Equations 5.69) (5.⋅ ------.70) Simultaneous solution of (5. Our discussion will be restricted to second order ODEs with constant coefficients.+ α ( t ) ----2 dt dt (5. we must first find the natural response. method of undetermined coefficients b. Example 5. The method of variation of parameters replaces the constants k 1 and k 2 by two variables u 1 and u 2 that satisfy the following three relations: y = u1 y1 + u2 y2 du1 du2 -----.67) provided that the general form of the natural response is known.67) will yield the total solution.y1 + -------y = 0 dt dt 2 du1 dy1 du 2 dy 2 ------.+ β(t)y = f ( t) ------.68) and (5. This method will work with all linear equations including those with variable coefficients such as dy d2 y .

by substitution into (5.⋅ ------.72) a.⋅ ------.( – 3e – 3t ) = 12 dt dt (5.74) Then by (5. we find du1 ⁄ dt and du 2 ⁄ dt by Cramer’s rule as follows: 0 e – 3t – 3t – 3t – 3t du 1 12 – 3e .73) b.70).( – e – t ) + ------.= f( t) dt dt dt dt or du 1 du 2 ------. the natural response is yN = k1 e + k2 e –t –3 t (5. Using the method of undetermined coefficients we let y F = k 3 (a constant).69). the total solution is y = u1 y1 + u2 y2 or y = u1 e + u2 e –t – 3t (5.e + -------e = 0 dt dt (5. Third Edition Copyright © Orchard Publications 5−21 .68). du1 dy1 du 2 dy 2 ------.+ ------. from (5.76) and from (5.78) –e –t – 3e – 3t and Numerical Analysis Using MATLAB® and Excel®.= -----------------------------------------.Using the Method of Variation of Parameters for the Forced Response the roots s 1 = – 1 and s 2 = – 3 .= 6e t ------. du 2 du 1 ------.= ------------------------------– 12e – 12e .72) and we let the solutions y 1 and y 2 be represented as y1 = e –t and y 2 = e – 3t (5. Therefore. With the method of variation of parameters we begin with the natural response found above as (5.75) Also.77) Next. Then.71) we obtain k 3 = 4 and thus the total solution is y ( t ) = yN + yF = k1 e + k2 e –t –3 t +4 (5.y 1 + -------y = 0 dt dt 2 or du 1 – t du 2 – 3t ------.= ----------------–t – 3t – 4t – 4t – 4t dt e – 3e + e – 2e e (5.

85) We let y 1 = cos 2t and y 2 = sin 2t Then. The characteristic equation is s + 4 = 0 from which s = ± j2 and thus the natural response is yN = k1 e j2t 2 + k2 e – j 2t (5.79) and substitution into (5.13 Find the total solution of d y ------. Check with MATLAB: y=dsolve('D2y+4*Dy+3*y=12') y = (4*exp(t)+C1*exp(-3*t)*exp(t)+C2)/exp(t) Example 5.Chapter 5 Differential Equations.82) in terms of the constants k 1 and k 2 by any method.84) (5. by (5. State Variables.81) = ( 6e + k 1 ) e + ( – 2 e + k 2 ) e – 3t = 6 + k1 e – 2 + k2 e = k1 e + k2 e –t – 3t +4 We observe that the last expression in (5.= -------------------. and State Equations e –t 0 –t –t du 2 12e .75) yields u 1 = 6 e dt = 6e + k 1 y = u1 e + u2 e –t –t – 3t t ∫ t t u 2 = – 6 e dt = – 2 e + k 2 –t 3t – 3t ∫ 3t 3t (5.= -------------------------------– 4t – 4t dt – 2e – 2e (5.73) of part (a). integration of (5.79) Now.80) (5. therefore.78) and (5.81) is the same as (5.+ 4y = tan 2t 2 dt 2 (5. we will use the method of variation of parameters. Solution: This ODE cannot be solved by the method of undetermined coefficients.68) the solution is y = u1 y 1 + u 2 y 2 = u1 cos 2t + u 2 sin 2t 5−22 Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications .= – 6 e 3t –e 12.83) (5.

= f ( t ) = ------. Third Edition Copyright © Orchard Publications 5−23 .= -------------------------------------------------2 2 dt (5.dt = -----------– -u 1 = – -4 2 cos 2t 4 ∫ 2 (5.cos 2t ln ( sec 2t + tan 2t ) + k 1 cos 2t + k 2 sin 2t = – -4 (5.+ k 2 sin 2t 4 4 4 1 .88) and (5.y1 + -------y = 0 dt dt 2 or du 2 du1 ------.= -----------------------------------------------------.sin 2t = 0 dt dt (5.86) and from (5.cos 2t + ------.70). du1 du 2 ------.89) and substitution into (5.88) and cos 2t 0 du 2 sin 2t – 2 sin 2t tan 2t = -----------------.= ----------------2 2 cos 2t sin 2t 2 cos 2t dt 2 cos 2t + 2 sin 2t – 2 sin 2t 2 cos 2t 0 sin 2t 2 (5. we find du1 ⁄ dt and du 2 ⁄ dt by Cramer’s rule as follows: sin 2t – -------------2 du1 tan 2t 2 cos 2t cos 2t – sin 2t ------.87) Next.91) 1 cos 2t . integration of (5.⋅ ------.69). from (5.90) (5.– -.( 2 cos 2t ) = tan 2t .( – 2 sin 2t ) + ------dt dt dt dt dt dt (5.ln ( sec 2t + tan 2t ) + k 1 .89) Now.92) Check with MATLAB: y=dsolve('D2y+4*y=tan(2*t)') y = -1/4*cos(2*t)*log((1+sin(2*t))/cos(2*t))+C1*cos(2*t)+C2*sin(2*t) Numerical Analysis Using MATLAB® and Excel®.= ------------------------------------------.85) yields sin 2t 1 -------------sin 2t 1 .sin 2t dt = – ------------u 2 = -+ k2 2 4 ∫ sin 2t cos 2t 1 sin 2t cos 2t .Using the Method of Variation of Parameters for the Forced Response Also.+ ------.⋅ ------.cos 2t ln ( sec 2t + tan 2t ) + k 1 cos 2t – -------------------------y = u1 y 1 + u 2 y 2 = -------------------------. du 2 du1 dy1 du 2 dy 2 du1 ------.

Third Edition Copyright © Orchard Publications . An nth−order differential equation can be resolved to n first−order simultaneous differential equations with a set of auxiliary variables called state variables.= x x 2 = ---dt dt Then.97) (5. our discussion will be limited to linear. or simply state equations.jωe .14 A system is described by the integro−differential equation 1 di .= ------.---.95) or 2 1 R di 1 d t . these will be discussed in Chapter 9.= – --2 L L dt LC dt jωt (5.– ------. However.+ --Ri + L ---dt C t ∫ – ∞ i dt = e jωt (5. and State Equations 5.7 Expressing Differential Equations in State Equation Form A first order differential equation with constant coefficients has the form dy .di i = ------2 L L dt LC dt jωt (5. Example 5.Chapter 5 Differential Equations.+ a0 y ( t ) = x ( t ) a 1 ----dt (5.99) and dx 1 di ·1 . · 2 = d 2 i ⁄ dt 2 x 5−24 Numerical Analysis Using MATLAB® and Excel®.98) (5. The state variable method is best illustrated through several examples presented in this chapter. State equations can also be solved with numerical methods such as Taylor series and Runge− Kutta methods.93) In a second order differential equation the highest order is a second derivative. State Variables.i + ------. time−invariant systems. The state variable method offers the advantage that it can also be used with non−linear and time−varying systems. The resulting first−order differential equations are called state space equations.jωe ---.96) Next.94) Differentiating both sides and dividing by L we obtain 2 1 1t+R d . we define two state variables x 1 and x 2 such that x1 = i (5.+ --------.

jωe jωt x L LC L (5. and u = any input -.102) 0 where · · = x1 . we obtain the state equations · 1 = x2 x 1 1 ·2 = – R -. b= 1 . Therefore.101) We usually write (5.x 2 – -----.105) can be realized with the block diagram of Figure 5. the state representation of a system can be described by the pair of the of the state space equations · = Ax + bu x y = Cx + du (5.99). · x + A d u b + Σ ∫ dt x C + Σ + y Figure 5. Third Edition Copyright © Orchard Publications 5−25 .96) through (5.x2 -.x 1 + -. Thus.jωe jωt L L (5.101) in a compact form as · = Ax + bu x (5. –R -L x = x1 x2 .4.100) as 0 ·1 x = 1 ·2 – -----x LC 1 x 0 1 + 1 u R – -.105) Numerical Analysis Using MATLAB® and Excel®. and d is a column vector.Expressing Differential Equations in State Equation Form · denotes the derivative of the state variable x . Block diagram for the realization of the state equations of (5.104) where C is another matrix. x ·2 x A = 0 1 – -----LC 1 . we write the state equations of (5.103) The output y ( t ) is expressed by the state equation y = Cx + du (5.100) It is convenient and customary to express the state equations in matrix form. where x k k From (5.105) The state space equations of (5.jωe jωt L (5.1.

Chapter 5 Differential Equations.109) In compact form.+ a 2 ------.109) is written as · = Ax + bu x (5. State Variables.108) and in matrix form ·1 x ·2 x ·3 x ·4 x 0 0 = 0 –a0 1 0 0 –a1 0 1 0 –a2 0 0 1 –a3 x1 0 x2 + 0 u(t) x3 0 1 x4 (5.107) We observe that · 1 = x2 x · 2 = x3 x · 3 = x4 x d y · 4 = –a0 x1 –a1 x2 – a2 x3 –a3 x4 + u ( t ) --------. Solution: The differential equation of (5.+ a0 y ( t ) = u ( t ) .+ a 1 ------------3 -------3 2 4 dt dt dt dt 3 2 (5.106) is of fourth−order. and we relate them to the terms of the given differential equation as x1 = y ( t ) dy x 2 = ----dt d y x 3 = -------2 dt 2 d y x 4 = -------3 dt 3 (5. x 2. x 3 .106) where y ( t ) is the output and u ( t ) is any input. therefore. We denote the state variables as x 1.110) where 5−26 Numerical Analysis Using MATLAB® and Excel®. and x 4 . we must define four state variables that will be used with the resulting four first−order state equations. and State Equations We will learn how to solve the matrix equations of (5.= x 4 dt 4 (5. (5. Example 5. Express (5.106) as a set of state equations.105) in the subsequent sections. Third Edition Copyright © Orchard Publications .15 A fourth−order system is described by the differential equation 4 y d y dy d y+a d .

112) We will now prove that the solution of the first state equation in (5. Next. if non−zero. we must show that (5. and k 2 are scalar constants.114) is zero since the upper and lower limits of integration are the same. and the initial condition.113) with respect to t and we obtain d.111) where α .113). we differentiate (5.111) is x(t) = e α ( t – t0 ) x0 + e αt ∫t e 0 t – ατ β u ( τ ) dτ (5.113) Proof: First. we must prove that (5.Solution of Single State Equations ·1 x ·2 x ·3 x ·4 x 0 0 A= 0 –a0 1 0 0 –a1 0 1 0 –a2 0 0 . 0 1 · = x .111). Therefore. (5. 1 –a3 x1 x= x2 x3 x4 . β .8 Solution of Single State Equations Let us consider the state equations · = αx + βu x y = k1 x + k2 u (5. Third Edition Copyright © Orchard Publications 5−27 .113) satisfies also the first equation in (5. To prove this.114) reduces to x 0 since e α ( t0 – t0 ) x0 = e x0 = x0 0 (5.⎧ α t · ( t ) = ---e x (e x 0 ) + ---dt dt ⎨ ⎩ ∫t e 0 t – ατ β u ( τ ) dτ ⎬ ⎫ ⎭ Numerical Analysis Using MATLAB® and Excel®.113) satisfies the initial condition of (5. k 1 . and u = u ( t ) 5.115) The second term of (5. Then. x ( t0 ) = e α ( t0 – t0 ) x0 + e αt ∫t t0 0 e –α τ β u ( τ ) dτ (5.α ( t – t0 ) d.112). This is done by substitution of t = t 0 in (5. is denoted as x0 = x ( t0 ) (5.114) reduces to x ( t 0 ) = x 0 and thus the initial condition is satisfied. 0 b= 0.114) The first term in the right side of (5.

111). Therefore. · = αx + βu x and this is the same as the first equation of (5. Third Edition Copyright © Orchard Publications . and State Equations · ( t ) = α e α ( t – t0 ) x + α e α t x 0 = α e α ( t – t0 ) ∫t ∫t t 0 t 0 e e – ατ β u ( τ ) dτ + e [ e αt – ατ βu(τ)] τ = t βu(t) x0 + e αt t 0 – ατ β u ( τ ) dτ + e e α t –α t · ( t ) = α eα ( t – t0 ) x0 + x ∫t e α(t – τ) β u ( τ ) dτ + β u ( t ) (5. if α and β are scalar constants.9 The State Transition Matrix Let us again consider the state equations pair · = Ax + bu x y = Cx + du (5. and b and d are column vectors with two or more rows.122) x ( t0 ) = x0 is obtained from the relation At 5−28 Numerical Analysis Using MATLAB® and Excel®.111) is an algebraic equation whose coefficients are scalar constants. State Variables. the solution of · = αx + βu x (5.113).Chapter 5 Differential Equations. In this section we will introduce the state transition matrix e . and we will prove that the solution of the matrix differential equation · = Ax + bu (5.116) are the same as the right side of the assumed solution of (5.117) (5.118) with initial condition x0 = x ( t0 ) is obtained from the relation x( t) = e α ( t – t0 ) x0 + e αt ∫t e 0 t –α τ β u ( τ ) dτ (5.120) where for two or more simultaneous differential equations A and C are 2 × 2 or higher order matrices.116) We observe that the bracketed terms of (5. The second equation of (5.121) x with initial conditions (5.119) 5. In summary.

if it is related to the matrix A as the matrix power series ϕ(t) ≡ e At 1 22 1 33 1 nn .34). From (5. Third Edition Copyright © Orchard Publications 5−29 . is said to be the state transition matrix of (5.124) where I is the n × n identity matrix.120) is also satisfied.127).124).123) is the solution of the first equation of (5.127) To prove that (5.124) we obtain d At ---.e = 0 + A ⋅ 1 + A2 t + … = A + A2 t + … ϕ' ( t ) = ---dt (5. d At ---.The State Transition Matrix x(t) = e A ( t – t0 ) x0 + e At ∫t e 0 t –A τ bu ( τ ) dτ (5.e = Ae At dt Numerical Analysis Using MATLAB® and Excel®.A t + … + ---. To prove that the first equation of (5.123) Proof: Let A be any n × n matrix whose elements are constants.126) and by comparison with (5. we differentiate the assumed solution x( t) = e A ( t – t0 ) x0 + e At ∫t e 0 t –A τ bu ( τ ) dτ with respect to t and we use (5. another n × n matrix denoted as ϕ ( t ) . The integral is zero since the upper and lower limits of integration are the same.e = Ae At dt (5. we find that ϕ(0) = e A0 = I + A0 + … = I (5. The initial condition is satisfied from the relation x ( t0 ) = e A ( t0 – t0 ) x0 + e At 0 ∫t t0 0 e –A τ bu ( τ ) dτ = e A0 x 0 + 0 = Ix 0 = x 0 (5. Then.125) Differentiation of (5.120).124) with respect to t yields d At . we must prove that it satisfies both the initial condition and the matrix differential equation.128) where we have used (5.A t + ----A t = I + At + ---3! 2! n! (5. that is.125) for the initial condition.

129) reduces to · ( t ) = Ax + bu x In summary.130) with initial condition x0 = x ( t0 ) (5. At 5. the eigenvalues λ i . the expression (5. (5. and I be the n × n identity matrix. ….Chapter 5 Differential Equations. This theorem states that a matrix can be expressed as an ( n – 1 ) th degree polynomial in terms of the matrix A as e At At = a0 I + a1 A + a2 A + … + an – 1 A 2 n–1 (5. 2. Third Edition Copyright © Orchard Publications . Evaluation of the state transition matrix e is based on the Cayley−Hamilton theorem. n ≥ 2 . the solution of second or higher order systems using the state variable method.131) is x(t) = e A ( t – t0 ) x0 + e At ∫t e 0 t –A τ bu ( τ ) dτ (5.132) Therefore. entails the computation of the state transition matrix e . The roots of the polynomial of (5. By definition.129) as x ( t ) . and b is a column vector with n elements. the solution of · ( t ) = Ax + bu x (5. or complex numbers. n of A are the roots of the nth order polynomial det [ A – λ I ] = 0 At Let A be an n × n matrix.132). · ( t ) = Ae A ( t – t0 ) x 0 + Ae At x t ∫t e 0 –A τ bu ( τ ) dτ + e e At – A t bu ( t ) or · ( t ) = A e A ( t – t0 ) x 0 + e At x ∫t e 0 t –A τ bu ( τ ) dτ + e e At – A t bu ( t ) (5.133) We recall that expansion of a determinant produces a polynomial.134) 5−30 Numerical Analysis Using MATLAB® and Excel®.129) We recognize the bracketed terms in (5. and State Equations Then. if A is an n × n matrix whose elements are constants. State Variables. and the last term as bu ( t ) . Thus.10 Computation of the State Transition Matrix e i = 1.133) can be real (unequal or equal). and integration of (5.

that is. det [ A – λ I ] = 0 At given that A = – 2 1 0 –1 ⎧ ⎫ 1 det [ A – λ I ] = det ⎨ – 2 1 – λ 1 0 ⎬ = det – 2 – λ = 0 0 1 ⎭ 0 –1–λ ⎩ 0 –1 or Therefore. we must find the coefficients a i of (5.16 Compute the state transition matrix e Solution: We must first find the eigenvalues λ of the given matrix A . = (– 2 – λ)(– 1 – λ) = 0 (λ + 1)(λ + 2) = 0 λ 1 = – 1 and λ 2 = – 2 (5. The proof can be found in Linear Algebra and Matrix Theory textbooks.Computation of the State Transition Matrix where the coefficients a i are functions of the eigenvalues λ . Numerical Analysis Using MATLAB® and Excel®. we only need to consider the first two terms of that relation.136) Next. We accept (5.134). These are found from the expansion of For this example.135) = e λn t Example 5. if all eigenvalues of a given matrix A are distinct. Since A is a 2 × 2 matrix. that is. Since the coefficients a i are functions of the eigenvalues λ . we must consider the following cases: Case I: Distinct Eigenvalues (Real or Complex) If λ 1 ≠ λ 2 ≠ λ 3 ≠ … ≠ λ n . Third Edition Copyright © Orchard Publications 5−31 .134) without proving it. the coefficients a i are found from the simultaneous solution of the following system of equations: a0 + a1 λ1 + a2 λ1 + … + an – 1 λ1 a0 + a1 λ2 + a2 λ2 + … + an – 1 λ2 … a0 + a1 λn + a2 λn + … + an – 1 λn 2 n–1 2 2 n–1 n–1 = e = e λ1 t λ2 t (5.

141) If A matrix is a 3 × 3 matrix. We can write [ A – λ I ] at once by subtracting λ from each of the main diagonal elements of A . a0 + a1 λ1 = e a0 + a1 λ2 = e λ1 t λ2 t or a0 + a1 ( –1 ) = e a0 + a1 ( –2 ) = e –t – 2t (5. and so on. and so on.138) yields a 0 = 2e – e a1 = e – e –t –t – 2t – 2t (5. if it is a 3 × 3 matrix. we use only the first 2 terms of the right side of the state transition matrix e At = a0 I + a1 A + a2 A + … + an – 1 A 2 n–1 (5.138) Simultaneous solution of (5.135). State Variables. Third Edition Copyright © Orchard Publications . otherwise we refer to Case II. If the eigenvalues are distinct. If the dimension of A is a 2 × 2 matrix. e At = ( 2e – e –t – 2t ) 1 0 0 1 –t + (e – e –t – 2t ) –2 1 0 –1 or e At = e – 2t e –e e –t – 2t (5. we compute the state transition matrix e procedure: At for a given matrix A using the following 1. 5−32 Numerical Analysis Using MATLAB® and Excel®. it will yield three eigenvalues. 2.137) The coefficients a 0 and a 1 are found from (5.139) and by substitution into (5. If the dimension of A is a 2 × 2 matrix. it will yield two eigenvalues. we use the first 3 terms.137). For this example. We find the eigenvalues λ from det [ A – λ I ] = 0 .140) 0 In summary. we perform steps 2 through 4 below. and State Equations e At = a0 I + a1 A (5.Chapter 5 Differential Equations.

144).00 lambda3 = 1.141).. r=roots(p).. r(2)).142) Solution: 1. 4. r(3)) lambda1 = 3.2f \t'. We obtain [ A – λ I ] at once. Then.144) We will use MATLAB roots(p) function to obtain the roots of (5.00 lambda2 = 2. by subtracting λ from each of the main diagonal elements of A . We first compute the eigenvalues from det [ A – λ I ] = 0 .17 Compute the state transition matrix e At given that 7 –5 4 –1 8 –3 5 A = 0 2 (5.00 and thus the eigenvalues are Numerical Analysis Using MATLAB® and Excel®. We obtain the a i coefficients from a0 + a1 λ1 + a2 λ1 + … + an – 1 λ1 a0 + a1 λ2 + a2 λ2 + … + an – 1 λ2 … a0 + a1 λn + a2 λn + … + an – 1 λn 2 n–1 2 2 n–1 n–1 λ1 t λ2 t = e = e = e λn t We use as many equations as the number of the eigenvalues. We substitute the a i coefficients into the state transition matrix of (5.143) and expansion of this determinant yields the polynomial λ – 6 λ + 11 λ – 6 = 0 3 2 (5. and we simplify. fprintf(' \n'). fprintf('lambda2 = %5.2f'. fprintf('lambda1 = %5. det [ A – λ I ] = det 5–λ 0 2 7 –5 = 0 4–λ –1 8 –3–λ (5. Example 5. Third Edition Copyright © Orchard Publications 5−33 .. and we solve for the coefficients ai . r(1)).Computation of the State Transition Matrix 3. fprintf('lambda3 = %5.2f \t'. p=[1 −6 11 −6].

. fprintf(' \n'). and to perform the matrix multiplications. B=sym('[1 1 1.147).148) 4.Chapter 5 Differential Equations.. 1 2 4. 2 8 −3]. The script is shown below.147) We will use the following MATLAB script for the solution of (5. a 0 = 3e – 3e + e t 2t 3t 5 . and State Equations λ1 = 1 λ2 = 2 λ3 = 3 (5.. We obtain the coefficients a 0. disp(a(3)) a0 = 3*exp(t)-3*exp(2*t)+exp(3*t) a1 = -5/2*exp(t)+4*exp(2*t)-3/2*exp(3*t) a2 = 1/2*exp(t)-exp(2*t)+1/2*exp(3*t) Thus. b=sym('[exp(t). 0 4 −1.. disp('a1 = '). State Variables. a1 = −5/2*exp(t)−3/2*exp(3*t)+4*exp(2*t). a=B\b.145) 2...e t + 4e 2t – 3 -. disp(a(1)). 1 3 9]'). syms t.134). disp(a(2)). a2 = 1/2*exp(t)+1/2*exp(3*t)−exp(2*t).e 3t a 1 = – -2 2 1 t 2t 1 3t -e . Third Edition Copyright © Orchard Publications . We also use MATLAB to perform the substitution into the state transition matrix.e – e + -a 2 = -2 2 (5. exp(2*t). A = [5 7 −5. a0 = 3*exp(t)+exp(3*t)−3*exp(2*t). we need to use the first 3 terms of (5.. disp('a0 = ').146) 3. Since A is a 3 × 3 matrix. exp(3*t)]').. eAt=a0*eye(3)+a1*A+a2*A^2 5−34 Numerical Analysis Using MATLAB® and Excel®. that is. e At = a0 I + a1 A + a2 A 2 (5. a 1.. disp('a2 = '). and a 2 from a0 + a1 λ1 + a2 λ1 = e a0 + a1 λ2 + a2 λ2 = e a0 + a1 λ3 + a2 λ3 = e 2 2 2 λ1 t λ2 t λ3 t or a0 + a1 + a2 = e a 0 + 2a 1 + 4a 2 = e a 0 + 3a 1 + 9a 2 = e t 2t 3t (5.

we will assume that the polynomial of det [ A – λ I ] = 0 (5. Example 5.18 Compute the state transition matrix e At given that A = –1 0 2 –1 Solution: 1.151) are found from the simultaneous solution of the system of equations of (5. λn (5. λm + 1 .Computation of the State Transition Matrix eAt = [-2*exp(t)+2*exp(2*t)+exp(3*t).152) below.150) The coefficients a i of the state transition matrix e At = a0 I + a1 A + a2 A + … + an – 1 A 2 n–1 (5. In other words. Third Edition Copyright © Orchard Publications 5−35 . 2*exp(t)-3*exp(2*t)+exp(3*t)] [-3*exp(t)+4*exp(2*t)-exp(3*t). 6*exp(t)-6*exp(2*t)+exp(3*t)] Thus.-9*exp(t)+10*exp(2*t)-exp(3*t). We first find the eigenvalues λ of the matrix A and these are found from the polynomial of det [ A – λ I ] = 0 . Numerical Analysis Using MATLAB® and Excel®. the roots are λ1 = λ2 = λ3 … = λm .-3*exp(t)+5*exp(2*t)-exp(3*t).-6*exp(t)+5*exp(2*t)+exp(3*t). e At – 2e + 2e + e = – e + 2e – e t 2t t 2t t 2t 3t – 6 e + 5e + e – 3e + 5e – e t 2t t 2t 3t t 2t 3t 4e – 3e – e 2e – 3e + e 6e – 6e + e t 2t t 2t t 2t 3t 3t 3t 3t 3t – 3e + 4e – e – 9e + 10e – e 3t Case II: Multiple Eigenvalues In this case.149) has n roots. For this example. 4*exp(t)-3*exp(2*t)-exp(3*t)] [-exp(t)+2*exp(2*t)-exp(3*t). and m of these roots are equal.

Chapter 5 Differential Equations.153) 3.λ1 t d 2 n–1 ------e ( a 0 + a 1 λ 1 + a 2 λ 1 + … + a n – 1 λ 1 ) = ------2 2 d λ1 d λ1 … d d n–1 -------------.( a0 + a1 λ1 + a2 λ2 -e ) = -------------1 + … + an – 1λ1 m–1 m–1 d λ1 d λ1 a0 + a1 λm + 1 + a2 λm + 1 + … + an – 1 λm + 1 = e a 0 + a 1 λn + a 2 λ n + … + a n – 1 λ n 2 n–1 2 n–1 m–1 m–1 2 2 (5.λ1 t -------e ( a 0 + a 1 λ 1 ) = -------d λ1 d λ1 or a0 + a1 λ1 = e λ1 t λ1 t a 1 = te and by substitution with λ 1 = λ 2 = – 1 . and State Equations a0 + a1 λ1 + a2 λ1 + … + an – 1 λ1 2 n–1 = e λ1 t d d λ1 t n–1 -------. 2. For this example.( a0 + a1 λ1 + a2 λ2 -e ) = ------1 + … + an – 1λ1 d λ1 d λ1 d . we obtain 5−36 Numerical Analysis Using MATLAB® and Excel®. we only need the first two terms of the state transition matrix.152) λ1 t λ m + 1t … = e λn t 0 = 0 det [ A – λ I ] = det – 1 – λ 2 –1–λ = (– 1 – λ)(– 1 – λ) = 0 = (λ + 1) = 0 λ1 = λ2 = –1 2 and thus. We find a 0 and a 1 from (5. State Variables. a0 + a1 λ1 = e λ1 t dd . that is.152). Since A is a 2 × 2 matrix. Third Edition Copyright © Orchard Publications . e At = a0 I + a1 A (5.

and we will briefly discuss eigenvectors on the next section. We will first use MATLAB to verify the values of the eigenvalues found in Examples 5.0000 2. lambda=eig(B) lambda = 1.17: B = [5 7 −5. 2 8 −3].0000 3.18.154) into (5.16: A= [−2 1.Computation of the State Transition Matrix a0 – a1 = e –t –t a 1 = te Simultaneous solution of the last two equations yields a 0 = e + te a 1 = te –t –t –t (5.0000 For Example 5.16 through 5. Third Edition Copyright © Orchard Publications 5−37 .155) We can use the MATLAB eig(x) function to find the eigenvalues of an n × n matrix. lambda=eig(A) lambda = -2 -1 For Example 5. To find out how it is used. 0 4 −1. lambda=eig(C) lambda = -1 -1 Numerical Analysis Using MATLAB® and Excel®. we obtain e At –t –t = ( e + te ) 1 0 0 0 + te – t – 1 0 = e At = e – t – t 2 –1 1 2te e –t (5. 2 −1].18: C = [−1 0. 0 −1]. By substitution of (5.154) 4. we invoke the help eig command.153). For Example 5.

158) ( a 22 – λ ) x 2 … … … … ( a nn – λ ) x n (5.156) where A is an n × n matrix. We can express this relation in matrix form as a 11 a 12 … a 1n x 1 a 21 a 22 … a 2n x 2 … … … … … a n1 a n2 … a nn x n = λ x1 x2 … xn (5.160) Expansion of the determinant of (5. the roots λ of the characteristic equation are the eigenvalues of the matrix A .11 Eigenvectors Consider the relation AX = λ X (5. there is a non−trivial solution of the column vector X .160) results in a polynomial equation of degree n in λ . i.158) to be a non−zero vector and the product ( A – λ I ) X to be zero. X is a column vector. and corresponding to each eigenvalue λ . Third Edition Copyright © Orchard Publications .159) will have non−trivial solutions if and only if its determinant is zero*. This is because we want the vector X in (5.160) in a compact form as det ( A – λ I ) = 0 (5. We can express (5.159) The equations of (5.e. and it is called the characteristic equation. and State Equations 5. *. State Variables.157) We write (5. if ( a 11 – λ ) det a 21 … an 1 a 12 … an2 … a1n a2n = 0 ( a 22 – λ ) … … … … ( a nn – λ ) (5.Chapter 5 Differential Equations.. that is.157) as or ( a 11 – λ ) x 1 a 21 x 1 … an 1 x1 ( A –λ I ) X = 0 a 12 x 2 … an2 x2 … a1 n xn a2 n xn = 0 (5.161) As we know. 5−38 Numerical Analysis Using MATLAB® and Excel®. and λ is a scalar number.

19 Given the matrix 5 A = 0 2 7 –5 4 –1 8 –3 a. Third Edition Copyright © Orchard Publications 5−39 . Example 5. so that the sum of the squares of their components is equal to unity. that is.Eigenvectors eigenvalue. Find eigenvectors corresponding to each eigenvalue of A c. The example which follows. We begin with and we let Then. illustrates the relationships between a matrix A . Numerical Analysis Using MATLAB® and Excel®. Find the eigenvalues of A b. This is done by dividing each component of the eigenvector by the square root of the sum of the squares of their components. and I is the identity matrix. its eigenvalues. there is a different eigenvector for each Two vectors X and Y are said to be orthogonal if their inner (dot) product is zero. they are normalized to unit length. A set of eigenvectors constitutes an orthonormal basis if the set is normalized (expressed as unit eigenvectors) and these vector are mutually orthogonal. T T X ≠ 0 . where we found the eigenvalues to be λ1 = 1 λ2 = 2 AX = λ X x1 X = x2 x3 λ3 = 3 b. This is the same matrix as in Example 5. Solution: a. Obviously. An orthonormal basis can be formed with the Gram− Schmidt Orthogonalization Procedure. In many engineering applications the unit eigenvectors are chosen such that X ⋅ X = I where X is the transpose of the eigenvector X . it is discussed in Chapter 14. Eigenvectors are generally expressed as unit eigenvectors. and eigenvectors. This vector X is called eigenvector. Form a set of unit eigenvectors using the eigenvectors of part (b).17.

163) Equating corresponding rows and rearranging.162) or 5x 1 0 2x 1 (5.164) reduces to 4x 1 + 7x 2 – 5x 3 = 0 3x 2 – x 3 = 0 2x 1 + 8x 2 – 4x 3 = 0 (5.Chapter 5 Differential Equations.167) since any eigenvector is a scalar multiple of the last vector in (5. Then. and solve x 1 and x 3 in terms of x 2 . we obtain x 1 = x 2 .164) For λ = 1 . or MATLAB. say x 2 . we obtain x 1 = 2x 2 . Similarly. for λ = 2 . (5.166) Since the unknowns x 1. and x 3 = 2x 2 . we can assume that one of these. we obtain ( 5 – λ ) x1 0 2x 1 7x 2 ( 4 – λ ) x2 8x 2 – 5x 3 –x3 –( 3 – λ ) x3 0 = 0 0 (5. Therefore. is known. an eigenvector for λ = 1 is x1 Xλ = 1 = x2 = x3 2x 2 x2 3x 2 2 2 = x2 1 = 1 3 3 (5. and x 3 = 3x 2 . Third Edition Copyright © Orchard Publications . and x 3 are scalars. we obtain the indeterminate values x1 = 0 ⁄ 0 x2 = 0 ⁄ 0 x3 = 0 ⁄ 0 (5. and State Equations 5 0 2 x1 7 –5 x1 4 –1 x2 = λ x2 8 –3 x3 x3 7x 2 – 5x 3 4x 2 –x3 8x 2 – 3x 3 λ x1 = λ x2 λ x3 (5. Then. x 2. State Variables. an eigenvector for λ = 2 is 5−40 Numerical Analysis Using MATLAB® and Excel®.165) By Crame’s rule.167).

170) We observe that for the first unit eigenvector the sum of the squares is unity. that is.+ ----.+ --------. We find the unit eigenvectors by dividing the components of each vector by the square root of the sum of the squares of the components. 2 ⎞2 ⎛ 1 ⎞2 ⎛ 3 ⎞2 4 1 9 ⎛ --------. we obtain x 1 = – x 2 . and x 3 = x 2 . These are: 2 +1 +3 = 1 +1 +2 = ( –1 ) + 1 + 1 = 2 2 2 2 2 2 2 2 2 14 6 3 The unit eigenvectors are 2--------14 1Unit X λ = 1 = --------14 3--------14 1-----6 1Unit X λ = 2 = -----6 2-----6 –1 -----3 1Unit X λ = 3 = -----3 1-----3 (5.+ ----.Eigenvectors x1 Xλ = 2 = x2 = x3 x2 x2 2x 2 1 1 = x2 1 = 1 2 2 (5. Then. Numerical Analysis Using MATLAB® and Excel®.= 1 ⎝ 14 ⎠ ⎝ 14 ⎠ ⎝ 14 ⎠ 14 14 14 (5. Third Edition Copyright © Orchard Publications 5−41 .168) Finally. for λ = 3 .169) c.+ --------= ----.171) and the same is true for the other two unit eigenvectors in (5. an eigenvector for λ = 3 is x1 Xλ = 3 = x2 = x3 –x2 x2 x2 = x2 –1 –1 = 1 1 1 1 (5.170).

also known as the complementary solution. The total solution of the non−homogeneous ODE is the summation of the natural and forces responses.+ b0 x . …. x .+ bm – 1 d ---------------. is referred to as the natural response. the initial conditions must be satisfied. Third Edition Copyright © Orchard Publications . If the forcing function is zero. Also. if y = f ( x) .+ … + a 1 dy --------. is not zero. that is. and State Equations 5. and t is said to be a partial differential equation and it is abbreviated as PDE.12 Summary • Differential equations are classified by: Type − Ordinary or Partial Order − The highest order derivative which is included in the differential equation Degree − The exponent of the highest power of the highest order derivative after the differential equation has been cleared of any fractions or radicals in the dependent variable and its derivatives • If the dependent variable y is a function of only a single variable x . the differential equation which relates y and x is said to be an ordinary differential equation and it is abbreviated as ODE. and is denoted as y N ( t ) or simply y N . • If the dependent variable y is a function of two or more variables such as y = f ( x. 5−42 Numerical Analysis Using MATLAB® and Excel®. the differential equation that relates y . the differential equation is referred to as homogeneous differential equation. in engineering the solution of the homogeneous ODE. • Generally. t ) . and is denoted as y F ( t ) or simply y F . • The most general solution of an homogeneous ODE is the linear combination yH ( t ) = k1 y1 ( t ) + k2 y2 ( t ) + k3 y3 ( t ) + … + kn yn ( t ) where the subscript H is used to denote homogeneous and k 1.Chapter 5 Differential Equations. k n are arbitrary constants. The particular solution of a non−homogeneous ODE is be referred to as the forced response.+ a0 y = bm d ----. k 2. that is.+ … + b 1 dx a n -------n n–1 m n–1 dt dt dt dt dt dt n n–1 m m–1 is a non−homogeneous differential equation if the right side. known as forcing function. where x and t are independent variables. k 3. • A function y = f ( x) is a solution of a differential equation if the latter is satisfied when y and its derivatives are replaced throughout by f ( x) and its corresponding derivatives. State Variables.+ an – 1 d --------------. • The ODE d y y x x ----.

a 1. …. • For an nth order homogeneous differential equation the solutions are y 1 = k 1 e . …. contains no arbitrary constants. • In certain non−homogeneous ODEs.1.Summary y( t) = y Natural Response +y Forced Response = yN + yF The natural response y N contains arbitrary constants and these can be evaluated from the given initial conditions. …. • With the method of undetermined coefficients. The form of the forced response for second order non−homogeneous ODEs is given in Table 5. the right side f ( t ) cannot be determined by the method of undetermined coefficients. For these ODEs we must use the method of variation of parameNumerical Analysis Using MATLAB® and Excel®. the forced response is a function similar to the right side of the non−homogeneous ODE. these occur as complex conjugate pairs. It is imperative to remember that the arbitrary constants of the natural response must be evaluated from the total response. however. then another root is s 2 = – α – j β . the n solutions of the natural response are independent and the most general solution is: yN = k1 e s1 t + k2 e s2 t + … + kn e sn t • If the solution of the characteristic equation contains m equal roots. y 2 = k 2 e . Thus. for two complex conjugate roots we evaluate the constants from the expressions k1 e s1 t + k2 e s2 t = e –α t ( k 3 cos β t + k 4 sin β t ) = e –α t k 5 cos ( β t + ϕ ) • The forced response of a non−homogeneous ODE can be found by the method of undetermined coefficients or the method of variation of parameters. if one root is s 1 = – α + j β where α and β are real numbers. s 2. s n are the solutions of the characteristic equation an s + an – 1 s n n–1 + … + a1 s + a0 = 0 and a n. y n = k n e s1 t s2 t s3 t sn t where s 1. The forced response y F . the most general solution has the form: yN = ( k1 + k2 t + … + km t m–1 )e s1 t + kn – i e s2 t + … + kn e sn t • If the characteristic equation contains complex roots. y 3 = k 3 e . Third Edition Copyright © Orchard Publications 5−43 . Then. a 0 are the constant coefficients of the ODE • If the roots of the characteristic equation are distinct. a n – 1.

and k 2 are scalar constants. which when substituted into the first will yield the total solution. or simply state equations. the method of variation of parameters replaces the constants k 1 and k 2 by two variables u 1 and u 2 that satisfy the following three relations: y = u1 y1 + u2 y2 du2 du1 -----. The resulting first−order differential equations are called state space equations.= f (t) dt dt dt dt Simultaneous solution of last two expressions above will yield the values of du1 ⁄ dt and du 2 ⁄ dt . β . integration of these will produce u 1 and u 2 .+ ------. State Variables. then. Third Edition Copyright © Orchard Publications . • An nth−order differential equation can be resolved to n first−order simultaneous differential equations with a set of auxiliary variables called state variables. the solution of the first state equation above is x( t) = e α ( t – t0 ) t x0 + e αt ∫t e 0 – ατ β u ( τ ) dτ • In a system of state equations of the form · = Ax + bu x y = Cx + du 5−44 Numerical Analysis Using MATLAB® and Excel®. if non−zero is denoted as x 0 = x ( t 0 ) .⋅ ------. • The state representation of a system can be described by the pair of the of the state space equations · = Ax + bu x y = Cx + du • In a system of state equations of the form · = αx + βu x y = k1 x + k2 u where α .⋅ ------.Chapter 5 Differential Equations. and the initial condition. and State Equations ters.y1 + -------y = 0 dt dt 2 du1 dy1 du 2 dy 2 ------. k 1 . This method will work with all linear equations including those with variable coefficients provided that the general form of the natural response is known. • For second order ODEs with constant coefficients.

if all eigenvalues of a given matrix A are distinct.A t + … + ---= I + At + ---2! n! 3! and I is the n × n identity matrix. if λ 1 = λ 2 = λ 3 … = λ m . that is. λ m + 1 .A t + ----A t . 2. i = 1. the coefficients a i of the state transition matrix e At = a0 I + a1 A + a2 A + … + an – 1 A 2 n–1 Numerical Analysis Using MATLAB® and Excel®. and m of these roots are equal. the eigenvalues λ i . the coefficients a i are found from the simultaneous solution of the following system of equations: a0 + a1 λ1 + a2 λ1 + … + an – 1 λ1 a0 + a1 λ2 + a2 λ2 + … + an – 1 λ2 … a0 + a1 λn + a2 λn + … + an – 1 λn 2 n–1 2 2 n–1 n–1 λ1 t λ2 t = e = e = e λn t • If the polynomial of det [ A – λ I ] = 0 has n roots. • If λ 1 ≠ λ 2 ≠ λ 3 ≠ … ≠ λ n . This theorem states that a matrix can be expressed as an ( n – 1 ) th degree polynomial in terms of the matrix A as e At At = a0 I + a1 A + a2 A + … + an – 1 A 2 n–1 where the coefficients a i are functions of the eigenvalues λ . and b and d are column vectors with two or more rows. • If A is an n × n matrix. that is. Third Edition Copyright © Orchard Publications 5−45 . n of A are the roots of the nth order polynomial det [ A – λ I ] = 0 • Evaluation of the state transition matrix e is based on the Cayley−Hamilton theorem. and I be the n × n identity matrix. λ n .Summary where for two or more simultaneous differential equations A and C are 2 × 2 or higher order matrices. …. the solution of the matrix · = Ax + bu with initial conditions x ( t ) = x is obtained from the differential equation x 0 0 relation x( t) = e A ( t – t0 ) x0 + e At ∫t e 0 t –A τ bu ( τ ) dτ where the state transition matrix e ϕ(t) ≡ e At At is defined as the matrix power series 1 22 1 33 1 nn .

and λ is a scalar number. and State Equations are found from the simultaneous solution of the system of equations below.( a0 + a1 λ1 + a2 λ2 -e 1 ) = -------------1 + … + an – 1 λ1 m–1 m–1 d λ1 d λ1 m–1 m–1 2 2 a0 + a1 λm + 1 + a2 λm + 1 + … + an – 1 λm + 1 = e a 0 + a 1 λn + a 2 λ n + … + a n – 1 λ n 2 n–1 2 n–1 λ m + 1t … = e λn t • We can use the MATLAB eig(x) function to find the eigenvalues of an n × n matrix. Obviously. 5−46 Numerical Analysis Using MATLAB® and Excel®. X is a non−zero column vector. they are normalized to unit length. so that the sum of the squares of their components is equal to unity.Chapter 5 Differential Equations. This is done by dividing each component of the eigenvector by the square root of the sum of the squares of their components. there is a different eigenvector for each eigenvalue. the vector X is called eigenvector. a0 + a1 λ1 + a2 λ1 + … + an – 1 λ1 2 n–1 = e λ1 t dd . Eigenvectors are generally expressed as unit eigenvectors. State Variables. that is. Third Edition Copyright © Orchard Publications . • If A is an n × n matrix.λ1 t 2 n–1 -------e ( a 0 + a 1 λ 1 + a 2 λ 1 + … + a n – 1 λ 1 ) = ------d λ1 d λ1 d λ1 t d 2 n–1 ------( a 0 + a 1 λ 1 + a 2 λ 1 + … + a n – 1 λ 1 ) = ------e 2 2 d λ1 d λ1 … λ t d d n–1 -------------.

+ y = sec t 2 dt 2 5 Express the integro−differential equation below as a matrix of state equations where k 1. d ------.+ 4 ----2 dt 2 2 dt dy y . ·1 x ·2 x ·3 x ·4 x x1 0 1 0 0 0 x = 0 0 1 0 ⋅ 2 + 0 u(t) x3 0 0 0 1 0 –1 –2 –3 –4 1 x4 7. B . and k 3 are constants. and C below.( cos 2t + 1 ) 3.Exercises 5.+ 3y = t – 1 1. Third Edition Copyright © Orchard Publications 5−47 . Numerical Analysis Using MATLAB® and Excel®.+ 4 ----2 dt dt 2 dy 1 2 2 d y .+ 2 ----2 dt dt 2 y 4.+ y = cos t Hint: Use cos t = -. d ------. and let x(y) = y(t) .+ k2 v + k1 ------+ k 3 ----2 dt dt ∫0 v dt t = sin 3t + cos 3t 6. d ------. 2 dv dv . dy y . k 2.+ 3y = 4e – t 2. Express the matrix of the state equations below as a single differential equation. Compute the eigenvalues of the matrices A . A = 1 2 3 –1 B = a 0 –a b 0 1 0 C = 0 0 1 – 6 – 11 – 6 Hint: One of the eigenvalues of matrix C is – 1 .13 Exercises Solve the following ODEs by any method and verify your answers with MATLAB. ------.

and State Equations 8. Third Edition Copyright © Orchard Publications .Chapter 5 Differential Equations. Compute e At given that A = 0 1 0 0 0 1 – 6 – 11 – 6 5−48 Numerical Analysis Using MATLAB® and Excel®. State Variables.

The characteristic equation of the homogeneous part is s + 4s + 3 = 0 from which s 1 = – 1 and s 2 = – 3 .14 Solutions to End−of−Chapter Exercises 1. we refer to Table 5.1 and we assume a solution of –t –t –t the form y F = k 3 te where we multiplied e by t to avoid the duplication with k 1 e . Thus y N = k 1 e + k 2 e –t – 3t 2 . y=simple(y) y = -7/9+1/3*t+C1/exp(t)+C2/exp(t)^3 2. y = k1 e + k2 e –t – 3t 1 -t – 7 + --3 9 Check with MATLAB: y=dsolve('D2y+4*Dy+3*y=t−1’). we refer to Table 5. Third Edition Copyright © Orchard Publications 5−49 .1 and we assume a solution of the form y F = k 3 t + k 4 and the total solution is y = k1 e + k2 e –t – 3t + k3 t + k4 The first and second derivatives of y are dy ⁄ dt = – k 1 e – 3k 2 e 2 2 –t –t – 3t + k3 – 3t d y ⁄ dt = k 1 e + 9k 2 e and by substitution into the given ODE k 1 e + 9k 2 e –t – 3t + 4 ( – k 1 e – 3k 2 e –t – 3t + k3 ) + 3 ( k1 e + k2 e –t – 3t + k3 t + k4 ) = t – 1 Equating like terms we obtain 4k 3 + 3k 3 t + 3k 4 = t – 1 3k 3 t = t 4k 3 + 3k 4 = – 1 and simultaneous solution of the last two yields k 3 = 1 ⁄ 3 and k 4 = – 7 ⁄ 9 .Solutions to End−of−Chapter Exercises 5. Therefore. The characteristic equation of the homogeneous part is the same as for Exercise 1 and thus yN = k1 e + k2 e –t – 3t . For the forced response. By substitution of this assumed solution into the given ODE and using MATLAB to find the first and second derivatives we obtain: y = k1 e + k2 e –t – 3t + k 3 te –t Numerical Analysis Using MATLAB® and Excel®. For the forced response.

f1=simple(y1) % Define symbolic variables % Assumed form of total solution % Compute and simplify first derivative f1 = -k3*exp(-t)*(-1+t) Thus. Third Edition Copyright © Orchard Publications . Therefore. y1=diff(y0). 3. the first derivative of y F is dyF ⁄ dt = k 3 e – k 3 te y2=diff(y0. syms t k3 y0=k3*t*exp(−t).Chapter 5 Differential Equations. State Variables. The characteristic equation yields two equal roots s 1 = s 2 = – 1 and thus the natural response has the form y N = k 1 e + k 2 te –t –t For the forced response we assume a solution of the form y F = k 3 cos 2t + k 4 sin 2t + k5 5−50 Numerical Analysis Using MATLAB® and Excel®. y = k1 e + k2 e –t + 2te –t Check with MATLAB: y=dsolve('D2y+4*Dy+3*y=4*exp(−t)'). y=simple(y) 2*t/exp(t)-1/exp(t)+C1/exp(t)+C2/exp(t)^3 We observe that the second and third terms of the displayed expression above have the same form and thus they can be combined to form a single term C3/exp(t). f2=simple(y2) –t –t % Compute and simplify second derivative f2 = k3*exp(-t)*(-2+t) and the second derivative of y is d yF ⁄ dt = – 2k 3 e + k 3 te f=y2+4*y1+3*y0.2). and State Equations We will use MATLAB to find the first and second derivatives of this expression. f=simple(f) 2 2 –t –t % Form and simplify the left side of the given ODE f = 2*k3/exp(t) and by substitution into the given ODE 2k 3 e –t = 4e – 3t –t or k 3 = 2 .

Third Edition Copyright © Orchard Publications 5−51 . f1=simple(y1) % Define symbolic variables % Assumed form of total solution % Compute and simplify first derivative f1 = -2*k3*sin(2*t)+2*k4*cos(2*t) Thus. y1=diff(y0).2).Solutions to End−of−Chapter Exercises We will use MATLAB to find the first and second derivatives of this expression. the forced response is y F = ( – 3 ⁄ 50 ) cos 2t + ( 4 ⁄ 50 ) sin 2t + 1 ⁄ 2 and the total response is –t – t 1 3 cos 2t – 4 sin 2t . Therefore. syms t k1 k2 k3 k4 k5 y0=k3*cos(2*t)+k4*sin(2*t)+k5. the first derivative of y F is dyF ⁄ dt = – 2k 3 sin 2t + 2k 4 cos 2t y2=diff(y0.– -------------------------------------y = k 1 e + k 2 te + -2 50 Check with MATLAB: Numerical Analysis Using MATLAB® and Excel®. f=simple(f) % Form and simplify the left side of the given ODE 2 2 f = -3*k3*cos(2*t)-3*k4*sin(2*t)-4*k3*sin(2*t)+4*k4*cos(2*t)+k5 Simplifying this expression and equating with the right side of the given ODE we obtain: cos 2t + 1 -( – 3k 3 + 4k 4 ) cos 2t – ( 4k 3 + 3k 4 ) sin 2t + k 5 = ------------2 2 Equating like terms and solving for the k terms we obtain – 3k 3 + 4k 4 = 1 ⁄ 2 – 4k 3 – 3k 4 = 0 k5 = 1 ⁄ 2 Simultaneous solution of the first two equations above yields k 3 = – 3 ⁄ 50 and k 4 = 4 ⁄ 50 . f2=simple(y2) % Compute and simplify second derivative f2 = -4*k3*cos(2*t)-4*k4*sin(2*t) and the second derivative of y is d yF ⁄ dt = – 4k 3 cos 2t – 4k 4 sin 2t f=y2+2*y1+y0.

Third Edition Copyright © Orchard Publications .= ---------------------------------------.+ ------.= f ( t ) = ------.cos t + ------. if not impossible. Therefore. State Variables.= – tan t (2) 2 2 cos t sin t dt 1 cos t + sin t – sin t cos t 0 sin t and cos t 0 du 2 – sin t sec t 1 ------. du1 du 2 ------.= -----------------------------------. from (5. f=simple(y) f = -3/50*cos(2*t)+2/25*sin(2*t)+1/2+C1*exp(-t)+C2*exp(-t)*t 4.⋅ ------.⋅ ------.= 1 dt 1 1 (3) Integration of (2) and (3) above and substitution into (1) yields u1 = ∫ ( – tan t ) dt = – ( – ln cos t ) + k 1 = ln cos t + k 1 u2 = ∫ dt = t + k2 5−52 Numerical Analysis Using MATLAB® and Excel®. du1 dy1 du 2 dy 2 du 2 du1 ------. by (5. The characteristic equation is s + 1 = 0 from which s = ± j and thus the natural response is yN = k1 e + k2 e jt –j t 2 We let y 1 = cos t and y 2 = sin t Then.= ---------------------------------------. It is very difficult.Chapter 5 Differential Equations. to assume a solution for the forced response of this ODE.69).= -.( – sin t ) + ------dt dt dt dt dt dt Next.( cos t ) = sec t . and State Equations y=dsolve('D2y+2*Dy+y=cos(2*t)/2+1/2').y + -------y = 0 dt 1 dt 2 or du 2 du1 ------. we will use the method of variation of parameters.68) the solution is y = u1 y 1 + u 2 y 2 = u1 cos t + u 2 sin t (1) Also.= -----------.sin t = 0 dt dt and from (5.70). we find du1 ⁄ dt and du 2 ⁄ dt by Cramer’s rule as follows: sin t – --------du1 sec t cos t cos t – tan t .

– k 2 dv ------3 ------2 3 dt 1 dt dt 2 We let v = x1 · dv ----. and by substitution into (1) · x 3 = – k 1 x 1 – k 2 x 2 – k 3 x 3 + 3 ( cos 3t – sin 3t ) Thus.= x3 3 dt 2 · dv ------.Solutions to End−of−Chapter Exercises y = u1 y 1 + u 2 y 2 = ( ln cos t + k 1 ) cos t + ( t + k 2 ) sin t = k 1 cos t + k 2 sin t + t sin t + cos t ( ln cos t ) Check with MATLAB: y=dsolve('D2y+y=sec(t)').+ k 3 ------+ k 2 ----2 3 dt dt dt 2 or 3 dv .+ k 1 v = 3 cos 3t – 3 sin 3t = 3 ( cos 3t – sin 3t ) ------. Differentiating the given integro−differential equation with respect to t we obtain 3 dv dv dv . the state equations are · x1 = x2 · x2 = x3 · x 3 = – k 1 x 1 – k 2 x 2 – k 3 x 3 + 3 ( cos 3t – sin 3t ) and in matrix form · x1 x1 0 1 0 0 · = 0 0 1 ⋅ x + 0 ⋅ 3 ( cos 3t – sin 3t ) x2 2 –k1 –k2 –k3 · 1 x3 x3 Numerical Analysis Using MATLAB® and Excel®.= x2 = x1 dt 3 · dv ------. Third Edition Copyright © Orchard Publications 5−53 .– k v + 3 ( cos 3t – sin 3t ) (1) .= x3 = x2 2 dt Then.= – k dv ----. f=simple(y) f = sin(t)*t+log(cos(t))*cos(t)+C1*sin(t)+C2*cos(t) 5.

2 λ + 6 λ + 11 λ + 6 --------------------------------------------. Third Edition Copyright © Orchard Publications .+ 2 dy ------3 2 4 dt dt dt dt 7. and State Equations 6. Then. λ2 = –2 λ1 = –3 a. a. – 1 – λ + λ + λ – 6 = 0 . Then. 0 1 0 C = 0 0 1 – 6 – 11 – 6 ⎛ 0 1 0 1 0 0⎞ ⎜ ⎟ det ( C – λ I ) = det ⎜ 0 0 1 –λ 0 1 0 ⎟ ⎜ ⎟ 0 0 1⎠ ⎝ – 6 – 11 – 6 = det –λ 1 0 0 –λ 1 =0 – 6 – 11 – 6 – λ λ ( – 6 – λ ) – 6 – ( – 11 ) ( – λ ) = λ + 6 λ + 11 λ + 6 = 0 and it is given that λ 1 = – 1 . and thus λ 1 = a λ2 = b c. State Variables. A = 1 2 3 –1 ⎛ ⎞ 2 det ( A – λ I ) = det ⎜ 1 2 – λ 1 0 ⎟ = det 1 – λ = 0 ⎝ 3 –1 0 1⎠ 3 –1 –λ 2 2 ( 1 – λ ) ( – 1 – λ ) – 6 = 0 .+ y = u(t) ------.Chapter 5 Differential Equations.+ 3 dy ------. λ = 7 .+ 4 dy ----. Expansion of the given matrix yields · x1 = x2 · x2 = x3 · x3 = x2 3 · x 4 = – x 1 – 2x 2 – 3x 3 – 4x 4 + u ( t ) 2 Letting x = y we obtain 4 dy .= λ + 5λ + 6 ⇒ (λ + 1)(λ + 2)(λ + 3) = 0 (λ + 1) 3 2 2 3 2 and thus λ 1 = – 1 8. B = a 0 –a b ⎛ ⎞ det ( B – λ I ) = det ⎜ a 0 – λ 1 0 ⎟ = det a – λ 0 = 0 ⎝ –a b 0 1⎠ –a b – λ ( a – λ ) ( b – λ ) = 0 . Matrix A is the same as Matrix C in Exercise 7. λ1 = –1 λ2 = –2 λ1 = –3 5−54 Numerical Analysis Using MATLAB® and Excel®. and thus λ 1 = 7 λ2 = – 7 b.

1 −3 9]. a0=3*exp(−t)−3*exp(−2*t)+exp(−3*t). 5/2*exp(-t)-4*exp(-2*t)+3/ 2*exp(-3*t). A=[0 1 0. 5/2*exp(-t)-16*exp(2*t)+27/2*exp(-3*t). A=[1 −1 1.Solutions to End−of−Chapter Exercises and since A is a 3 × 3 matrix the state transition matrix is e At = a0 I + a1 A + a2 A 2 2 2 λ1 t λ2 t λ3 t 2 (1) –t – 2t – 3t Then. disp('a2 = ').5e – 2t + 0.. fprintf(' \n').. exp(−3*t)]'). disp('a1 = '). a2=1/2*exp(−t)−exp(−2*t)+1/2*exp(−3*t). fprintf(' \n'). disp(x(2)).. a 0 = 3e – 3e –t –t –t – 2t + 3e – 3t – 3t a 1 = 2.. 0 0 1. disp(x(1)).. exp(−2*t).. Third Edition Copyright © Orchard Publications 5−55 .. -1/2*exp(-t)+2*exp(-2*t)-3/2*exp(-3*t)] [3*exp(-t)-12*exp(-2*t)+9*exp(-3*t).5e – 4e a 2 = 0. 1 −2 4. a=sym('[exp(−t).. disp('a0 = ').5e – e – 2t + 1. 1/2*exp(-t)-4*exp(-2*t)+9/2*exp(3*t)] Then. a1=5/2*exp(−t)−4*exp(−2*t)+3/2*exp(−3*t)...5e – 3t Now. disp(x(3)) a0 = 3*exp(-t)-3*exp(-2*t)+exp(-3*t) a1 = 5/2*exp(-t)-4*exp(-2*t)+3/2*exp(-3*t) a2 = 1/2*exp(-t)-exp(-2*t)+1/2*exp(-3*t) Thus. a0 + a1 λ1 + a2 λ1 = e a0 + a1 λ2 + a2 λ2 = e a0 + a1 λ3 + a2 λ3 = e ⇒ a0 – a1 + a2 = e ⇒ a 0 – 2a 1 + 4a 2 = e ⇒ a 0 – 3a 1 + 9a 2 = e syms t. Numerical Analysis Using MATLAB® and Excel®.. eAt=a0*eye(3)+a1*A+a2*A^2 eAt = [3*exp(-t)-3*exp(-2*t)+exp(-3*t).. -5/2*exp(-t)+8*exp(2*t)-9/2*exp(-3*t). 1/2*exp(-t)-exp(-2*t)+1/2*exp(-3*t)] [-3*exp(-t)+6*exp(-2*t)-3*exp(-3*t). x=A\a. we compute e At of (1) with the following MATLAB code: syms t. −6 −11 −6].

5e – 3t 5−56 Numerical Analysis Using MATLAB® and Excel®.5e – 4e – 2t – 1.5 e + 2e 0. State Variables.5e – e –t –t –t – 2t + 0.Chapter 5 Differential Equations.5e – 3t – 3t – 3t 0. and State Equations 3e – 3e 3e – 12e –t –t – 2t e At +e – 3t 2.5 e + 8e 2.5e – 3t – 3t = – 3 e – t + 6e – 2t – 3e –3t – 2t – 2.5e – 16e –t – 2t – 4.5e – 4e –t –t – 2t + 1.5e – 2t + 4.5e – 0.5e + 9e – 3t – 2t + 13. Third Edition Copyright © Orchard Publications .

Thus. the terms with the coefficients a 2 and b 2 together. In general. The terms with the coefficients a 1 and b 1 together. and represents the DC (average) component of f ( t ) . Third Edition Copyright © Orchard Publications 6−1 . and so on.1 Wave Analysis The French mathematician Fourier found that any periodic waveform. a third harmonic of 3 MHz . that is.. For example. the term a 0 ⁄ 2 is the average value of v ( t ) or i ( t ) .1) or 1 -a + f ( t ) = -2 0 n=1 ∑ ( a cos n ω t + b sin n ω t ) n n ∞ (6.e. a series of sinusoids with frequencies 1 MHz . any periodic waveform f ( t ) can be expressed as 1 . and Maclaurin Series T his chapter is an introduction to Fourier and power series. Several examples are presented to illustrate the approach. we discuss the different types of symmetry and how they can be used to predict the terms that may be present. if f ( t ) represents some voltage v ( t ) . or current i ( t ) . Numerical Analysis Using MATLAB® and Excel®. The alternate trigonometric and the exponential forms are also presented. and so on. a second harmonic of 2 MHz . 3 MHz . represent the second harmonic component 2 ω . We begin with the definition of sinusoids that are harmonically related and the procedure for determining the coefficients of the trigonometric form of the series. Likewise. contains the fundamental frequency of 1 MHz . 2 MHz . represent the fundamental frequency component ω *. i. a waveform that repeats itself after some time. Taylor. * We recall that k 1 cos ω t + k2 sin ω t = k cos ( ω t + θ ) where θ is a constant. can be expressed as a series of harmonically related sinusoids. We conclude with a discussion on power series expansion with the Taylor and Maclaurin series.a + a 1 cos ω t + a 2 cos 2 ω t + a 3 cos 3 ω t + a 4 cos 4 ω t + … f ( t ) = -2 0 + b 1 sin ω t + b 2 sin 2 ω t + b 3 sin 3 ω t + b 4 sin 4 ω t + … (6. Then.2) where the first term a 0 ⁄ 2 is a constant. sinusoids whose frequencies are multiples of a fundamental frequency (or first harmonic). and so on. 6.Chapter 6 Fourier.

5) is also is zero since 2π 2π sin mt dt = 0 (6.Chapter 6 Fourier. and for convenience.3) and (6. the second harmonic. we have assumed that ω = 1 . and Maclaurin Series Since any periodic waveform f ( t ) can be expressed as a Fourier series.4) are zero since the net area over the 0 to 2 π area is zero.3) (6. the fundamental.1. Then.1. the sum of two or more sinusoids of different frequencies produce a waveform that is not a sinusoid as shown in Figure 6. Let us consider the functions sin mt and cos m t where m and n are any integers. the product of the sine and cosine functions under the integral evaluated from 0 to 2 π is zero. and so on. must produce the waveform f ( t ) . 2nd Harmonic Total 3 2 1 0 -1 -2 -3 0 2 4 6 8 10 12 Fundamental 3rd Harmonic Figure 6.5) 2π cos m t dt = 0 ( sin mt ) ( cos nt ) dt = 0 The integrals of (6.2 Evaluation of the Coefficients Evaluations of a i and b i coefficients of (6. This will be shown shortly.1) is not a difficult task because the sine and cosine are orthogonal functions. Summation of a fundamental. Generally. that is. The inte- 6−2 Numerical Analysis Using MATLAB® and Excel®. it follows that the sum of the DC . ∫0 ∫0 ∫0 gral of (6.4) (6. second and third harmonic 6. Taylor. Third Edition Copyright © Orchard Publications .

∫0 since 2π ( sin mt ) ( sin nt ) dt = 0 (6.Evaluation of the Coefficients 1 . where m = 2 and n = 3 . sin x sin x ⋅ cos x cos x Figure 6.6) can also be confirmed graphically as shown in Figure 6.6) 1 . We observe that the net shaded area above and below the time axis is zero.2.2. Third Edition Copyright © Orchard Publications 6−3 . then. sin 2x sin 3x sin 2x ⋅ sin 3x Numerical Analysis Using MATLAB® and Excel®. where we observe that the net shaded area above and below the time axis is zero. if m and n are different integers. Graphical proof of ∫0 2π ( sin mt ) ( cos nt ) dt = 0 Moreover.3.[ sin ( x + y ) + sin ( x – y ) ] sin x cos y = -2 This is also obvious from the plot of Figure 6.[ cos ( x – y ) – cos ( x – y ) ] ( sin x ) ( sin y ) = -2 The integral of (6.

Chapter 6 Fourier.8) ∫0 2π ( cos m t ) dt = π 2 (6.[ cos ( x + y ) + cos ( x – y ) ] ( cos x ) ( cos y ) = -2 The integral of (6. if m and n are different integers. ∫0 since 2π ( cos m t ) ( cos nt ) dt = 0 (6.8) and (6. Graphical proof of ∫0 2π ( sin mt ) ( sin nt ) dt = 0 for m = 2 and n = 3 Also.9) The integrals of (6.7) can also be confirmed graphically as shown in Figure 6. We observe that the net shaded area above and below the time axis is zero.6) and (6.7) 1 . The simpli* We will discuss orthogonal functions in Chapter 14 6−4 Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications . cos 3x cos 2x cos 2x ⋅ cos 3x Figure 6.5 and 6. then. and Maclaurin Series Figure 6.4. where m = 2 and n = 3 . ∫0 and 2π ( sin mt ) dt = π 2 (6. if in (6. m = n .6.3.9) can also be seen to be true graphically with the plots of Figures 6. It was stated earlier that the sine and cosine functions are orthogonal* to each other.4.7). Graphical proof of ∫0 2π ( cos m t ) ( cos nt ) dt = 0 for m = 2 and n = 3 However. then. Taylor.

Graphical proof of ∫0 2π ( sin mt ) dt = π 2 cos x cos x 2 Figure 6. Then. 1 . sin x sin x 2 Figure 6. for simplicity.5. 1 .10) by sin 2t . we multiply both sides of (6.a sin 2t + a 1 cos t sin 2t + a 2 cos 2t sin 2t + a 3 cos 3t sin 2t + a 4 cos 4t sin 2t + … f ( t ) sin 2t = -2 0 b 1 sin t sin 2t + b 2 ( sin 2t ) + b 3 sin 3t sin 2t + b 4 sin 4t sin 2t + … 2 Numerical Analysis Using MATLAB® and Excel®.6.1). Graphical proof of ∫0 2π ( cos m t ) dt = π 2 In (6.a + a 1 cos t + a 2 cos 2t + a 3 cos 3t + a 4 cos 4t + … f ( t ) = -2 0 + b 1 sin t + b 2 sin 2t + b 3 sin 3t + b 4 sin 4t + … (6. becomes apparent in the discussion that follows. say b 2 . Third Edition Copyright © Orchard Publications 6−5 .10) To evaluate any coefficient.Evaluation of the Coefficients fication obtained by application of the orthogonality properties of the sine and cosine functions. Then. we let ω = 1 .

14) 6−6 Numerical Analysis Using MATLAB® and Excel®. a n .12) ∫0 2π f ( t ) cos nt dt (6.a 0 = ----2 2π 1 a n = -π ∫0 2π f ( t ) dt (6. and we integrate over the period 0 to 2 π . The remaining coefficients can be evaluated similarly. Third Edition Copyright © Orchard Publications . Then. Therefore. Taylor. we multiply both sides of the above expression by dt . and b n are found from the following relations. 1 1-.Chapter 6 Fourier.11) sin 3t sin 2t dt 2π sin 4t sin 2t dt + … We observe that every term on the right side of (6. ∫0 2π 1 -a f ( t ) sin 2t dt = -2 0 ∫0 2π sin 2t dt + a 1 ∫0 2π cos t sin 2t dt + a 2 ∫0 2π cos 2t sin 2t dt + a3 + b1 + b4 ∫0 2π cos 3t sin 2t dt + a 4 sin t sin 2t dt + b 2 ∫0 2π cos 4t sin 2t dt + … 2 ∫0 ∫0 2π ∫0 2π ( sin 2t ) dt + b 3 ∫0 2π (6.13) 1 b n = -π ∫0 2π f ( t ) sin nt dt (6.11) reduces to ∫0 or 2π f ( t ) sin 2t dt = b 2 ∫0 2π ( sin 2t ) dt = b 2 π 2 1 b 2 = -π ∫0 2π f ( t ) sin 2t dt and thus we can evaluate this integral for any given function f ( t ) . (6.11) except the term b2 ∫0 2π ( sin 2t ) dt 2 is zero as we found in (6. The coefficients a 0 .6) and (6.7). and Maclaurin Series Next.

if it is an even function. all even (even cosine and even sine) harmonics will be zero. We will discuss three types of symmetry that can be used to facilitate the computation of the trigonometric Fourier series form. are even functions. only odd (odd cosine and odd sine) harmonics will be present.– ---t. if it is an odd function. that is. f ( –t ) = f ( t ) (6.12) yields the average ( DC ) value of f ( t ) . Still other waveforms have or have not DC components. Third Edition Copyright © Orchard Publications 6−7 . cosine. do not have the average.7.+ ---cos t = 1 – ---2! 4! 6! 2 4 6 Other examples of even functions are shown in Figure 6. if f ( t ) is an odd function. Numerical Analysis Using MATLAB® and Excel®. In other words. In other words. 2.Symmetry The integral of (6. In other words. it is possible to predict which terms will be present in the trigonometric Fourier series. Odd symmetry − If a waveform has odd symmetry. the cosine function is an even function because it can be written as the power series* t . For instance. will be zero. Thus. the series will consist of cosine terms only. the most common waveforms used in science and engineering. and a 0 may or may not be zero. A function f ( t ) is an even function of time if the following relation holds. if in an even function we replace t with – t . Even symmetry − If a waveform has even symmetry. Some waveforms have cosine terms only. 6. all the a i coefficients including a 0 .+ … t . and with or without constants. Half−wave symmetry − If a waveform has half−wave symmetry (to be defined shortly). that is. and sine terms all present. if f ( t ) is an even function. These are: 1. * We will discuss power series later in this chapter. the function f ( t ) does not change. Fortunately. all the b i coefficients will be zero. 3. the series will consist of sine terms only. while others have sine terms only. polynomials with even exponents only. and odd functions have nothing to do with odd harmonics.15) that is.6.3 Symmetry With a few exceptions such as the waveform of Example 6. by observing whether or not the given waveform possesses some kind of symmetry. We will now define even and odd functions and we should remember that even functions have nothing to do with even harmonics.

However. 2 The product of two even or two odd functions is an even function. if in an odd function we replace t with – t . Taylor. and an odd function with the subscript o . and Maclaurin Series f(t) t2 0 f(t) k t 0 t2 + k t f(t) 0 t Figure 6. f(t) mt 0 f(t) t3 t 0 f(t) t 0 t Figure 6. we obtain the negative of the function f ( t ) .7. if f ( 0 ) = 0 . the sine function is an odd function because it can be written as the power series t t t . Also. f e ( t ) and f o ( t ) will be used to represent even and odd functions of time respectively. f ( 0 ) = 0 . the reverse is not always true. we will denote an even function with the subscript e . For instance. An example of this is the function f ( t ) = t in Figure 6. Thus. is an odd function. polynomials with odd exponents only. Third Edition Copyright © Orchard Publications .8.Chapter 6 Fourier.+ ---.17) Numerical Analysis Using MATLAB® and Excel®.16) that is.– ----+… sin t = t – ---3! 5! 7! 3 5 7 Other examples of odd functions are shown in Figure 6. Examples of odd functions We observe that for odd functions. we should not conclude that f ( t ) is an odd function.7. Examples of even functions A function f ( t ) is an odd function of time if the following relation holds. Thus.8. and no constants are odd functions. Henceforth. ∫– T 6−8 T f e ( t ) dt = 2 ∫0 T f e ( t ) dt (6. that is. –f ( –t ) = f ( t ) (6. and the product of an even function times an odd function.

Symmetry and T ∫– T f o ( t ) dt = 0 (6.20) By addition of (6. We will test the waveforms of Figures 6. the function with value f ( t ) at any time t .16) with (6. a 0 = 0 . we recall that any periodic function with period T . To understand half−wave symmetry. but inverted. will have the same value again at a later time t + T. and therefore.18) A function f ( t ) that is neither even nor odd can be expressed as 1 . Numerical Analysis Using MATLAB® and Excel®. It is also an odd function and has half − wave symmetry since – f ( – t ) = f ( t ) and –f ( t + T ⁄ 2 ) = f ( t ) .21) that is. A periodic waveform with period T . 1. is expressed as f(t) = f(t + T) (6.22) that is.[ f ( t ) + f ( –t ) ] f e ( t ) = -2 (6. Third Edition Copyright © Orchard Publications 6−9 . any function of time can be expressed as the sum of an even and an odd function. has half−wave symmetry if –f ( t + T ⁄ 2 ) = f ( t ) (6. Square waveform For the waveform of Figure 6.19) or as 1 .9 through 6.9.17).[ f ( t ) –f ( –t ) ] f o ( t ) = -2 (6.13 for any of the three types of symmetry.23) that is. we get f ( t ) = fe ( t ) + fo ( t ) (6. the average value over one period T is zero. the shape of the negative half−cycle of the waveform is the same as that of the positive half−cycle.

10. Square waveform with ordinate axis shifted If we shift the ordinate axis π ⁄ 2 radians to the right. Sawtooth waveform For the sawtooth waveform of Figure 6. If there is half−wave symmetry. but has no half−wave symmetry 6−10 Numerical Analysis Using MATLAB® and Excel®. and observe the values of f ( t ) at the left and right points on the time axis. T A −π/2 π/2 2π π −2π −π 0 ωt T/2 −A T/2 Figure 6. a 0 = 0 . a 0 = 0 . these will always be equal but will have opposite signs as we slide the half−period T ⁄ 2 length to the left or to the right on the time axis at non−zero values of f ( t ) . we see that the square waveform now becomes an even function and has half−wave symmetry since f ( – t ) = f ( t ) and – f ( t + T ⁄ 2 ) = f ( t ) .Chapter 6 Fourier. the waveform will have neither odd nor even symmetry. if the ordinate axis is shifted by any other value other than an odd multiple of π ⁄ 2 . Third Edition Copyright © Orchard Publications . It is also an odd function because – f ( – t ) = f ( t ) .9. Taylor. Square waveform test for symmetry An easy method to test for half−wave symmetry is to choose any half−period T ⁄ 2 length on the time axis as shown in Figure 6. as shown in Figure 6.11. the average value over one period T is zero and therefore. Also. 2. Square waveform with ordinate shifted by π ⁄ 2 Obviously. such as f ( a ) and f ( b ) .9. 3. and Maclaurin Series T A T/2 f(b) π 2π 0 f(a) T/2 ωt −A Figure 6.10.

Third Edition Copyright © Orchard Publications 6−11 .Symmetry since – f ( t + T ⁄ 2 ) ≠ f ( t ) T A −2π −π 0 T/2 π T/2 2π ωt −A Figure 6. These waveforms can be either odd or even depending on the position of the ordinate. when separated by T ⁄ 2 . Triangular waveform For this triangular waveform of Figure 6. and third harmonic of a typical sinewave where the half period T ⁄ 2 . when separated by T ⁄ 2 are equal and opposite. Sawtooth waveform test for symmetry 4. It is also an odd function since – f ( – t ) = f ( t ) . are equal and opposite.13 shows a fundamental. Moreover.11. The second harmonic has no half−wave symmetry because the ordinates b on the left and b on the right. This is necessary in order to test the fundamental. and third harmonics for half−wave symmetry. Triangular waveform test for symmetry 5. all three waveforms have zero average value unless the abscissa axis is shifted up or down. there are not opposite in sign. the average value over one period T is zero and therefore. The third harmonic has half−wave symmetry since the c and – c values. Fundamental. although are equal. a 0 = 0 . Numerical Analysis Using MATLAB® and Excel®. Second and Third Harmonics of a Sinusoid Figure 6.12. it has half−wave symmetry because – f ( t + T ⁄ 2 ) = f ( t ) T A −2π −π T/2 0 π T/2 2π ωt −A Figure 6. The fundamental has half−wave symmetry since the a and – a values. second. is chosen as the half period of the period of the fundamental frequency. Also. second.12.

Of course.12) through (6. or an even function. Also. Fundamental. the limits of integration for the coefficients a n and b n are given as 0 to 2 π . and also that the product of two odd functions results also in an even function. that is. one period T . the product of two even functions is another even function.Chapter 6 Fourier.4 Waveforms in Trigonometric Form of Fourier Series We will now derive the trigonometric Fourier series of the most common periodic waveforms.1 Compute the trigonometric Fourier series of the square waveform of Figure 6. we can choose the limits of integration from 0 to π ⁄ 2 and multiply the integral by 4 .5 -1 c a b −a 0 b T/2 (for fundamental) 2 4 T/2 6 (for 2nd harmonic) 8 T/2 10 (for 3rd harmonic) −c 12 Figure 6. Moreover. and third harmonic test for symmetry In the expressions of the integrals in (6. Example 6. or has half−wave symmetry. it is important to remember that when using these shortcuts. we can compute the non−zero coefficients a n and b n by integrating from 0 to π only. Third Edition Copyright © Orchard Publications .5 0 -0.13. 6. and Maclaurin Series 3rd harmonic 2nd harmonic Fundamental 1 0. Taylor. The proof is based on the fact that. if the given waveform is an odd function.14). 6−12 Numerical Analysis Using MATLAB® and Excel®. This point will be illustrated in Example 6. Page 6−6. However. if the waveform has half− wave symmetry and is also an odd or an even function. we must evaluate the coefficients a n and b n for the integer values of n that will result in non−zero coefficients.2. second. and multiply the integral by 2 .14. we can choose the limits of integration as – π to + π .

Moreover.24). the terms inside the parentheses on the second line of (6.25) The b i coefficients are found from (6. 1 b n = -π ∫0 2π 1 f ( t ) sin nt dt = -π ∫0 π A sin nt dt + ∫π 2π Aπ ( – A ) sin nt dt = ----( – cos n t 0 + cos nt nπ 2π ) π AA= ----( 1 – 2 cos n π + cos 2n π ) ( – cos n π + 1 + cos 2n π – cos n π ) = ----nπ nπ (6. for brevity.24) and since n is an integer (positive or negative) or zero.( sin n π – 0 – sin n2 π + sin n π ) = ----. Square waveform for Example 6. as expected. Also. since the square waveform has odd symmetry. only odd harmonics will be present since this waveform has half−wave symmetry.( sin nt π ( – A ) cos nt dt = ----– sin nt 0 nπ 2π ) π A A . by inspection. but if we attempt to verify this using (6.12).14. all a i coefficients are zero.24) are zero and therefore. we will assume that ω = 1 . this waveform is an odd function.Waveforms in Trigonometric Form of Fourier Series T A π 0 −A 2π ωt Figure 6. To work around this problem. we will evaluate a 0 directly from (6.26) For n = even .1 Solution: The trigonometric series will consist of sine terms only because. that is.14). Third Edition Copyright © Orchard Publications 6−13 . we will get the indeterminate form 0 ⁄ 0 .(π – 0 – 2π + π) = 0 ( – A ) dt = --π (6. 1 a 0 = -π ∫0 π A dt + ∫π 2π A . we will compute all coefficients to verify this. the average ( DC ) value is zero. Then.( 2 sin n π – sin n2 π ) = ----nπ nπ (6. as we already know. Also.26) yields Numerical Analysis Using MATLAB® and Excel®. (6. However. The a i coefficients are found from 1 a n = -π ∫0 2π 1 f ( t ) cos nt dt = -π ∫0 π A cos nt dt + ∫π 2π A .

Taylor.⎛ – cos n π ) = -----⎠ 2 nπ ⎝ (6.sin ω t + -f ( t ) = -----⎠ π 5 3 π ⎝ n = odd ∑ 1 -. and multiplying the result by 4 .sin 3 ω t + -. and multiply the integral by 4 . Solution: Since the waveform is an odd function and has half−wave symmetry. We will apply this property to the following example.sin 5 ω t + …⎞ = -----.21) reduces to A-----( 1 + 2 + 1 ) = 4A b n = ----nπ nπ 4A b 1 = -----π 4A b 3 = -----3π 4A b 5 = -----5π and thus and so on.28) For n = odd .2 Compute the trigonometric Fourier series of the square waveform of Example 1 by integrating from 0 to π ⁄ 2 .sin n ω t n (6. since the square waveform has half−wave symmetry. 1 b n = 4 -π ∫0 π⁄2 4A . Example 6. and it is also an odd or an even function. For n = odd . Therefore. if the given waveform has half−wave symmetry. we can integrate from 0 to π ⁄ 2 . we are only concerned with the odd b n coefficients.( – cos n t f ( t ) sin nt dt = -----nπ π⁄2 0 4A -. Then. the trigonometric Fourier series for the square waveform with odd symmetry is 4A 1 1 4A ⎛ .+ 1⎞ . Third Edition Copyright © Orchard Publications . (6.27) It was stated above that.( 1 –2 + 1 ) = 0 b n = ----nπ as expected. (6. and Maclaurin Series A .Chapter 6 Fourier.28) becomes 6−14 Numerical Analysis Using MATLAB® and Excel®.

Also. 9. However. (6.30) becomes Numerical Analysis Using MATLAB® and Excel®.( sin nt A cos nt dt = -----nπ π⁄2 ) 0 4A --⎞ . 5. will alternate between +1 and – 1 depending For n = odd . The a n coefficients are found from 1 a n = 4 -π ∫0 π⁄2 4 f ( t ) cos nt dt = -π ∫0 π⁄2 4A . Therefore. by inspection. and thus all even harmonics are zero as expected. 13 . Thus. Example 6.3 Compute the trigonometric Fourier series of the square waveform of Figure 6. and has become an even function.( – 0 + 1 ) = 4A -----b n = -----nπ nπ (6.29) as before. it still has half−wave symmetry. and multiply the integral by 4 . the average ( DC ) value is zero.15. all a n coefficients are zero.30) We observe that for n = even . except that the ordinate has been shifted to the right by π ⁄ 2 radians. Waveform for Example 6.15. Third Edition Copyright © Orchard Publications 6−15 . the trigonometric Fourier series will consist of odd cosine terms only. and thus the series is the same as in Example 1.. 4A a n = ± -----nπ (6.31) For n = 1. Solution: This is the same waveform as in Example 6.Waveforms in Trigonometric Form of Fourier Series 4A .30) that sin n -π 2 on the odd integer assigned to n . .3 Since the waveform has half−wave symmetry and is an even function.1. and so on. we observe from (6. A T π/2 0 π 3π / 2 2π ωt −A Figure 6. it will suffice to integrate from 0 to π ⁄ 2 .⎛ sin n π = -----2⎠ nπ ⎝ (6.

3 is the same as of Example 6.33) becomes 1 1 4A . we can use the result of Example 6. 4A ⎛ 1 1 .34) and using the identities sin ( x + π ⁄ 2 ) = cos x . 7. the trigonometric Fourier series for the square waveform with even symmetry is 1 4A 4A ⎛ .sin ⎛ ωτ + -. and afterwards we want to recompute the series with reference to a different ordinate. but shifted to the right by π ⁄ 2 radians.+… ----.cos 3 ω t + -f ( t ) = -----⎠ 5 π π ⎝ 3 n = odd ∑ ( –1 ) (n – 1 ) ---------------2 1 -.cos 5 ωτ – … .cos ωτ – -f ( τ ) = -----5 3 π (6. we rewrite (6. Therefore. 11.33) and substitute ω t with ω t + π ⁄ 2 . Taylor.sin 5 ⎛ ωτ + π . 15 . and so on.sin ⎛ 3 ωτ + 3 = ---------.35) and this is the same as (6.e. With this substitution. we can use the above procedure to save time. and Maclaurin Series 4A a n = -----nπ and for n = 3. Third Edition Copyright © Orchard Publications . relation (6.sin 3 ω t + -.cos 5 ω t – …⎞ = -----.cos 3 ωτ + -.32) Alternate Solution: Since the waveform of Example 6.. 6−16 Numerical Analysis Using MATLAB® and Excel®.34) as 1 1 4A .1.Chapter 6 Fourier. and so on.1.27). that is. sin ( x + 3 π ⁄ 2 ) = – cos x .+ -. it becomes – 4A a n = ---------nπ Then.sin ⎛ ωτ + π --⎞ + … --⎞ + ---⎞ + -f ( τ ) = -----⎝ ⎝ ⎠ ⎝ ⎠ 5 3 π 2⎠ 2 2 4A π⎞ 1 π⎞ π⎞ 1 .sin 3 ⎛ ωτ + π .cos n ω t n (6. i. if we compute the trigonometric Fourier series with reference to one ordinate.sin 5 ω t + …⎞ f ( t ) = -----⎝ ⎠ π 3 5 (6.sin ω t + -.+ -⎝ π 2⎠ 2⎠ 5 ⎝ 2⎠ 3 ⎝ (6. we let ω t = ωτ + π ⁄ 2 .cos ω t – 1 -.sin ⎛ 5 ωτ + 5 .

1 .cos ax 2 a a (6. we will assume that ω = 1 . As before.4 Compute the trigonometric Fourier series of the sawtooth waveform of Figure 6. we can choose the limits from – π to + π .t – 2A ⎩π 0<t<π π < t < 2π However. we only need to evaluate the b n coefficients.Waveforms in Trigonometric Form of Fourier Series Example 6. By inspection. Accordingly. the DC component is zero.sin a x – x = ----. this is what we will do. Sawtooth waveform Solution: This waveform is an odd function but has no half−wave symmetry. since the waveform is an odd function. If we choose the limits of integration from 0 to 2 π we will need to perform two integrations since A ⎧ ---t π ⎪ f(t) = ⎨ ⎪A --.16. From tables of integrals. and multiply the integral by 2 . we can integrate from 0 to π . and thus we will only need one integration since A -t f ( t ) = --π –π < t < π Better yet. ∫ x sin ax dx Then.36) Numerical Analysis Using MATLAB® and Excel®. T A − 2π −π π 0 −A 2π ωt Figure 6.16. Third Edition Copyright © Orchard Publications 6−17 . therefore. it contains sine terms only with both odd and even harmonics.

17. Then. 2.37) reduces to 2A . sin n π = 0 .( sin n π – n π cos n π ) = ---------n2 π2 We observe that: 1. (6.⎛ sin ω t – 1 f ( t ) = -----⎠ π 4 3 2 π ⎝ ∑ ( –1 ) n–1 1 -. Triangular waveform for Example 6.sin 4 ω t + …⎞ = 2A -. Then. 2A .cos nt⎞ . cos n π = – 1 .sin 3 ω t – 1 -. A −2π −π −A T ωt 0 π/2 π 2π Figure 6. we only need to evaluate the b n coefficients.5 Solution: This waveform is an odd function and has half−wave symmetry. then. the even harmonics have negative coefficients. and Maclaurin Series 2 b n = -π ∫0 π A --.37) 2A .( – n π ) = – 2A -----b n = ---------n2 π2 nπ that is. If n = even . Third Edition Copyright © Orchard Publications . Assume ω = 1 . Taylor.sin 2 ω t + 1 . sin n π = 0 and cos n π = 1 . We will choose the limits of integration from 0 to π ⁄ 2 . the odd harmonics have positive coefficients. and will multiply the 6−18 Numerical Analysis Using MATLAB® and Excel®.( n π ) = 2A b n = --------------n2 π2 nπ that is. the trigonometric Fourier series will contain sine terms only with odd harmonics.5 Find the trigonometric Fourier series of the triangular waveform of Figure 6.( sin nt – nt cos nt ) = ---------n2 π2 π 0 2A .sin n ω t n (6.38) Example 6.⎛ ---t sin nt dt = -----sin nt – -⎠ n π2 ⎝ n2 π 0 (6. Accordingly.Chapter 6 Fourier. If n = odd .17. the trigonometric Fourier series for the sawtooth waveform with odd symmetry is 2A ------.t sin nt dt = 2A -----π π2 ∫0 π 1t 2A . Thus.

39) π⁄2 0 ∫0 π⁄2 2A -----.sin 7 ω t + …⎞ = -----. Third Edition Copyright © Orchard Publications 6−19 . Assume ω = 1 .18.Waveforms in Trigonometric Form of Fourier Series integral by 4 .sin 5 ω t – ----. 11. b = ---------n ⎪ 2 2 n π ⎪ π . … then. 7.⎛ ---t sin nt dt = -----sin nt – -2 ⎝ 2 ⎠ n n π 8A = ---------( sin nt – nt cos nt ) 2 2 n π π⎞ π π 8A ⎛ . Solution: The waveform for this example is shown in Figure 6.= ⎨ sin n -2 8A ⎪ – 1 for n = 3. ∫ x sin ax dx Then.– n -= ---------sin n -2 2 ⎝ 2⎠ 2 2 n π (6. 4 b n = -π 1x = --sin a x – -. 9.cos nt⎞ .sin ω t – 1 -. Numerical Analysis Using MATLAB® and Excel®. … then.t sin nt dt = 8A -----2 π π π⁄2 0 ∫0 1t 8A .40) We are only interested in the odd integers of n . From tables of integrals. By inspection.41) 2 n Example 6.6 A half−wave rectification waveform is defined as ⎧ sin ω t f (t) = ⎨ ⎩ 0 0 < ωt < π π < ωt < 2π (6. the trigonometric Fourier series for the triangular waveform with odd symmetry is 8A 1 1 8A ⎛ .sin n ω t (6. the sine term yields 8A⎧ 1 for n = 1.= 0 cos n -2 For odd integers of n . and we observe that: π . b = – ---------n ⎪ 2 2 n π ⎩ Thus.cos n -.sin 3 ω t + ----( –1 ) f ( t ) = -----2 ⎝ 2 ⎠ 49 25 9 π n = odd π ∑ (n – 1) ---------------2 1 ---. 5.42) Express f ( t ) as a trigonometric Fourier series. the DC component is zero.cos ax 2 a a π⁄2 (6.

⎧ cos n π cos n π 2 ⎫ A ⎧ – cos n π – cos n π 2 ⎫ .⎬ = – ----2π ⎩ 1–n 1+n 1–n n+1 ⎭ (6.18. Therefore. cos ( m + n ) x cos ( m – n ) x – -----------------------------. and from tables of integrals A a n = --π ∫0 π A sin t cos nt dt + --π ∫π 2π 0 cos nt dt ∫ ( sin mx ) ( cos nx ) dx Then. we expect all terms to be present.6 By inspection.Chapter 6 Fourier.– ----------.--------------------------.( m2 ≠ n2 ) = – -----------------------------2( m + n) 2(m – n) π ⎧ A cos ( 1 + n )t cos ( 1 – n ) t + ---------------------------. and Maclaurin Series –2 π –π 0 π 2π 3π Figure 6.+ -----------------.+ ------------= ----n≠1 2 2 2⎠ ⎝ π 2π ⎝ (1 – n ) ⎠ 1–n 1–n (6. Taylor.+ -------------. f ( t ) for Example 6.+ ----------. the average is a non−zero value.⎨ -----------------.⎛ -----------------------.–1 a n = --2 π ⎨ 1+n 1–n ⎩ ⎫ ⎬ 0⎭ A ⎧ cos ( π – n π ) cos ( π + n π ) 1 1 ⎫ . Third Edition Copyright © Orchard Publications . by substitution into (6. A.--------------------------------------------------------------------------------------.⎨ ---------------------------.– ------------a n = – ----⎬ = ----⎨ -------------2 2⎬ 2 π 1+n 2π ⎩ 1 – n 1+n 1–n ⎭ 1–n ⎭ ⎩ 1–n cos n π + 1 ⎞ A ⎛ cos n π + n cos n π + cos n π – n cos n π 2 -⎞ = A --.44) 6−20 Numerical Analysis Using MATLAB® and Excel®. and the waveform has neither odd nor even symmetry.+ ----------------------------.43) Using the trigonometric identities and we obtain and cos ( x – y ) = cos x cos y + sin xsiny cos ( x + y ) = cos x cos y – sin x sin y cos ( π – n π ) = cos π cos n π + sin π sin n π = – cos n π cos ( π + n π ) = cos π cos n π – sin π sin n π = – cos n π Then.43). The a n coefficients are found from 1 a n = -π ∫0 2π f ( t ) cos nt dt For this example.+ ------------.

However.= 0 2 π(1 – 3 ) (6. we get cos 2 π + 1 A . we get 2AA ( cos 4 π + 1 ) = – -------a 4 = --------------------------------2 15 π π(1 – 4 ) 2AA ( cos 6 π + 1 ) = – -------a 6 = --------------------------------2 35 π π(1 – 6 ) 2AA ( cos 8 π + 1 ) = – -------a 8 = --------------------------------2 63 π π(1 – 8 ) (6. 1 b n = A -π ∫0 2π A f ( t ) sin nt dt = --π ∫0 π A sin t sin nt dt + --π ∫π 2π 0 sin nt dt Numerical Analysis Using MATLAB® and Excel®. First.( sin t ) 2 a 1 = ----2π π (6. therefore. from (6.44) with n = 2. we will evaluate the integral A a 1 = --π ∫0 sin t cos t dt 12 ( sin ax ) = ----2a = 0 0 π From tables of integrals.50) (6. we will evaluate a 0 to obtain the DC value. Third Edition Copyright © Orchard Publications 6−21 . the DC value is 1 -.Waveforms in Trigonometric Form of Fourier Series Next.51) and so on. ∫ ( sin ax ) ( cos ax ) dx A . By substitution of n = 0 .44).48) We see that for odd integers of n . a n = 0 . For this example. 4.44) to obtain the value of a 1 . 5. Now. … .45) We cannot use (6.a0 = A --2 π (6. except a 1 . we can evaluate all the a n coefficients. for n = even . we need to evaluate the b n coefficients. we get a 0 = 2A ⁄ π Therefore.47) (6.46) From (6.⎞ = – 2A -----a 2 = --⎝ π ( 1 – 22 ) ⎠ 3π A ( cos 3 π + 1 ) a 3 = --------------------------------. and thus.49) (6. 3.⎛ -----------------------.

+ … .– ---------------------------.45) and (6.19 where the ordinate was arbitrarily chosen as shown. We will find b 1 by direct substitution into (6.sin t – --f ( t ) = --.– -------------------------b n = --π 2⎨ 1+n 1–n ⎩ ⎫ ⎬ 0⎭ A sin ( 1 – n )π sin ( 1 + n )π . 7 0 . except b 1 . 8 0 .------------.53) Example 6. A b 1 = --π ∫0 π A 2 t sin 2t . we find that the trigonometric Fourier series for the half−wave rectification waveform with no symmetry is A cos 2t cos 4t cos 6t cos 8t A A .+ ------------. Full−wave rectified waveform with even symmetry 6−22 Numerical Analysis Using MATLAB® and Excel®.– -----------π 2 4 π 0 A sin 2 π A . Third Edition Copyright © Orchard Publications . 9 0 .-------------------------.47) through (6.– ----------------------------2(m + n) 2( m – n) π A 1 ⎧ sin ( 1 – n ) t sin ( 1 + n ) t .= --π 2 4 2 (6. Assume ω = 1 .7 A full−wave rectification waveform is defined as f ( t ) = A sin ω t (6.( m2 ≠ n2 ) = ----------------------------.--------------------------= ----. 6 A 0 .54) Express f ( t ) as a trigonometric Fourier series.+ ------------. 5 0 .14) for n = 1 .52).π = ---.19. 2 0 .– 0 + 0 = 0 ( n ≠ 1 ) 2π 1–n 1+n that is.+ ------------. 4 0 . 1 0 . 3 0 . 1 0 –2 π 0 2 –π 4 6 0 8 π 1 0 1 2 2π Figure 6.Chapter 6 Fourier.52) Combining (6. are zero. sin ( m – n ) x sin ( m + n ) x .+ --π 2 63 3 15 35 π (6.-( sin t ) dt = --. Thus.⋅ -. and Maclaurin Series and from tables of integrals.– ------------. all the b n coefficients. ∫ ( sin mx ) ( sin nx ) dx Therefore. Taylor. Solution: The waveform is shown in Figure 6.

– ----------------------------. The a n coefficients are found from 1 a n = -π ∫0 2π f ( t ) cos nt dt where for this example. from (6.– . we will evaluate a 0 to Numerical Analysis Using MATLAB® and Excel®. except a 1 .+ ------------------------------------π n+1 n–1 (6.55) and from tables of integrals.⎨ ---------------------------⋅1 – --------------------------a n = -----π 2⎩ n–1 n+1 ⎫ ⎬ 0⎭ A ⎧ cos ( n – 1 )π cos ( n + 1 )π 1 1 ⎫ . Therefore.57) Now. (6.1 = ---------------------------------------. the average is a non−zero value. 1 a n = -π ∫ 2A A sin t cos nt dt = -----π –π π ∫0 π sin t cos nt dt (6. ∫ ( sin mx ) ( cos nx ) dx Since we express (6.---------------------------.= --------------------------–1 a n = --2 π π n–1 n+1 n –1 – 2A ( cos n π + 1 ) = ---------------------------------------n≠1 2 π(n – 1) (6.– -----------------------------2(m + n) 2(n – m) cos ( x – y ) = cos ( y – x ) = cos x cos y + sin xsiny π ⎧ cos ( n – 1 ) t cos ( n + 1 ) t 2A -. we observe that the waveform has even symmetry. First. We choose the period of the input sinusoid so that the output will be expressed in terms of the fundamental frequency. we expect only cosine terms to be present.Waveforms in Trigonometric Form of Fourier Series By inspection.57).– ----------.56) simplifies to 2 + ( n – 1 ) cos n π – ( n + 1 ) cos n π A A + cos n π + cos n π . we can evaluate all the a n coefficients.1 -----------------------------------------------------------------------------------------------------------.56) To simplify the last expression in (6.⎬ = --π⎨ n – 1 n + 1 n – 1 n + 1 ⎭ ⎩ A – cos ( n π + π ) cos ( n π – π ) – 1 . We also choose the limits of integration as – π and + π . we make use of the trigonometric identities and cos ( n π + π ) = cos n π cos π – sin n π sin π = – cos n π cos ( n π – π ) = cos n π cos π + sin n π sin π = – cos n π Then. Third Edition Copyright © Orchard Publications 6−23 .55) as cos ( m – n ) x cos ( m + n ) x .( m2 ≠ n2 ) = -----------------------------.56).– ----------.

( sin ax ) 2 = ----2a From tables of integrals.– -----f ( t ) = 2A π ⎩ 3 63 35 15 π ⎭ (6.= – -------2 15 π π(4 – 1) – 2A ( cos 6 π + 1 ) 4A a 6 = ---------------------------------------.= – 4A 2 3π π(2 – 1) 4A– 2A ( cos 4 π + 1 ) a 4 = ---------------------------------------.Chapter 6 Fourier.60) (6.60) through (6.61) (6. and Maclaurin Series obtain the DC value.64) Therefore.63) and so on. we get -----a 0 = 4A π Therefore. By substitution of n = 0 .+ ----------------.= – -------2 35 π π(6 – 1) – 2A ( cos 8 π + 1 ) 4A a 8 = ---------------------------------------. from (6.58) and (6.62) (6. ∫ ( sin ax ) ( cos ax ) dx and thus. a n = 0 .58) From (6. combining the terms of (6.57) we observe that for all n = odd . other than n = 1 . Third Edition Copyright © Orchard Publications . To obtain the value of a 1 .59) For n = even .= – -------2 63 π π(8 – 1) (6.⎨ ----------------. the trigonometric form of the Fourier series for the full−wave rectification waveform with even symmetry is 6−24 Numerical Analysis Using MATLAB® and Excel®.63) we get ⎫ 4A ⎧ cos 2 ω t cos 4 ω t cos 6 ω t cos 8 ω t .a = 2A -----2 0 π (6. 12 ( sin t ) a 1 = ----2π π = 0 0 (6.+ ----------------. Then. we must evaluate the integral 1 a 1 = -π π ∫0 sin t cos t dt 1 . Taylor.+ …⎬ . the DC value is 1 -.57) we get 2A ( cos 2 π + 1 ) -----a2 = – ---------------------------------------.+ ---------------------.

and the sum is combined to a single term.5 Alternate Forms of the Trigonometric Fourier Series We recall that the trigonometric Fourier series is expressed as 1 .cos n ω t 2 n ( – 1 ) n = 2. we still need to compute the a n and b n coefficients separately.66) as Numerical Analysis Using MATLAB® and Excel®. 2.– -----π π 1 -----------------.20 for the derivation of the alternate forms. either cosine or sine.65) This series of (6. it is very unlikely that a Fourier series will consist of even harmonic terms only. This is because we chose the period to be from – π and + π . we rewrite (6. Derivation of the alternate form of the trigonometric Fourier series cos θ n = sin ϕ n We assume ω = 1 . ϕn cn = bn an + bn bn bn .a + a 1 cos ω t + a 2 cos 2 ω t + a 3 cos 3 ω t + a 4 cos 4 ω t + … f ( t ) = -2 0 + b 1 sin ω t + b 2 sin 2 ω t + b 3 sin 3 ω t + b 4 sin 4 ω t + … (6. Third Edition Copyright © Orchard Publications 6−25 . … . 3. the period is defined as the shortest period of repetition. and for n = 1. In any waveform where the period is chosen appropriately. it may be advantageous of using the alternate form of the trigonometric Fourier series where the cosine and sine terms of the same frequency are grouped together. 6. 4. Generally. 6.65) shows that there is no component of the fundamental frequency.= ----sin θ n = --------------------cn an + bn cn θn an an an . … ∑ ∞ (6.Alternate Forms of the Trigonometric Fourier Series 2A 4A f ( t ) = -----.66) If a given waveform does not have any kind of symmetry.20. We use the triangle shown in Figure 6.= ----cos θ n = --------------------cn an + bn an bn ϕ n = atan ----θ n = atan ----bn an Figure 6. However.

1 --a + 2 0 n=1 ∑ ∞ b n⎞ 1 -a + c n cos ⎛ n ω t – atan ---. Third Edition Copyright © Orchard Publications . Taylor.21. Since it is customary to use the cosine function in the time domain to phasor transformation.sin t⎞ + c 2 ⎛ ---.⇔ -⎝ ⎠ 2 0 an n=1 ∑ cn ∠– atan ---an ∞ bn (6.68) can be expressed as phasors.cos nt + ---.67) and (6. where ω ≠ 1 .a0 + c1 ⎛ f ( t ) = -2 ⎝ ⎠ sin ( t + ϕ 1 ) ⎧ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎩ c2 ⎛ ⎝ ⎧ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎩ sin ( 2t + ϕ 2 ) ⎧ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎩ sin ( nt + ϕ n ) an sin ϕ 2 cos 2t + cos ϕ 2 sin 2t ⎞ + … + cn ⎛ ⎠ ⎝ sin ϕ n cos nt + cos ϕ n sin nt ⎞ ⎠ and. we get 1 -a + f ( t ) = -2 0 ⎧ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎩ cos ( nt – θ n ) ∞ cos θ n cos nt + sin θ n sin nt ⎞ ⎠ n=1 ∑ 1 -a + c n cos ( n ω t – θ n ) = -2 0 n=1 n ω t – atan ----⎞ ∑ cn cos ⎛ ⎝ a n⎠ ∞ (6.68) The series of (6. for ω ≠ 1 .69) Example 6.sin 2t⎞ + … f ( t ) = -cos t + ---cos 2t + ---⎝ c1 ⎠ ⎝ c2 ⎠ 2 c1 c2 bn an .a + c1 ⎛ = -2 0 ⎝ + cn ⎛ ⎝ ⎧ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎩ cos ( t – θ 1 ) ⎧ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎩ cos ( 2t – θ 2 ) bn cos θ 1 cos t + sin θ 1 sin t ⎞ + c2 ⎛ ⎠ ⎝ cos θ 2 cos 2t + sin θ 2 sin 2t ⎞ +… ⎠ and.a 0 + c 1 ⎛ ---. we get 1 -a + f ( t ) = -2 0 n=1 ∑ ∞ 1 -a + c n sin ( n ω t + ϕ n ) = -2 0 n=1 n ω t + atan ----⎞ ∑ cn sin ⎛ ⎝ b n⎠ ∞ (6. in general. in general. 6−26 Numerical Analysis Using MATLAB® and Excel®.8 Find the first 5 terms of the alternate form of the trigonometric Fourier series for the waveform of Figure 6. and Maclaurin Series a1 a2 b1 b2 1 .63) below. we choose to use the transformation of (6. sin ϕ 1 cos t + cos ϕ 1 sin t 1 ⎞ .67) Similarly.sin nt⎞ + c n ⎛ ---⎝ cn ⎠ cn 1 .Chapter 6 Fourier.

and we will combine them to get an expression in the form of (6.+ 2 π – -. by inspection the DC value is not zero.+ ----.72) The DC value is 1 1 -. 1 a n = -π ∫0 π⁄2 1 ( 3 ) cos nt dt + -π ∫ 2π 3 . We will compute the a n and b n coefficients.a = ----2 0 2π 1 = ----2π ∫0 π⁄2 1 ( 3 ) dt + ----2π ∫ π ⁄ 2 ( 1 ) dt 2π 1 . 2 a 1 = -π 2a 3 = – ----3π (6.21.sin nt + ----nπ 2π π⁄2 π 2 π 1 π 1 3 . Waveform for Example 6.71) and (6. For n = odd .8 The given waveform has no symmetry.sin nt ( 1 ) cos nt dt = ----n π π⁄2 π⁄2 0 1 .sin n2 π – ----.sin n -.( π + 2 π ) = -⎝2 ⎠ 2 π 2 2 (6. we expect both cosine and sine functions with odd and even terms present.( 3t = ----2π π⁄2 0 +t 2π ) π⁄2 1 3 π π⎞ ----⎛3 ----.Alternate Forms of the Trigonometric Fourier Series f( t) 3 2 1 ω = 1 t π/2 π 3π/2 2π Solution: Figure 6. Also.sin n -= ----2 2 nπ nπ 2 nπ nπ (6.63). the DC value. Then.70) We observe that for n = even . Third Edition Copyright © Orchard Publications 6−27 .sin n -.= ----. a n = 0 .= . thus.73) The b n coefficients are Numerical Analysis Using MATLAB® and Excel®.

cos nt + ----nπ 2π π⁄2 Then. 2.+ ⎛ -.+ ----.cos nt ( 1 ) sin nt dt = ----nπ π⁄2 π⁄2 0 –1 .= --------a 3 – jb 3 = – ----. we get: 2 2 -a 1 – jb 1 = --–j.= ----.81) (6.76) (6.= c n ∠– atan ---an a n + b n ∠– θ n = a n – jb n (6.cos ( 3 ω t – 135 ° ) .cos ( 2 ω t – 90 ° ) .80) 1 1 1 .Chapter 6 Fourier.∠– 45 ° ⎝ π⎠ ⎝ π⎠ 2 2 Similarly. 3.– j ----3π 3π 3π 3π 11 1.69).79) Thus.75) (6. Third Edition Copyright © Orchard Publications .cos n2 π + -----.∠– 135 ° ⇔ --------. and Maclaurin Series 1 b n = -π ∫0 π⁄2 1 ( 3 ) sin nt dt + -π ∫ 2π –3 . 2 2 2 8 --------.∠– 45 ° = --------π π 2 π (6. we find that the alternate form of the trigonometric Fourier series representing the waveform of this example is 6−28 Numerical Analysis Using MATLAB® and Excel®.80) through (6.= -a 2 – jb 2 = 0 – j -π π π 2 2 2 2 2 2 .82) and (6.cos ( ω t – 45 ° ) .cos n π = ----( 3 – cos n2 π ) = ----cos n -2 nπ nπ 2 nπ nπ nπ nπ b1 = 2 ⁄ π b2 = 1 ⁄ π b3 = 2 ⁄ 3 π b4 = 1 ⁄ 2 π (6. for n = 1.= π π = 2⎞ 2⎞ ⎛ -.78) ∑ cn ∠– atan ---an ∞ From (6. 1 --a + 2 0 where n=1 ∑ ∞ b n⎞ 1 -a + c n cos ⎛ n ω t – atan ---.83).⇔ -⎝ ⎠ 2 0 an bn 2 2 a n + b n ∠– atan ---.73) and (6. Taylor.∠– 90 ° ⇔ -. 1π 1–1 –3 23. and 4 .∠– 45 ° ⇔ 2 ----.∠– 90 ° ⇔ ----a 4 – jb 4 = 0 – j ----2π 2π 2π (6.+ ----.74) (6.83) Combining the terms of (6.= an 2 2 bn n=1 bn .77) (6.----cos ( 4 ω t – 90 ° ) = .

+ ---. we get a 2 b 2 ⎞ – j2 ω t ⎛ a 1 b 1 ⎞ – j ω t 1 a 1 b 1⎞ j ω t ⎛ a 2 b 2⎞ j2 ω t . The advantage of the exponential form is that we only need to perform one integration rather than two. 1 e +e e +e .87) and grouping terms with same exponents.+ -f(t) = 3 2 π [2 2 cos ( ω t – 45 ° ) + cos ( 2 ω t – 90 ° ) 1 2 2 .86) into f ( t ) .= -2 n 2⎝ n j ⎠ b n⎞ 1 1⎛ . Moreover.e .89) (6.a + ---C n = -.( a + jb n ) .= -2 n n 2⎝ n j ⎠ 1 -a C 0 = -2 0 (6.90) (6. Third Edition Copyright © Orchard Publications 6−29 .a – ---C – n = -. Thus.e + ---.– ---+ -⎝ 2 j2 ⎠ ⎝ 2 j2 ⎠ ⎝ 2 j2⎠ ⎝ 2 j2⎠ 2 (6.cos ( 3 ω t – 135 ° ) + -+ --------2 3 ] (6. in most cases the integration is simpler.( a –j b ) .88) The terms of (6.91) Numerical Analysis Using MATLAB® and Excel®.6 The Exponential Form of the Fourier Series The Fourier series are often expressed in exponential form.The Exponential Form of the Fourier Series 1 -.88) in parentheses are usually denoted as b n⎞ 1 1⎛ . one for the a n .– ---. and another for the b n coefficients in the trigonometric form of the series. The exponential form is derived from the trigonometric form by substitution of +e --------------------------cos ω t = e 2 jωt –j ω t (6.a 0 + ⎛ ---f ( t ) = … + ⎛ --.a + a 1 ⎛ --------------------------.cos ( 4 ω t – 90 ° ) + … .⎞ + a 2 ⎛ --------------------------------⎞ + f ( t ) = -⎝ ⎠ ⎝ ⎠ 2 0 2 2 jωt –j ω t j2 ω t – j2 ω t e – e -⎞ e –e -⎞ + … + b 2 ⎛ ------------------------------… + b 1 ⎛ -------------------------⎝ ⎠ ⎝ ⎠ j2 j2 jωt –j ω t j2 ω t – j2 ω t (6.e + ---.84) 6.e + --.85) and e –e sin ω t = -------------------------j2 jωt –j ω t (6.

that is.97) are zero except the last. Third Edition Copyright © Orchard Publications .( a – jb n + a n + jb n ) C n + C – n = -2 n 6−30 Numerical Analysis Using MATLAB® and Excel®. for ω ≠ 1 .93) We can derive a general expression for the complex coefficients C n . Therefore. Then. (6.96) We can derive the trigonometric Fourier series from the exponential series by addition and subtraction of the exponential form coefficients C n and C –n . by multiplying both sides of (6. as we did in the derivation of the a n and b n 2π 2π 2π 2π coefficients of the trigonometric form.Chapter 6 Fourier. in general. ∫0 2π f(t )e – jnt dt = … + + + ∫0 C –2 e C0 e C2 e – j2t – jnt e dt + 2π ∫0 C–1 e e dt – jt – jnt dt (6. Thus. 1 C n = ----2π ∫0 ∫0 2π f(t)e – jn ω t d( ω t ) (6.94) ∫0 ∫0 – jnt dt + ∫0 C1 e e jt – jnt 2π j2t – jnt e dt + … + ∫0 ∫0 Cn e jnt – jnt e dt We observe that all the integrals on the right side of (6. and Maclaurin Series Then. Taylor. C – n = C n∗ (6. from (6.88) is written as f ( t ) = … + C –2 e – j2 ω t + C –1 e –j ω t + C0 + C1 e jωt + C2 e j2 ω t +… (6.89) and (6.90).92) by e – jn ω t and integrating over one period. ∫0 or 2π f(t )e – jnt dt = ∫0 2π Cn e jnt – jnt e dt = 2π C n dt = 2 π C n 1C n = ----2π ∫0 2π f(t )e – jnt dt and.95) or 1 C n = -T T f(t)e – jn ω t d( ω t ) (6. except C 0 . 1 .92) We must remember that the C i coefficients. are complex and occur in complex conjugate pairs. with ω = 1 .

98) Similarly. in (6. all coefficients C i are imaginary Since odd functions have no cosine terms. 4. For odd functions. both C –n and C n are real. the b n coefficients in (6.89) and (6. 3. all even harmonics are zero.The Exponential Form of the Fourier Series or an = Cn + C–n (6. Therefore. Therefore. C –n = C n∗ always This can be seen in (6.= -C – n = -⎝ ⎠ 2 n 2 j (6.100) and (6.( a –j b ) . all coefficients C i are real We recall from (6.100) and b n⎞ 1 1⎛ .= -C n = -⎝ ⎠ 2 n n 2 j (6. Therefore.a n – ---. f ( t ) is complex. If there is no symmetry. For even functions. 1 .a n + ---. and thus. If there is half−wave symmetry.101) are zero.( a + jb n ) .101) Since even functions have no sine terms.101) are zero.101) Numerical Analysis Using MATLAB® and Excel®. 5.97) (6.100) and (6.101) the coefficients a n and b n are both zero for n = even . the a n coefficients in (6. Third Edition Copyright © Orchard Publications 6−31 . C n = 0 for n = even We recall from the trigonometric Fourier series that if there is half−wave symmetry.( a – jb – a – j b ) C n – C – n = -2 n n n n or bn = j ( Cn – C–n ) (6. 2. both C –n and C n are imaginary.90) that b n⎞ 1 1⎛ .100) and (6. both C –n and C n are also zero for n = even .100) and (6.99) Symmetry in Exponential Series 1.

For n ≠ 0 . and its DC component is zero. T A π 0 2π ωt −A Figure 6. Assume that ω = 1 . and C 0 = 0 .( 1 + e – jn2 π – 2e – jn π ) = ----------= ----------2j π n 2j π n (6.9 Solution: This is the same waveform as in Example 6.22 below. and as we know.22. we get 1C n = ----2π ∫0 2π f(t)e – jnt 1dt = ----2π ∫0 π Ae – jnt 1dt + ----2π ∫π 2π –A e – jnt dt and for n = 0 . it is an odd function. n = even Cn 2 A A . C n = 0 for n = even .Chapter 6 Fourier. then. Waveform for Example 6. Third Edition Copyright © Orchard Publications .e – jnt + ------– jn 2π π A – jn2 π – jn π A 1 A –jn π -(e .102) For n = even . e – jn π = 1 .( e – jn π – 1 ) = ----------. and Maclaurin Series Example 6.103) 6−32 Numerical Analysis Using MATLAB® and Excel®. has half− wave symmetry. Therefore. 1 C 0 = ----2π ∫0 π Ae dt + –0 as expected. Using (6.1. the C n coefficients will be imaginary.( e –jn π – 1 ) .( 1 – e – jn π + e – jn2 π – e – jn π ) . Taylor.-------(e = ----– 1 ) + ---–e ) = ----------jn 2j π n 2 π – jn 2 A A . 1C n = ----2π ∫π 2π A –0 .95) with ω = 1 .------e dt = ----2 π – jn π 0 –A .– jnt 1.( 1 – 1 )2 = 0 = ----------2j π n 2j π n (6.9 Compute the exponential Fourier series for the square waveform of Figure 6.(π – 2π + π) = 0 ( – A ) e dt = ----2π ∫0 π Ae – jnt dt + ∫π 2π –A e – jnt A.

For n = odd .22) are the same. e – jn π = – 1 . it is useful to plot the amplitudes of the harmonics on a frequency scale that shows the first (fundamental frequency) harmonic.105) The minus (−) sign of the first two terms within the parentheses results from the fact that C –n = C n∗ . Line spectrum for square waveform of Example 6. 2 A .1. Numerical Analysis Using MATLAB® and Excel®.7 Line Spectra When the Fourier series are known. Figure 6. We observe that f ( t ) is purely imaginary. we group the two terms for which n = 3 . as expected. since the waveform is an odd function.23 shows the line spectrum of the square waveform of Example 6.Line Spectra as expected.23. and this will produce the third harmonic sin 3 ω t . and so on.105) and (6. and the higher harmonics times the amplitude of the fundamental.– jn π AA2 2 2A ( – 1 – 1 ) = ----------( – 2 ) = ------= ----------(e – 1 ) = ----------2j π n 2j π n 2j π n jπn n = odd Cn (6. bn 4/π 0 1 3 5 7 9 nωt Figure 6. that is.1 * An instrument that displays the spectral lines of a waveform.104) Using (6. Third Edition Copyright © Orchard Publications 6−33 .… – --e f ( t ) = -----–e +e + -3 jπ ⎝ 3 j3 ω t ⎞ ⎠ 2A = -----jπ n = odd ∑ 1 --e n jn ω t (6. f ( t ) = … + C–2 e – j2 ω t + C–1 e –j ω t + C0 + C1 e jωt + C2 e j2 ω t +… we obtain the exponential Fourier series for the square waveform with odd symmetry as 1 – j3 ω t – j ω t 2A ⎛ jωt 1 -e . it follows that C –3 = C 3∗ = – 2A ⁄ j3 π . this will produce the fundamental frequency sin ω t . Such a plot is known as line spectrum and shows the spectral lines that would be displayed by a spectrum analyzer*. Then. Therefore. 6.105) for which n = 1 . For instance. we group the two terms inside the parentheses of (6. To prove that (6. since C 3 = 2A ⁄ j3 π .92).

Waveform for Example 6.106) The value of the average ( DC component) is found by letting n = 0 .10 Compute the exponential Fourier series for the waveform of Figure 6.25. the components of the exponential Fourier series are found from 1C n = ----2π Figure 6. and Maclaurin Series Figure 6. As shown in Figure 6.106) we get 6−34 Numerical Analysis Using MATLAB® and Excel®.6.6 The line spectra of other waveforms can be easily constructed from the Fourier series. Assume ω = 1 . To determine how faithfully such pulses will be transmitted.Chapter 6 Fourier. Third Edition Copyright © Orchard Publications .24 shows the line spectrum for the half−wave rectification waveform of Example 6. Thus. Line spectrum for half−wave rectifier of Example 6.25. k is the ratio of the pulse repetition time to the duration of each pulse. it is necessary to know the frequency components. For this example. A T T /κ −2π −π −π/κ 0 π/κ π 2π ωt Solution: This recurrent rectangular pulse is used extensively in digital communications systems.11 ∫– π π Ae – jnt Adt = ----2π ∫– π ⁄ k e π⁄k – jnt dt (6. Taylor. Then. and plot its line spectra. from (6.25.24. the pulse duration is T ⁄ k . is k times the pulse duration. In other words. the recurrence interval (period) T . A/ 2 A/π DC 2 4 6 8 nωt 0 1 Figure 6. Example 6.

e –e -⎞ .2]) 1. integration of (6.4 -4 -3 -2 -1 0 1 2 3 4 K=2 Figure 6.106) yields A .6 0.4 1.26.2 0 -0./(2.[−4 4 −0.*x)'.107) For the values for n ≠ 0 .109) The relation of (6. Line spectrum of (6.Line Spectra At C 0 = ----2π π⁄k –π ⁄ k A.⋅ ------------------------k nπ ⁄ k (6./(10.26 through 6./(5.[−4 4 −0.*x).2]) fplot('sin(10. sin ( n π ⁄ k ) A sin ( n π ⁄ k ) .+ -.⎛ π π ⎞ A = -----.108) f(t) = ∑ n = –∞ ∞ A sin ( n π ⁄ k ) --.*x). for k = 2 .*x)'.28. k = 5 and k = 10 respectively by using the MATLAB scripts below.– jnt e C n = -------------– jn2 π π⁄k –π ⁄ k nπ AA.*x).109) has the sin x ⁄ x form.4 1.= ----= ----⋅ sin ⎛ ----⋅ -----------------------------------⎝ k⎠ nπ nπ j2 jn π ⁄ k – jn π ⁄ k and thus.4 0.*x)'. Third Edition Copyright © Orchard Publications 6−35 .= --.109) for k = 2 Numerical Analysis Using MATLAB® and Excel®.2]) fplot('sin(5.4 1.⋅ ------------------------= A ------------------------nπ k nπ ⁄ k (6.8 0. fplot('sin(2. and the line spectrum is shown in Figures 6.= --2π⎝ k k ⎠ k (6.2 -0.2 1 0.[−4 4 −0.

Chapter 6 Fourier.27. Third Edition Copyright © Orchard Publications . it is necessary to construct the Fourier expansion of a function based on observed values instead of an analytic expression.6 0. 6.8 0.8 0. Taylor. Examples are meteorological or economic quantities 6−36 Numerical Analysis Using MATLAB® and Excel®.112) for k = 10 The spectral lines are separated by the distance 1 ⁄ k and thus.2 0 -0.2 1 0.4 -4 -3 -2 -1 0 1 2 3 4 K=10 Figure 6.8 Numerical Evaluation of Fourier Coefficients Quite often. Line spectrum of (6. Line spectrum of (6. the lines get closer together while the lines are further apart as k gets smaller.2 -0. as k gets larger.6 0.109) for k = 5 1.28.2 -0. and Maclaurin Series 1.4 0.2 1 0.4 0.4 -4 -3 -2 -1 0 1 2 3 4 K=5 Figure 6.2 0 -0.

using a spreadsheet. we multiply degrees by π (3.29.1459. y sin 2x . y sin 3x .. To compute the coefficients of the higher order harmonics. f(x) x Figure 6. the more pulses we use. a month or even a year. we form the products y cos 2x . In Columns D and E we enter the values of cos x and the product y cos x respectively. Waveform whose analytical expression is unknown Since the arguments of the sine and the cosine are in radians. and enter these values in Column A of the spreadsheet. Numerical Analysis Using MATLAB® and Excel®.31 is a partial table showing the computation of the coefficients of a clipped sine waveform. Third Edition Copyright © Orchard Publications 6−37 . will work for both the waveforms that have an analytical solution and those that do not.5 ° and it is convenient to start at the zero point of the waveform. Then. and Figure 6. and we divide by π to obtain the coefficients a 1 and b 1 . we can divide the period 0 ° to 360 ° in 2. Consider the waveform of f ( x ) shown in Figure 6.. In these situations. were we have divided it into small pulses of width Δ x . we need to evaluate the integral(s) using numerical integration. we multiply these by Δ x . and we enter these in subsequent columns of the spreadsheet. Next. we form the sums of y cos x and y sin x .) and divide by 180 to perform the conversion. we enter the values of sin x and y sin x in Columns F and G respectively. y cos 3x . We enter these in Column B and we denote them as x . The procedure presented here. a week. If the time axis is in degrees. Similarly. we can choose Δ x to be 2. Figure 6. the better the approximation. such as Microsoft Excel. In Column C we enter the corresponding values of y = f ( x ) as measured from the waveform.5 ° intervals. Obviously. and so on. The complete tables extend to the seventh harmonic to the right and to 360 ° down. Even though we may already know the Fourier series from analytical methods.30 is a partial table showing the computation of the coefficients of the square waveform.Numerical Evaluation of Fourier Coefficients whose period may be a day.29. we can use this procedure to check our results.

0 DC= a1= a2= a3= a4= a5= a6= a7= b1= b2= b3= b4= b5= b6= b7= 1.985 cos3x ycos3x 1.991 0. Third Edition Copyright © Orchard Publications 0.259 0.793 0.654 0.0 2.866 0.500 ysin3x 0.383 0.044 0.866 0.342 0.991 0.0 32.643 0.462 0.793 0.574 0.611 0.676 0.423 0.567 0.924 0.5 10.766 0.131 0.180 0.259 0.866 0.000 0.996 0.985 0.000 1.216 0.5 45.087 0.000 0.999 0.906 0.866 0.996 1.940 0.259 0.737 0.000 1.342 0.174 0.000 1.423 0.906 0.044 0.000 1.707 -0.259 0.609 0.985 0.924 0.500 x(deg) x(rad) Numerical Analysis Using MATLAB® and Excel®.991 0.537 0.500 0.000 0.991 1.000 1.259 0.131 0.342 0.906 0.819 0.966 0.793 -0.829 0.924 0.000 1.609 -0.574 0.500 -0.609 0.044 0.044 1.537 0.436 0.174 0.866 0.866 0.000 0.044 0.707 0.500 0.866 sin3x 0.000 1.044 0.0 7.609 0.000 0.000 1.609 0.0 4.707 0.643 0.996 0.349 0.5 20.954 0.383 0.131 0.000 1.819 0.254 0.0 47.044 0.044 0.766 0.940 0.996 0.500 0.131 0.0 0.707 0.087 0.044 0.676 0.707 0.000 0.866 0.383 0.000 -0.966 0.766 0.044 0.793 -0.305 0.866 0.259 0.609 0.0 -1.819 0.5 40. Numerical computation of the coefficients of the square waveform (partial listing) y=f(x) 0.707 0.000 0.966 0.766 0.087 0.707 0.462 0.966 0.996 0.044 0.000 1.301 0.423 0.000 1.383 0.000 0.707 0.000 0.5 30.0 42.707 0.707 0.676 0.0 0.0 17. and Maclaurin Series Figure 6.044 0.6−38 Analytical: 1.676 0.000 1.044 0.966 0.924 0.924 0.996 1.985 0.044 0.766 0.500 -0.996 0.259 -0.044 0.976 0.342 0.873 .044 0.423 0.966 0.259 -0.737 0. Taylor.174 0.574 0.906 0.5 Square waveform f(t)=4(sinwt/p+sin3wt/3p+sin5wt/5p+ ….000 -0.000 0.000 1.707 0.000 0.985 0.866 0.0 27.0 12.643 0.218 0.574 0.991 0.819 0.044 0.742 0.924 0.000 0.643 0.999 0.000 0.737 0.0 2.737 0.707 -0.044 0.500 0.976 0.793 0.000 0.087 -0.259 0.609 0.000 0.301 0.000 0.574 0.707 0.000 0.906 0.707 0.000 0.216 0.000 Chapter 6 Fourier.131 -0.000 0.906 0.5*a0 0.000 1.000 0.766 0.524 0.793 0.0 22.991 1.423 0.383 -0.000 1.000 0.843 0.044 0.966 0.087 0.609 -0.793 0.643 0.0 -0.174 0.609 0.087 0.0 37.500 0.383 0.766 0.643 0.000 1.383 0.000 0.259 0.966 0.966 0.940 0.966 0.766 1.0 8.044 0.5 15.131 0.924 0.000 0.000 1.000 0.000 0.131 -0.174 0.131 0.866 0.0 6.843 0.866 0.985 cosx ycosx sinx ysinx cos2x ycox2x sin2x ysin2x 0.044 0.887 0.087 0.500 0.819 0.383 -0.991 0.) Numerical: Average= 1.643 0.985 0.174 0.985 0.342 0.500 0.793 0.887 0.500 0.996 0.5 0.5 0.866 0.5 5.000 0.342 0.044 0.262 0.259 0.259 0.966 0.5 35.966 0.000 0.793 0.954 0.574 0.175 0.819 0.424 0.087 -0.000 0.000 -0.30.940 0.174 0.273 0.087 0.5 -1.698 0.940 0.393 0.5 25.131 0.793 0.707 0.924 0.000 0.707 0.643 0.866 0.000 -0.480 0.423 0.5 50.000 1.500 0.000 1.609 0.785 0.000 1.044 0.500 0.940 0.991 0.174 0.

354 0.966 0.793 -0.480 0.246 0.022 0.015 0.Analytical: Sine wave clipped at π/6.470 0.010 0.793 0.500 0.369 0.433 0.676 0.5 30.000 0.500 0.643 0.999 0. 5π/6 etc.000 0.146 0.383 0.174 0.000 0.000 0.498 0.793 0.000 -0.179 0.462 0.397 -0.216 0.028 b6= 0.676 0.022 0.000 0.0 8.084 0.218 0.000 0.500 0.000 b5= 0.500 0.483 0.422 0.5 -1.574 0.224 0.643 0.354 0.0 0.087 0.0 6.31.985 0.0 7.766 0.500 0.321 0.742 0.065 -0.483 0.378 0.017 0.5 45.342 0.5 40.017 0.462 0.146 0.000 0.5 0.5 10.000 -0.044 0.131 0.985 0.500 0.342 0.349 0.924 0.996 0.5 1.707 0.250 0.220 0.000 0.022 0.383 0.996 0.766 0.171 0.707 0.259 0.496 0.0 37.304 0.250 0.287 0.000 Numerical Analysis Using MATLAB® and Excel®.211 0.707 0.050 0.008 0.013 0.265 0.966 0.500 0.006 0.5 35.985 0.707 0.000 0.000 -0.766 0.271 0.492 0.216 0.000 -0.906 0.060 0.0 12.090 0.354 -0.022 0.150 0.009 0.707 0.500 0.991 0.183 0.163 0.0 0.500 0.453 0. Numerical computation of the coefficients of a clipped sine waveform (partial listing) x(deg) x(rad) Numerical Evaluation of Fourier Coefficients 0.000 0.954 0.737 0.0 4.0 27.843 0.087 0.966 0.940 0.866 0.044 0.397 0.262 0.138 b4= 0.175 0.940 0.5 15.000 0.131 0.183 0.433 sin3x 0.966 0.866 0.0 22.296 0.991 0.043 0.383 0.087 0.174 0.393 0.707 0.087 -0.030 0.305 0.737 0.654 0.609 0.000 0.304 0.996 0.171 0.301 0.500 -0.500 0. Numerical: f(t)=unknown 1.271 0.131 0.609 b2= 0.022 0.354 0.174 0.423 0.500 0.259 0.574 0.091 0.873 6−39 .0 -1.000 0.250 0.609 0.354 0.985 ycosx sinx ysinx cos2x ycox2x sin2x 0.183 0.991 1.213 0.000 0.250 -0.397 0.0 32.0 -0.087 0.500 0.707 -0.866 0.966 0.172 0. Third Edition Copyright © Orchard Publications y=f(x) cosx 1.259 0.287 0.324 0.015 0.5 20.129 0.262 0.011 0.338 0.174 0.500 0.087 0.574 0.611 0.321 0.785 0.022 0.132 0.866 0.966 0.500 ysin3x 0.524 0.018 0.866 0.0 DC= a1= a2= a3= a4= a5= a6= a7= b1= 0.609 0.0 2.131 -0.191 -0.5 50.342 0.410 0.044 -0.866 0.087 0.643 0.304 -0.000 0.086 0.369 0.109 0.047 0.819 0.766 0.462 0.537 0.458 0.129 0.567 0.059 0.272 0.259 0.087 0.259 0.866 0.000 b3= 0.5 2.383 0.129 0.000 0.000 0.321 0.433 0.410 0.087 0.643 0.383 0.002 0.423 0.131 0.793 0.436 0.500 0.500 0.000 0.492 cos3x ycos3x 1.131 0.0 17.000 0.129 -0.022 0.498 0.087 0.239 0.887 0.433 0.0 0.022 0.034 0.172 0.940 0.924 0.0 47.196 0.174 0.819 0.000 0.269 0.043 0.044 0.121 0.5 5.924 0.5 25.004 0.117 0.000 0.5*a0 ysin2x 0.044 0.996 1.301 0.354 0.976 0.338 0.383 -0.707 0.126 0.023 0.250 Figure 6.991 0.924 0.000 0.342 0.698 0.044 0.002 0.819 0.423 0.906 0.006 0.020 0.829 0.259 -0.004 0.259 0.423 0.000 0.383 0.171 0.000 b7= -0.383 1.500 0.067 0.793 0.0 42.408 0.022 0.906 0.000 0.609 -0.211 0.609 0.008 0.

that is. Third Edition Copyright © Orchard Publications .+ -----------( kv ) .+ ---.+ ---.116) yields.116) Substitution of (6.+ …⎞ ( kv ) . we rewrite (6.– ----+… cos x = 1 – ---2! 4! 6! 2 4 6 3 5 7 The following example illustrates the fact that a power series expansion can lead us to a Fourier Series.+ … – 1⎞ = a ⎛ kv + ( kv ) .– ----+… 3! 5! 7! x x x .114) suggests the power series expansion of (6. Example 6. Accordingly. and Maclaurin Series 6. the current i in a semiconductor diode can be approximated by the relation i = a(e kv – 1) (6.11 If the applied voltage v is small (no greater than 5 volts). v = V max cos ω t (6. Taylor.+ ( kv ) . Solution: The term e inside the parentheses of (6.110) Some familiar power series expansions for real values of x are x x x x e = 1 + x + ---.Chapter 6 Fourier.+ ----+… 2! 3! 4! 2 3 4 (6.9 Power Series Expansion of Functions A power series has the form ∞ ∑ ak x k=0 k = a0 + a1 x + a2 x + … 2 (6.113) x x x sin x = x – ---.+ ---.115) into (6.114) as kv ) .114) as a power series.115) Express the current i in (6.+ -----------( kv ) . and the input voltage is a sinusoid.111).114) where a and k are arbitrary constants.111) (6.112) (6. 6−40 Numerical Analysis Using MATLAB® and Excel®.+ ( ---------------------------------i = a ⎛ 1 + kv + -----------⎝ ⎠ ⎝ ⎠ 2! 3! 4! 2! 3! 4! 2 3 4 2 3 4 kv (6.

119) We recall that the series of (6.x + … + ----------------x f ( x ) = f ( 0 ) + f' ( 0 ) x + ----------2! n! (n) (6. a term of the fundamental frequency.cos 3x .118) into (6. and terms of all harmonic frequencies. To appreciate the usefulness and application of the Taylor series. (6.Taylor and Maclaurin Series ( kV p cos ω t ) ( kV p cos ω t ) ( kV p cos ω t ) .121) Relation (6.120) reduces to f (0) n f'' ( 0 ) 2 .118) 1 4 3 1 .119) is the trigonometric series form of the Fourier series. that is.+ -8 8 2 Then.110) with an = f (n) ( 0 ) ⁄ n! .cos 2x + -cos x = -.+ …⎞ i = a ⎛kV max cos ω t + ------------------------------⎝ ⎠ 2! 3! 4! 2 3 4 (6.117) and after simplification.cos x + 1 cos x = -4 4 (6.32. higher frequencies which are multiples of the fundamental frequency.10 Taylor and Maclaurin Series A function f ( x ) which possesses all derivatives up to order n at a point x = x 0 can be expanded in a Taylor series as f'' ( x 0 ) f ( x0 ) n .+ -2 2 3 3 -. We observe that it consists of a constant term.cos 2x cos x = -. Numerical Analysis Using MATLAB® and Excel®. 6. substitution of (6.+ ------------------------------. we obtain a series of the following form: i = A 0 + A 1 cos ω t + A 2 cos 2 ω t + A 3 cos 3 ω t + A 4 cos 4 ω t + … (6. and has the form of power series of (6.117) This expression can be simplified with the use of the following trigonometric identities: 2 1 1 .+ ------------------------------.121) is known as Maclaurin series. where i ( v ) represents some experimental data for the current−voltage ( i – v ) characteristics of a semiconductor diode operating at the 0 ≤ v ≤ 5 volts region. we will consider the plot of Figure 6.( x – x 0 ) 2 + … + ------------------f ( x ) = f ( x 0 ) + f' ( x 0 ) ( x – x 0 ) + ------------( x – x0 ) 2! n! (n) (6.120) If x 0 = 0 .cos 4x . Third Edition Copyright © Orchard Publications 6−41 .

Obviously. Third Edition Copyright © Orchard Publications . and we denote this first approximation as a 0 shown in Figure 6.33. We begin by referring to the power series of (6. We assume that the first n derivatives of the function i ( v ) exist at this point. Taylor. i 0 ) shown in Figure 6.33. where we observe that the first term on the right side is a constant. Approximation of the function i ( v ) by a power series 6−42 Numerical Analysis Using MATLAB® and Excel®.32. i i(v) P ( v 0.110). i 0 ) i0 0 v0 v Figure 6.34. the horizontal line i 0 passes through point P . we are seeking a constant that it will be the best approximation to the given curve in the vicinity of point P . suppose that we want to approximate the function i ( v ) by a power series. Therefore. in the neighborhood of some arbitrary point P ( v 0.Chapter 6 Fourier. and Maclaurin Series i i(v) 0 v Figure 6. Current-voltage (i-v) characteristics for a typical semiconductor diode Now.

the linear term a 0 + a 1 ( v – v 0 ) can be expressed as i ( v 0 ) + i' ( v 0 ) ( v – v 0 ) . This will be accomplished if the linear term has the same slope as the given function as shown in Figure 6.34.Taylor and Maclaurin Series i i(v) P ( v 0. we seek a linear term of the form a 0 + a 1 x . we want the linear term a 0 + a 1 ( v – v 0 ) to be the best approximation to the function i ( v ) in the vicinity of point P . Thus. First approximation of i ( v ) v The next term in the power series is the linear term a 1 x . we express the desired power series as f ( v ) = a0 + a1 ( v – v0 ) + a2 ( v – v 0 ) + a3 ( v – v0 ) + a4 ( v – v0 ) + … 2 3 4 (6. Third Edition Copyright © Orchard Publications 6−43 .122) Now. But since we want the power series to be a good approximation to the given function for some distance on either side of point P . i 0 ) i0 a0 + a0 ( v – v0 ) a0 a0 ( v – v0 ) v 0 v0 Figure 6. i i( v) P ( v 0. Accordingly.35. i 0 ) i0 a0 0 v0 Figure 6. Second approximation of i ( v ) It is evident that the slope of i ( v ) at v 0 is i' ( v 0 ) = a 1 and therefore.35. we are interested in the difference v – v 0 . Numerical Analysis Using MATLAB® and Excel®.

Chapter 6 Fourier. by a Taylor series as illustrated by the following example. Third Edition Copyright © Orchard Publications .12 Compute the first three terms of the Taylor series expansion for the function y = f ( x ) = tan x (6. and so on of (6. Solution: The Taylor series expansion about point a is given by f''' ( a ) f'' ( a ) .122). 6−44 Numerical Analysis Using MATLAB® and Excel®. Example 6. a 5 . i 0 ) a0 + a1 ( v – v0 ) a0 a1 ( v – v0 ) a2 ( v – v0 ) i0 2 0 v0 v Figure 6. a 4. that is. i'' ( v 0 ) i''' ( v 0 ) . we choose a 2 such that it matches the second derivative of the function i ( v ) in the vicinity of point P as shown in Figure 6. fifth.36.36.124) at a = π ⁄ 4 . When this is done.122) are found by matching the third.( x – a ) 2 + -----------f n ( x ) = f ( a ) + f' ( a ) ( x – a ) + ----------3! 2! (6. a 2 ( v – v 0 ) is a quadratic and therefore. Taylor. fourth. we obtain the following Taylor series. we must find the first and second derivatives of f ( x ) = tan x .( v – v0 )3 + … i ( v ) = i ( v 0 ) + i' ( v 0 ) ( v – v 0 ) + ------------2! 3! (6.( v – v 0 ) 2 + -------------. and Maclaurin Series The third term in (6.125) and since we are asked to compute the first three terms.( x – a )3 + … .123) We can also describe any function that has an analytical expression. and higher order derivatives of the given function with these coefficients. i i( v) a0 + a1 ( v – v0 ) + a2 ( v – v0 ) 2 2 P ( v 0. The remaining coefficients a 3. Third approximation of i ( v ) Then. 2a 2 = i'' ( v 0 ) or a 2 = i'' ( v 0 ) ⁄ 2 .

1/4 pi) 64 5 244 6 2176 7 + -.(x . pretty(z) 2 3 4 1 + 2x .130) We can also obtain a Taylor series expansion with the MATLAB taylor(f. 2 -. using the trigonometric identity sec x = tan x + 1 2 2 (6.129) and by substitution into (6.(x . at point a = π ⁄ 4 we have: π . y=tan(x). we get. so we let z = sec x .= 2 sec x ----dx dx (6. d tan x = sec x .⎞ = tan ⎛ π --⎞ = 1 f ( a ) = f ⎛ -⎝4⎠ ⎝4⎠ --⎞ = 1+1 = 2 f' ( a ) = f' ⎛ ⎝4⎠ π -. Then. and a defines the Taylor approximation about point a .a) function where f is a symbolic expression.125).n.1/4 pi) + --.1/4 pi) + 10/3(x . x=sym('x').Taylor and Maclaurin Series 2 2 From math tables. we get derivative of sec x .1/4 pi) + 8/3(x . 2 dz ----. Third Edition Copyright © Orchard Publications 6−45 .128) Now. using ----dx d 2 dz . the following MATLAB script computes the first 8 terms of the Taylor series expansion of y = f ( x ) = tan x about a = π ⁄ 4 .8.⎞ + 2⎛x – π --⎞ +… fn ( x ) = 1 + 2 ⎛ x – π ⎝ ⎠ ⎝ ⎠ 4 4 (6.sec x = 2 sec x ( sec x ⋅ tan x ) = 2 sec x ⋅ tan x ----. For example.126) Next.⎞ = 2(1 + 1)1 = 4 f'' ( a ) = f'' ⎛ ⎝4⎠ π 2 (6.= f'' ( x ) = 2 ( tan x + 1 ) tan x dx (6. n produces the first n terms in the series. so f ′ ( x ) = sec x .126).1/2 pi + 2(x . To find f'' ( x ) we need to find the first dx d 2 2 .127) into (6.sec x = sec x ⋅ tan x .1/4 pi) 15 45 315 Numerical Analysis Using MATLAB® and Excel®. A detailed description can be displayed with the help taylor command.pi/4).1/4 pi) + ---.127) and by substitution of (6.(x . z=taylor(y.

) n f (0f'' ( 0 ) 2 x x + … + ---------------f ( x ) = f ( 0 ) + f' ( 0 ) x + ----------n! 2! t t y = f (t) = e t (6. and Maclaurin Series Example 6. the constant n has a value between 1 and 2 depending on the material and physical structure of the diode. f' ( 0 ) = f'' ( 0 ) = f''' ( 0 ) = … = 1 and therefore.+ ----+… f n ( t ) = 1 + t + ---2! 3! 2 3 (6. Third Edition Copyright © Orchard Publications . then. Expand this relation into a power series that can be used to compute the current when the voltage is small and varies about v D = 0 . pretty(fn) 2 3 4 5 1 + t + 1/2 t + 1/6 t + 1/24 t + 1/120 t Example 6.133) MATLAB displays the same result. the instantaneous current i D and voltage v D are related as iD ( vD ) = ID e v D ⁄ nV T (6.Chapter 6 Fourier.132) For this function.132). and its value at room temperature is approximately 25 mV . Solution: A Maclaurin’s series has the form of (6. Solution: Since the voltage is small and varies about v D = 0 .13 Express the function in a Maclaurin’s series.14 In a semiconductor diode D . and V T is the thermal voltage which depends on the temperature. t t . we have f ( t ) = e and thus f ( 0 ) = 1 . that is. 6−46 Numerical Analysis Using MATLAB® and Excel®. fn=taylor(exp(t)). we can use the following Maclaurin’s series. Taylor.131) (n) (6. t=sym('t').134) where I D is the DC (average) component of the current. Since all derivatives are e .

These are: v ⁄ nV d 1 1 .138) Then.137).135) The first term i D ( 0 ) on the right side of (6.2 1v D + -----------v D + …⎟ i D ( v D ) = I D ⎜1 + --------2 2 ⎝ nV T ⎠ n VT (6.v + --------------.136) To compute the second and third terms of (6.Taylor and Maclaurin Series i'' D ( 0 ) 2 i''' D ( 0 ) 3 .135) we get ⎛ ⎞ 1 .135) is found by letting v D = 0 in (6. (6. we must find the first and second derivatives of (6. and (6.⋅ I D e D T and i' D ( 0 ) = ---------⋅I i' D ( v D ) = -------dv D nV T D nV T v D ⁄ nV T 1 d 1 -----------i'' D ( v D ) = -----------i I = ⋅ e and i''D ( 0 ) = -----------⋅ ID D D 2 2 2 2 2 n VT n VT d vD 2 (6. Third Edition Copyright © Orchard Publications 6−47 .vD + … i D ( v D ) = i D ( 0 ) + i' D ( 0 ) v D + -------------2! D 3! (6.139) Numerical Analysis Using MATLAB® and Excel®.138) into (6. Then. iD ( 0 ) = ID (6.134). by substitution of (6.136).135).134).i D = --------.137) (6.

Likewise. the series will consist of sine terms only. but inverted. a n . and b n are found from the following relations: 1 1 -. has half−wave symmetry if –f ( t + T ⁄ 2 ) = f ( t ) that is.a = ----2 0 2π 1 a n = -π ∫0 2π f ( t ) dt ∫0 2π f ( t ) cos nt dt 1 b n = -π ∫0 2π f ( t ) sin nt dt • If a waveform has odd symmetry. The coefficients a 0 . if it is an odd function. and represents the DC (average) component of f ( t ) . that is. Odd functions are those for which – f ( – t ) = f ( t ) . if it is an even function. Taylor. that is. the shape of the negative half−cycle of the waveform is the same as that of the positive half−cycle. the series will consist of cosine terms only. represent the fundamental frequency component ω . • If a waveform has even symmetry. and a 0 may or may not be zero.11 Summary • Any periodic waveform f ( t ) can be expressed as 1 -a + f ( t ) = -2 0 ∞ n=1 ∑ ( a cos n ω t + b sin n ω t ) n n where the first term a 0 ⁄ 2 is a constant.sin 5 ω t + …⎞ = 4A -. If a waveform has half−wave symmetry only odd (odd cosine and odd sine) harmonics will be present. • The trigonometric Fourier series for the square waveform with odd symmetry is 4A ------. Even functions are those for which f( –t ) = f( t ) • A periodic waveform with period T .sin 3 ω t + 1 .⎛ sin ω t + 1 f ( t ) = -----⎠ π 5 3 π ⎝ n = odd ∑ 1 -. In other words.sin n ω t n 6−48 Numerical Analysis Using MATLAB® and Excel®. the terms with the coefficients a 2 and b 2 together. The terms with the coefficients a 1 and b 1 together. and Maclaurin Series 6. represent the second harmonic component 2 ω .Chapter 6 Fourier. Third Edition Copyright © Orchard Publications . all even (even cosine and even sine) harmonics will be zero. and so on.

except C 0 .------------.a + ---. Third Edition Copyright © Orchard Publications 6−49 .sin n ω t n • The trigonometric Fourier series for the triangular waveform with odd symmetry is 8A 1 8A ⎛ 1 .sin 2 ω t + 1 .sin t – A --.sin 7 ω t + …⎞ = -----. C – n = C n∗ Numerical Analysis Using MATLAB® and Excel®.cos n ω t n • The trigonometric Fourier series for the sawtooth waveform with odd symmetry is 2A ------.⎛ a n – ---.cos ω t – 1 -. are complex.( a –j b ) .– -----π π 1 -----------------cos n ω t 2 (n – 1) n = 2.sin 3 ω t – 1 -.sin 3 ω t + ----( –1 ) 2 ⎝ 2 ⎠ 49 25 9 π n = odd π ∑ (n – 1) ---------------2 1 ---.sin 4 ω t + …⎞ = 2A -.sin n ω t 2 n • The trigonometric Fourier series for the half−wave rectification waveform with no symmetry is A cos 8t + … cos 6t + ------------cos 4t + ------------cos 2t + --------------.= -C n = -2 n n 2⎝ n j ⎠ 1 -a C 0 = -2 0 • The C i coefficients.sin 5 ω t – ----.Summary • The trigonometric Fourier series for the square waveform with even symmetry is 4A 1 4A ⎛ . … ∑ ∞ • The Fourier series are often expressed in exponential form as f ( t ) = … + C –2 e – j2 ω t + C –1 e –j ω t + C0 + C1 e jωt + C2 e j2 ω t +… where the C i coefficients are related to the trigonometric form coefficients as b n⎞ 1 1 .cos 3 ω t + -f ( t ) = -----⎠ π 5 π ⎝ 3 n = odd ∑ ( –1 ) (n – 1) ---------------2 1 -.= -C – n = -2 n 2⎝ j⎠ b n⎞ 1 1⎛ . and appear as complex conjugate pairs.( a + jb n ) .⎛ sin ω t – 1 f ( t ) = -----⎝ ⎠ π 4 3 2 π ∑ ( –1 ) n–1 1 -.cos 5 ω t – …⎞ = -----. 4. that is.sin ω t – 1 f ( t ) = ------.+ --f(t) = A π 2 63 35 15 3 π • The trigonometric Fourier series for the full−wave rectification waveform with even symmetry is 2A 4A f ( t ) = -----. 6.

**Chapter 6 Fourier, Taylor, and Maclaurin Series
**

• In general, for ω ≠ 1 , 1 C n = -T

∫0

T

f( t)e

– jn ω t

1 d( ω t ) = ----2π

∫0

2π

f(t )e

– jn ω t

d( ω t )

• We can derive the trigonometric Fourier series from the exponential series from the relations a n = Cn + C–n

and

b n = j ( Cn – C–n )

• For even functions, all coefficients C i are real • For odd functions, all coefficients C i are imaginary • If there is half−wave symmetry, C n = 0 for n = even • C –n = C n∗ always • A line spectrum is a plot that shows the amplitudes of the harmonics on a frequency scale. • The frequency components of a recurrent rectangular pulse follow a sin x ⁄ x form. • We can evaluate the Fourier coefficients of a function based on observed values instead of an

analytic expression using numerical evaluations with the aid of a spreadsheet.

• A power series has the form

∑ ak x k=0

in a Taylor series as

∞

k

= a0 + a1 x + a2 x + …

2

**• A function f ( x ) that possesses all derivatives up to order n at a point x = x 0 can be expanded f'' ( x 0 ) f ( x0 ) n - ( x – x 0 ) 2 + … + ------------------f ( x ) = f ( x 0 ) + f' ( x 0 ) ( x – x 0 ) + ------------( x – x0 ) 2! n!
**

(n)

**If x 0 = 0 , the series above reduces to
**

f (0) n f'' ( 0 ) 2 - x + … + ----------------x f ( x ) = f ( 0 ) + f' ( 0 ) x + ----------2! n!

(n)

**and this relation is known as Maclaurin series
**

• We can also obtain a Taylor series expansion with the MATLAB taylor(f,n,a) function where f is a symbolic expression, n produces the first n terms in the series, and a defines the Taylor approximation about point a .

6−50

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

Exercises 6.12 Exercises

1. Compute the first 5 components of the trigonometric Fourier series for the waveform below. Assume ω = 1 .

f( t) A

0

ωt

**2. Compute the first 5 components of the trigonometric Fourier series for the waveform below. Assume ω = 1 .
**

f( t) A 0 ωt

**3. Compute the first 5 components of the exponential Fourier series for the waveform below. Assume ω = 1 .
**

f(t)

A

0

ωt

**4. Compute the first 5 components of the exponential Fourier series for the waveform below. Assume ω = 1 .
**

f( t) A⁄2 0 –A ⁄ 2 ωt

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

6−51

**Chapter 6 Fourier, Taylor, and Maclaurin Series
**

5. Compute the first 5 components of the exponential Fourier series for the waveform below. Assume ω = 1 .

f(t)

A

0

ωt

**6. Compute the first 5 components of the exponential Fourier series for the waveform below. Assume ω = 1 .
**

f(t)

A ωt

0 −A Figure 6.37. Waveform for Exercise 6

**7. Compute the first 4 terms of the Maclaurin series for each of the following functions. a. f ( x ) = e
**

–x

b. f ( x ) = sin x

c. f ( x ) = sinh x

Confirm your answers with MATLAB. 8. Compute the first 4 terms of the Taylor series for each of the following functions.

1 - about a = – 1 a. f ( x ) = -x -b. f ( x ) = sin x about a = – π 4

**Confirm your answers with MATLAB. 9. In a non−linear device, the voltage and current are related as
**

v ⎞ 1.5 i ( v ) = k ⎛ 1 + --⎝ V⎠

where k is a constant and V is the DC component of the instantaneous voltage v . Expand this function into a power series that can be used to compute the current i , when the voltage v is small, and varies about v = 0 .

6−52

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Solutions to End−of−Chapter Exercises 6.13 Solutions to End−of−Chapter Exercises
**

1.

f(t) A A ---t π

–2 π

–π

0

π

2π

ωt

This is an even function; therefore, the series consists of cosine terms only. There is no half− wave symmetry and the average ( DC component) is not zero. We will integrate from 0 to π and multiply by 2 . Then,

2 a n = -π

∫0

π

From tables of integrals,

A --- t cos nt dt = 2A -----2 π π

∫0 t cos nt dt

π

(1)

∫ x cos ax dx

and thus (1) becomes

t 2A ⎛ 1 - sin nt ⎞ - ---- cos nt + -a n = -----2⎝ 2 ⎠ n π n

x 1- sin a x = ---cos ax + -2 a a

π

0

t 2A ⎛ 1 1 - sin nt π – ---- ---- cos n π + -= ------ – 0⎞ 2⎝ 2 ⎠ 2 n π n n

**and since sin nt π = 0 for all integer n ,
**

2A 2A ⎛ 1 1⎞ - ---- cos n π – ---- ( cos n π – 1 ) (2) - = ---------a n = -----2⎝ 2 2⎠ 2 2 π n n n π

**We cannot evaluate the average ( 1 ⁄ 2 ) ⁄ a 0 from (2); we must use (1). Then, for n = 0 ,
**

1 2A -- a 0 = -------2 2 2π

∫0

π

A t - ⋅ --t dt = ---2 π 2

2 π 0

A π - ⋅ ---= ---2 π 2

2

or

( 1 ⁄ 2 ) ⁄ a0 = A ⁄ 2

**We observe from (2) that for n = even , a n = even = 0 . Then,
**

– 4A 4A-, for n = 7, a = ---------– 4A -, for n = 5, a5 = – ---------------, for n = 3, a 3 = ---------for n = 1, a 1 = – 4A 3 2 2 2 2 2 2 2 π 3 π 5 π 7 π

and so on. Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

6−53

**Chapter 6 Fourier, Taylor, and Maclaurin Series
**

Therefore,

4A 111 -- cos 3t + ----- a 0 – 4A ------ ⎛ cos t + 1 --- – -----cos 7t + …⎞ = A cos 5t + ----f ( t ) = -2 ⎝ ⎠ π 49 25 9 2 2 π

∑

n = odd

∞

1---cos nt 2 n

2.

f( t) A 2A ------t π 0 π⁄2 π 3π ⁄ 2 π

ωt

This is an even function; therefore, the series consists of cosine terms only. There is no half− wave symmetry and the average ( DC component) is not zero.

1 Area 2 × [(A ⁄ 2) ⋅ (π ⁄ 2)] + Aπ 3A ⋅ ( π ⁄ 2 ) 3A - a = ---------------- = -------------------------- = ------ = --------------------------------------------------------------Average = -2 0 Period 2π 2π 4 2 a n = -π

∫0

π⁄2

2A 2 ------ t cos nt dt + -π π

and with

∫π ⁄ 2 A cos nt dt

π

(1)

∫ x cos ax dx

(1) simplifies to

**1x 1- sin a x = ---= ---cos ax + -( cos ax + ax sin ax ) 2 2 a a a
**

π⁄2

0

4A 1 - ---- ( cos nt + nt sin nt ) a n = -----2 2 π n

2A - sin nt + -----nπ

π π⁄2

2A ⎛ nπ nπ 4A ⎛ π π - sin n π – sin n - sin ----- cos n = --------------⎞ - – 1 – 0⎞ + ---------- + ----⎠ ⎠ nπ ⎝ 2 2⎝ 2 2 2 2 n π

**and since sin nt π = 0 for all integer n ,
**

4A ⎛ 4A π 4A 2A π π 2A π ⎞ - sin n - sin n - cos n - cos n ----- – -------------- = -------------- + ----------–1 - – -----a n = ---------2 ⎝ ⎠ 2 2 2 2 nπ 2 2 2 2 nπ 2 n π n π n π 4A 4A 2A - ( 0 – 1 ) = – 4A - ( – 1 – 1 ) = – -----for n = 1, a 1 = -----------, for n = 2, a 2 = -------2 2 2 2 π 4π π π 4A – 4A 4A - ( 0 – 1 ) = – --------(1 – 1) = 0 for n = 3, a 3 = --------, for n = 4, a 4 = ---------2 2 2 2 9π 9π 7 π

We observe that the fourth harmonic and all its multiples are zero. Therefore,

6−54

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Solutions to End−of−Chapter Exercises
**

4A -- cos 3t + …⎞ -- cos 2t + 1 - ⎛ cos t + 1 ------ – -----f ( t ) = 3A 2 ⎝ ⎠ 9 2 4 π f(t)

A

3.

0

π

2π

ωt

This is neither an even nor an odd function and has no half−wave symmetry; therefore, the series consists of both cosine and sine terms. The average ( DC component) is not zero. Then,

1 C n = ----2π

∫0

π

2π

f(t )e

– jn ω t

d( ω t )

and with ω = 1

1C n = ----2π

∫0

2π

f(t )e

– jnt

1dt = ----2π

∫0

Ae

– jnt

dt +

∫π

π

0

2π

0e

– jnt

Adt = ----2π

∫0 e

π

– jnt

dt

The DC value is

A C 0 = ----2π

∫0

π

A 0 -t e dt = ----2π

π

0

A = --2

For n ≠ 0

A C n = ----2π

∫0

π

e

– jnt

A – jnt -e dt = -------------– j2 n π

A - ( 1 – e –jn π ) = ----------j2n π

Recalling that

e

– jn π

= cos n π – j sin n π

– jn π

for n = even , e

– jn π

= 1 and for n = odd , e

= – 1 . Then,

and

A -(1 – 1) = 0 C n = even = ----------j2n π A A - [ 1 – ( – 1 ) ] = ------C n = odd = ----------j2n π jn π

**By substitution into the expression
**

f ( t ) = … + C –2 e

– j2 ω t

+ C –1 e

–j ω t

+ C0 + C1 e

jωt

+ C2 e

j2 ω t

+…

we find that Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

6−55

**Chapter 6 Fourier, Taylor, and Maclaurin Series
**

A⎛ 1 –j3 ω t –j ω t j ω t 1 j3 ω t - … – --e -e --- + ---f(t) = A –e + e + -+ …⎞ ⎠ 3 3 2 jπ ⎝

The minus (−) sign of the first two terms within the parentheses results from the fact that C – n = C n∗ . For instance, since C 1 = 2A ⁄ j π , it follows that C –1 = C 1∗ = – 2A ⁄ j π . We observe that f ( t ) is complex, as expected, since there is no symmetry. 4.

f( t) A⁄2 0 –A ⁄ 2 ωt

**This is the same waveform as in Exercise 3 where the DC component has been removed. Then,
**

1 – j3 ω t – j ω t A⎛ j ω t 1 j3 ω t -e - … – --e f ( t ) = ---–e + e + -+ …⎞ ⎠ 3 jπ ⎝ 3

It is also the same waveform as in Example 6.9, Page 6−32, except that the amplitude is halved. This waveform is an odd function and thus the expression for f ( t ) is imaginary. 5.

f(t)

A

–π –π ⁄ 2

0 π⁄2

π

ωt

This is the same waveform as in Exercise 3 where the vertical axis has been shifted to make the waveform an even function. Therefore, for this waveform C n is real. Then,

1C n = ----2π

∫– π

π

f(t )e

– jnt

Adt = ----2π

∫– π ⁄ 2 e

π⁄2

– jnt

dt

The DC value is

At C 0 = ----2π

π⁄2

–π ⁄ 2

A- ⎛ π π⎞ A = ------ + -- = --2 π ⎝ 2 2⎠ 2

For n ≠ 0

6−56

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Solutions to End−of−Chapter Exercises
**

AC n = ----2π

∫– π ⁄ 2

π⁄2

e

– jnt

A - – jnt e dt = -------------– j2 n π

π⁄2

–π ⁄ 2

A - – jn π ⁄ 2 jn π ⁄ 2 = -------------(e –e ) – j2 n π

– jn π ⁄ 2

–e A A ⎛e A π -⎞ = ----- sin n - ------------------------------------- ( e jn π ⁄ 2 – e – jn π ⁄ 2 ) = ----= --------------⎠ nπ j2 nπ ⎝ j2n π 2

jn π ⁄ 2

**and we observe that for n = even , C n = 0 For n = odd , C n alternates in plus (+) and minus (−) signs, that is,
**

Aif n = 1, 5, 9, … C n = ----nπ A C n = – ----- if n = 3, 7, 11, … nπ A f ( t ) = ---+ 2

Thus,

n = odd

∑

A- jn ω t⎞ ⎛ ± ----e ⎝ nπ ⎠

where the plus (+) sign is used with n = 1, 5, 9, … and the minus (−) sign is used with n = 3, 7, 11, … . We can express f ( t ) in a more compact form as

A f ( t ) = ---+ 2

6.

n = odd

∑

( –1 )

(n – 1) ⁄ 2

A jn ω t -----e nπ

f(t) –π ⁄ 2 –π −A

A

2A ------t – 1 π π

ωt

π⁄2 0

**We will find the exponential form coefficients C n from
**

1C n = ----2π

From tables of integrals

∫– π f ( t ) e

ax

π

– jnt

dt

∫

Then,

e xe dx = -----( ax – 1 ) 2 a

ax

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

6−57

**Chapter 6 Fourier, Taylor, and Maclaurin Series
**

1C n = ----2π

∫

– jnt ⎛ – 2A ------ t – 1⎞ e dt + ⎝ ⎠ π –π

0

------ t – 1⎞ e ∫0 ⎛ ⎝ π ⎠

π

2A

– jnt

dt

**Integrating and rearranging terms we get
**

e 4A 1- 4A -------– e - e jn π + e –jn π ⎞ 2A –e -⋅e --------------------------+ --------------------------- – -----– -------- + 2 - ⎛ n π ⋅ --------------------------C n = ----⎠ n 2 π n2 π n π ⎝ j2 j2 2 4A ⎛ nπ - sin n π⎞ = -------------– 1 + n π sin n π + cos n π – ----2 2⎝ ⎠ 2 2n π

jn π – jn π jn π – jn π

**and since sin n π = 0 for all integer n ,
**

2A - ( cos n π – 1 ) C n = ---------2 2 n π –4 A For n = even , C n = 0 and for n = odd , cos n π = – 1 , and C n = ---------2 2 n π

**Also, by inspection, the DC component C 0 = 0 . Then,
**

1 – j3 ω t –j ω t j ω t 1 j3 ω t 4A -e -e - ⎛ … + -+e + e + -+ …⎞ f ( t ) = – -----⎠ 2⎝ 9 9 π

The coefficients of the terms e

– j3 ω t

and e

–j ω t

are positive because all coefficients of C n are

real. This is to be expected since f ( t ) is an even function. It also has half−wave symmetry and thus C n = 0 for n = even as we’ve found. 7.

f (0) n f'' ( 0 ) - x 2 + … + ----------------x f ( x ) = f ( 0 ) + f' ( 0 ) x + ----------2! n!

(n)

a. f ( x ) = e , f ( 0 ) = 1 , f' ( x ) = – e , f' ( 0 ) = – 1 , f'' ( x ) = e , f'' ( 0 ) = 1 , f''' ( x ) = – e f''' ( 0 ) = – 1 , and so on. Therefore,

x x - – ----+… f n ( x ) = 1 – x + ---2! 3!

2 3

–x

–x

–x

–x

,

**MATLAB displays the same result.
**

x=sym('x'); fn=taylor(exp(−x)); pretty(fn) 2 1 - x + 1/2 x - 1/6 x 3 + 1/24 x 4 - 1/120 x 5

6−58

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Solutions to End−of−Chapter Exercises
**

b. f ( x ) = sin x ,

f (0) = 0 , f' ( x ) = cos x , f' ( 0 ) = 1 , f'' ( x ) = – sin x , f'' ( 0 ) = 0 , f''' ( x ) = – cos x , f''' ( 0 ) = – 1 , and so on. Therefore, x x x - + ---- – ----+… f n ( x ) = x – ---3! 5! 7!

3 5 7

**MATLAB displays the same result.
**

x=sym('x'); fn=taylor(sin(x)); pretty(fn) 3 x - 1/6 x + 1/120 x f'' ( x ) = sin hx , f'' ( 0 ) = 0 , 5

c. f ( x ) = sin hx ,

**f ( 0 ) = 0 , f' ( x ) = cos hx , f' ( 0 ) = 1 , f''' ( x ) = cos hx , f''' ( 0 ) = 1 , and so on. Therefore, x x x f n ( x ) = x + ---- + ---- + ----+… 3! 5! 7!
**

3 5 7

**MATLAB displays the same result.
**

x=sym('x'); fn=taylor(sinh(x)); pretty(fn) 3 x + 1/6 x + 1/120 x 5

8.

f'' ( a ) f''' ( a ) - ( x – a ) 2 + ------------ ( x – a )3 + … f n ( x ) = f ( a ) + f' ( a ) ( x – a ) + ----------2! 3!

a. f ( x ) = 1 ⁄ x , f ( a ) = f ( – 1 ) = – 1 , f' ( x ) = – 1 ⁄ x , f' ( a ) = f' ( – 1 ) = – 1 , f'' ( x ) = 2 ⁄ x ,

f'' ( a ) = f'' ( – 1 ) = – 2 , f''' ( x ) = – 6 ⁄ x , f''' ( a ) = f''' ( – 1 ) = – 6 , and so on. Therefore, fn ( x ) = –1 –( x + 1 ) – ( x + 1 ) – ( x + 1 ) + …

2 3 4

2

3

or

fn ( x ) = –2 –x – ( x + 1 ) – ( x + 1 ) + …

2

3

**MATLAB displays the same result.
**

x=sym('x'); y=1/x; z=taylor(y,4,−1); pretty(z) 2 -2 - x - (x + 1) - (x + 1) f' ( x ) = cos x , f' ( a ) = f' ( – π ⁄ 4 ) = 2 ⁄2, 3

b. f ( x ) = sin x ,

f ( a ) = f ( –π ⁄ 4 ) = – 2 ⁄ 2 ,

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

6−59

**Chapter 6 Fourier, Taylor, and Maclaurin Series
**

f'' ( x ) = – sin x , f'' ( a ) = f'' ( – π ⁄ 4 ) = 2 ⁄ 2, f''' ( x ) = – cos x ,

f''' ( a ) = f''' ( – π ⁄ 4 ) = – 2 ⁄ 2 , and so on. Therefore, f n ( x ) = – 2 ⁄ 2 + ( 2 ⁄ 2 ) ( x + π ⁄ 4 ) + ( 2 ⁄ 4 ) ( x + π ⁄ 4 ) – ( 2 ⁄ 12 ) ( x + π ⁄ 4 ) + …

2 3

**MATLAB displays the same result.
**

x=sym('x'); y=sin(x); z=taylor(y,4,−pi/4); pretty(z) 1/2 - 1/2 2 1/2 - 1/12 2 (x + 1/4 pi)

1.5 v -⎞ i ( v ) = k ⎛ 1 + --⎝ V⎠

1/2 + 1/2 2 (x + 1/4 pi) + 1/4 2 3

1/2 (x + 1/4 pi)

2

9.

**The Taylor series for this relation is
**

i''' ( v 0 ) i'' ( v 0 ) - ( v – v 0 ) 2 + -------------- ( v – v0 )3 + … i ( v ) = i ( v 0 ) + i' ( v 0 ) ( v – v 0 ) + ------------2! 3!

Since the voltage v is small, and varies about v = 0 , we expand this relation about v = 0 and the series reduces to the Maclaurin series below.

i'' ( 0 ) 2 -v + … i ( v ) = i ( 0 ) + i' ( 0 ) v + ----------2!

(1)

**By substitution of v = 0 into the given relation we get
**

i(0) = k

**The first and second derivatives of i are
**

3k ⎛ v⎞1 ⁄ 2 - 1 + --i' ( v ) = -----2V ⎝ V⎠ 3k ⎛ v ⎞ –1 ⁄ 2 - 1 + --i'' ( v ) = --------2⎝ V⎠ 4V 3k i' ( 0 ) = -----2V 3k i'' ( 0 ) = --------2 4V

**and by substitution into (1)
**

3k 2 3 2 3k 3 - v + --------- v + --------- v + … = k ⎛ 1 + ------ v + …⎞ i ( v ) = k + -----2 2 ⎝ ⎠ 2V 2V 8V 8V

6−60

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Solutions to End−of−Chapter Exercises
**

MATLAB displays the same result.

x=sym('x'); i=sym(‘i’); v=sym(‘v’); k=sym(‘k’); V=sym(‘V’);... i=k*(1+v/V)^1.5; z=taylor(i,4,0); pretty(z) 2 3 k v k v k v k + 3/2 --- + 3/8 ---- - 1/16 ---V 2 3 V V

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

6−61

2) Numerical Analysis Using MATLAB® and Excel®. Then. x n be some values of x in the interval x 0 ≤ x ≤ x n . Third Edition Copyright © Orchard Publications 7−1 . not necessarily consecutive values of x . x k ) f ( x i. x j ) = ---------------------------xi – xj (7. x k ) = --------------------------------------------xi – xk (7. x j. x j ) – f ( x j. and numerical integration. Finite Differences form the basis of numerical analysis as applied to other numerical methods such as curve fitting. within this interval. It is customary to show the independent variable x . TABLE 7. the first divided difference is defined as: f ( xi ) – f ( xj ) f ( x i. the second divided difference is defined as: f ( x i. x 2. x n – 1.1 The variable x and y = f ( x ) in tabular form x x0 x1 x2 f (x) f ( x0 ) f ( x1 ) f ( x2 ) … xn – 1 xn … f ( xn – 1 ) f ( xn ) Let x i and x j be any two.1) Likewise. data smoothing.1. ….Chapter 7 Finite Differences and Interpolation his chapter begins with finite differences and interpolation which is one of its most important applications. T 7. These applications are discussed in this and the next three chapters.1 Divided Differences Consider the continuous function y = f ( x ) and let x 0. numerical differentiation. and its corre- sponding values of y = f ( x ) in tabular form as in Table 7. x 1.

1 Form a difference table showing the values of x given as 0. the second the values of f ( x ) . TABLE 7. 1.2). 3 Solution: We construct Table 7.= 14 3–7 Likewise.3 with six columns. These differences are computed from (7. x 2 ) x2 f ( x2 ) f ( x 2. and the third through the sixth contain the values of the first through the fourth divided differences. and 9 . x 1 ) x1 f ( x1 ) f ( x 1.Chapter 7 Finite Differences and Interpolation The third. for the third divided difference we have 7−2 Numerical Analysis Using MATLAB® and Excel®. and other relations for higher order divided differences. x 3 ) f ( x 0. 3.2 Conventional presentation of divided differences x x0 f (x) f ( x0 ) f ( x 0. x 3 ) Example 7. and the first through the fourth divided differences.1) as 1 – 27 -------------.2. Third Edition Copyright © Orchard Publications . and so on divided differences. x 1.1). 4. the second value on the first divided difference is found from (7. x 1. 7. x 2. x 2. The divided differences are indicated in a difference table where each difference is placed between the values of the column immediately to the left of it as shown in Table 7.2) as 37 – 93 ----------------.= 13 1–3 and third value on the second divided difference is found from (7. are defined similarly. The first column contains the given values of x . the values of f ( x ) corresponding to y = f ( x ) = x . x 3 ) x3 f ( x3 ) f ( x 1. (7. 2. x 2 ) f ( x 0. fourth. For instance.

= 0 0–4 We observe that. Third Edition Copyright © Orchard Publications 7−3 . the denominators are all the same. the differences are sets of consecutive values. the second differences are Δ fk = Δ ( Δ fk ) = Δ fk + 1 – Δ fk 2 (7. they can be omitted.= 1 0–4 and for the fourth 1–1 ----------. These values are referred to as just the differences of the function.3 Divided differences for Example 7. 0. … (7. 1.4) Likewise. therefore. 2. the values of x in a table are equally spaced. If the constant difference between successive values of x is h .Divided Differences TABLE 7. all subsequent differences will be equal to zero. the typical value of x k is x k = x 0 + kh for k = … . if the values of the nth divided difference are the same. In this case. In most cases. as in the fifth column (third divided differences for this example). – 1.5) Numerical Analysis Using MATLAB® and Excel®.3) We can now express the first differences in terms of the difference operator Δ as Δ fk = fk + 1 – fk (7.1 Function x 0 1 3 4 7 9 Divided Differences 3 f( x) = x 0 First 1 Second Third Fourth 1 13 27 37 64 93 343 193 729 4 1 8 1 14 1 20 0 0 4 –8 ----------. Moreover.– 2.

and the first through the fourth differences.Chapter 7 Finite Differences and Interpolation and. for positive integer values of n Δ fk = Δ ( Δ n n–1 fk ) = Δ n–1 fk + 1 – Δ n–1 fk (7.6) The difference operator Δ obeys the law of exponents.2 Construct a difference table showing the values of x given as 1. in general.4. 6. that is. 7 and 8 . 5.4 Divided differences table in terms of the difference operator Δ Function x f First Second Differences Third Fourth … x0 f0 Δ f0 x1 f1 Δ f1 Δ f0 Δ f0 Δ f1 Δ f2 Δ f1 Δ f2 Δ f3 2 3 2 3 2 x2 f2 Δ f0 4 x3 f3 x4 … xn f4 fn Example 7.7) We construct the difference table in terms of the difference operator Δ as shown in Table 7. 3 7−4 Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications . TABLE 7. 4. 2. the values of f ( x ) corresponding to y = f ( x ) = x . 3. Δ ( Δ fk ) = Δ m n m+n fk (7.

8) n n–1 n n(n – 1) . TABLE 7. relation (7.10) Numerical Analysis Using MATLAB® and Excel®. 2. as expected.5. Using the binomial expansion we can show that n! ⎛ n⎞ = -------------------⎝ j⎠ j! ( n – j )! (7.2 Function k 1 2 3 4 5 6 7 8 xk 1 2 3 4 5 6 7 8 fk 1 7 8 19 27 37 64 61 125 91 216 127 343 169 512 4 Differences Δ fk Δ fk 2 Δ fk 3 Δ fk 4 … 12 6 18 6 24 6 30 6 36 6 42 0 0 0 0 We observe that the fourth differences Δ f k are zero.5 Difference table for Example 7.fk + n – 2 + … + ( –1 ) Δ f k = f k + n – nf k + n – 1 + ------------------nf k + 1 + ( – 1 ) f k 2! (7.Divided Differences Solution: Following the same procedure as in the previous example.9) For k = 0 . Third Edition Copyright © Orchard Publications 7−5 .9) reduces to Δ f0 = f2 – f1 Δ f 0 = f 2 – 2f 1 + f 0 Δ f 0 = f 3 – 3f 2 + 3f 1 – f 0 Δ f 0 = f 4 – 4f 3 + 6f 2 – 4 f 1 + f 0 4 3 2 (7. we construct Table 7. n = 1. 3 and 4 .

13) (x) –( n ) These expressions resemble the power functions x and x in elementary algebra.17) To compute the coefficient a 1 . we can express that polynomial as pn ( x ) = a0 + a1 ( x ) (1) + a2 ( x ) (2) + … + an ( x ) (n) (7.Chapter 7 Finite Differences and Interpolation It is interesting to observe that the first difference in (7. the nth differences of a polynomial of degree n are constant.16) reduces to a0 = pn ( 0 ) (7.11) As with derivatives.12) (7.13) we obtain and Δ(x) Δ(x) (n) = n(x) (n – 1) (7.12) and (7.10).16) and now our task is to compute the coefficients a k .15) n –n –( n ) = –n ( x ) –( n – 1 ) We observe that (7.14) we obtain Δ p n ( x ) = 1x a 1 + 2a 2 ( x ) 0 (1) + 3a 3 ( x ) (2) + … + na n ( x ) (n – 1) (7.18) and letting x = 0 . 7. is the difference quotient whose limit defines the derivative of a continuous function that is defined as Δx → 0 lim f ( x1 + Δ x ) – f ( x 1 ) Δy -----.15) are very similar to differentiation of x and x . it is desirable to express a polynomial p n ( x ) as a factorial polynomial. Third Edition Copyright © Orchard Publications . Using (7. we take the first difference of p n ( x ) in (7. Then. Using the difference operator Δ with (7. we find that 7−6 Numerical Analysis Using MATLAB® and Excel®. Occasionally.2 Factorial Polynomials The factorial polynomials are defined as and (x) (n) = x ( x – 1 ) ( x – 2 )… ( x – n + 1 ) 1 = --------------------------------------------------------( x – 1 ) ( x – 2 )… ( x + n ) n –n (7. in analogy with Maclaurin power series.14) and (7.= lim ------------------------------------------Δx Δx Δx → 0 (7.16). For x = 0 . relation (7.14) (7.

26) into (7. Δ pn ( 0 ) Δ pn ( 0 ) a 2 = ------------------. Third Edition Copyright © Orchard Publications 7−7 . We divide p n ( x ) in (7.26) By substitution of (7.20) and for x = 0 .27) Numerical Analysis Using MATLAB® and Excel®. q0 ( x ) = r1 + ( x – 1 ) q1 ( x ) (7. 1.23). ….25) 3.22) Factorial polynomials provide an easier method of constructing a difference table.16). and using the form of relation (7. n a j = -----------------j! j 2 (7. We divide q 1 ( x ) in (7. (7.21) In general.23) 2. Then. we obtain pn ( x ) = r0 + r1 ( x ) = r0 + r1 ( x ) (1) (1) + x ( x – 1 ) [ r2 + ( x – 2 ) q2 ( x ) ] + r2 ( x ) (2) + x ( x – 1 ) ( x – 2 ) q2 ( x ) (7.= ------------------2⋅1 2! Δ pn ( 0 ) . Then. 2. we obtain pn ( x ) = r0 + x [ r1 + ( x – 1 ) q1 ( x ) ] = r0 + r1 ( x ) (1) + x ( x – 1 ) q1 ( x ) (7.25) by ( x – 2 ) to obtain a quotient q 2 ( x ) and a remainder r 2 which turns out to be the constant term a 2 .24) into (7.19) (n – 2) Differencing again we obtain Δ p n ( x ) = 2 ⋅ 1a 2 + 3 ⋅ 2a 3 ( x ) 2 (1) + … + n ( n – 1 ) an ( x ) 2 (7.25).16) as p n ( x ) = r 0 + xq 0 ( x ) (7. and thus q1 ( x ) = r2 + ( x – 2 ) q2 ( x ) (7. we express (7. With this method we perform the following steps: 1.23) by ( x – 1 ) to obtain a quotient q 1 ( x ) and a remainder r 1 which turns out to be the constant term a 1 .for j = 0.Factorial Polynomials a1 = Δ pn ( 0 ) (7. We divide q 0 ( x ) in (7.16) by x to obtain a quotient q 0 ( x ) and a remainder r 0 which turns out to be the constant term a 0 .24) By substitution of (7.

[q2. Δ pn ( 0 ) r j = a j = -----------------j! (7. x % Computation of first quotient and remainder % Coefficients of second divisor.r2]=deconv(q1.22).30) or Δ p n ( 0 ) = j! r j j Example 7. then. i.e. [q3. [q4. [q0. construct the difference table with h = 1 . i. we obtain a new quotient whose degree is one less than preceding quotient and therefore.e. we must evaluate the remainders r 0.r3]=deconv(q2. we will use (7. x−1 % Computation of second quotient and remainder % Coefficients of third divisor. The MATLAB script is as follows: px=[1 −5 0 3 4].e.r4]=deconv(q3. d0=[1 0]. x−3 % Computation of fourth quotient and remainder % Coefficients of fifth (last) divisor.d3) d4=[1 −4].e. we will use the MATLAB deconv(p.q) function which divides the polynomial p by q. i. Solution: Since the highest power of the given polynomial p ( x ) is 4 . i.3 Express the algebraic polynomial p ( x ) = x – 5x + 3x + 4 4 3 (7. Third Edition Copyright © Orchard Publications .r1]=deconv(q0.28) to determine p n ( x ) . [q1.d2) d3=[1 −3].29) (7.d1) d2=[1 −2]. r 2.r0]=deconv(px. i. r 3 and r 4 . Then.d0) d1=[1 −1].Chapter 7 Finite Differences and Interpolation Continuing with the above procedure. r 1.d4) % Coefficients of given polynomial % Coefficients of first divisor.31) as a factorial polynomial. x−4 % Computation of fifth (last) quotient and remainder q0 = 7−8 Numerical Analysis Using MATLAB® and Excel®. We can compute the remainders by long division.28) and from (7. the process of finding new quotients and remainders terminates after ( n + 1 ) steps. x−2 % Computation of third quotient and remainder % Coefficients of fourth divisor. but for convenience. The general form of a factorial polynomial is pn ( x ) = r0 + r1 ( x ) (1) + r2 ( x ) (2) + … + rn – 1 ( x ) j (n – 1) + rn ( x ) (n) (7.16) and (7.e.

31). Numerical Analysis Using MATLAB® and Excel®.30) and (7. This can be easily done with the MATLAB collect(‘s_expr’) function. the MATLAB script is syms x. by expansion of the factorials using (7. the factorial polynomial is pn ( x ) = 4 – ( x ) (1) (2) (3) (4) -5 0 -4 0 -2 0 0 0 -4 0 3 0 4 -1 -8 1 – 8(x) + (x) + (x) (7. We will now compute the leading entries for the difference table using (7.32) We can verify that (7.32) is the same polynomial as (7.12). Third Edition Copyright © Orchard Publications 7−9 . For this example.Factorial Polynomials 1 r0 = 0 q1 = 1 r1 = 0 q2 = 1 r2 = 0 q3 = 1 r3 = 0 q4 = 0 r4 = 1 Therefore. where ‘s_expr’ is a symbolic expression. px=collect((x*(x−1)*(x−2)*(x−3))+(x*(x−1)*(x−2))−(8*x*(x−1))−x+4) px = x^4-5*x^3+3*x+4 We observe that this is the same algebraic polynomial as in (7.32).28). with reference to (7. Then.31).

7.6. 3. 2.33) 1. We enter the values of (7.8.Chapter 7 Finite Differences and Interpolation Δ p ( 0 ) = 0! ⋅ 4 = 4 Δ p ( 0 ) = 1! ⋅ ( – 1 ) = – 1 Δ p ( 0 ) = 2! ⋅ ( – 8 ) = – 16 Δ p ( 0 ) = 3! ⋅ 1 = 6 Δ p ( 0 ) = 4! ⋅ 1 = 24 Δ p ( 0 ) = 5! ⋅ 0 = 0 5 4 3 2 1 0 (7.6 Leading entries of (7. We obtain the next set of values by crisscross addition as shown in Table 7.33) in the appropriate spaces as shown in Table 7. The second crisscross addition extends the difference table as shown in Table 7. TABLE 7.7 Crisscross addition to find second set of values x p(x) 4 Δ Δ 2 Δ 3 Δ 4 Δ 5 −1 3 −16 −17 −10 30 24 6 24 0 7−10 Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications .33) in table form x p(x) 4 Δ −1 Δ 2 Δ 3 Δ 4 Δ 5 −16 6 24 0 TABLE 7.

This is shown in Table 7.9 Complete difference table for Example 7. Continuation of this procedure produces the complete difference table.Factorial Polynomials TABLE 7. Third Edition Copyright © Orchard Publications 7−11 .3 x p(x) 4 Δ Δ 2 Δ 3 Δ 4 Δ 5 −1 3 −16 −17 6 −14 −27 −41 −7 −48 67 19 −10 30 20 54 74 24 0 24 Numerical Analysis Using MATLAB® and Excel®. TABLE 7.9.8 Second crisscross addition to find third set of values x p(x) 4 Δ Δ 2 Δ 3 Δ 4 Δ 5 −1 3 −16 −17 6 −14 −27 −10 30 20 54 24 0 24 4.

by (7. Third Edition Copyright © Orchard Publications . we can find the antidifference of a factorial polynomial. In analogy with definite integrals for continuous functions. in finite differences we have the definite sum of p n ( x ) which for the interval a ≤ x ≤ a + ( n – 1 ) h is denoted as α + (n – 1)h x=α ∑ p n ( x ) = p n ( α ) + p n ( α + h ) + p n ( α + 2h ) + … + p n [ α + ( n – 1 ) h ] (7. We denote the antidifference as Δ p n ( x ) . in analogy with the fundamental theorem of integral calculus which states that 7−12 Numerical Analysis Using MATLAB® and Excel®.– 8 ------------. These are discussed below.37) where C is an arbitrary constant.34). its antidifference is (x) –1 (1) (x) (x) (x) .34) Example 7. It is computed from Δ (x) –1 –1 (n) (x) = -------------------(n + 1) (n + 1) (7. where we found that the corresponding factorial polynomial is pn ( x ) = 4 – ( x ) (1) – 8(x) (2) + (x) (3) + (x) (4) (7.3 Antidifferences We recall from elementary calculus that when we know the first derivative of a function. Antidifferences are very useful in finding sums of series.+ 4 ( x ) + C Δ p n ( x ) = -----------3 5 4 2 (5) (4) (3) (2) (7. By a similar method.– ------------.+ -----------.35) Solution: This is the same algebraic polynomial as that of the previous example.36) Then.4 Compute the antidifference of the algebraic polynomial p ( x ) = x – 5x + 3x + 4 4 3 (7. Before we present an example.38) Also. we can integrate or antidifferentiate to find the function. we need to review the definite sum and the fundamental theorem of sum calculus.Chapter 7 Finite Differences and Interpolation 7.

Antidifferences

∫a f ( x ) dx

α + (n – 1)h

b

= f(b) – f(a )

(7.39)

we have the fundamental theorem of sum calculus which states that

∑ x=α

pn ( x ) = Δ pn ( x )

–1

α + nh α

(7.40)

Example 7.5 Derive a simple expression, in closed form, that computes the sum of the cubes of the first n odd integers. Solution: An odd number can be expressed as 2m – 1 , and thus its cube is ( 2m – 1 ) . To use (7.40), we must express this term as a factorial polynomial. Recalling from (7.12) that

(x)

(n)

3

= x ( x – 1 ) ( x – 2 )… ( x – n + 1 )

(7.41)

**and using the MATLAB expand(f) function where f is a symbolic expression, we execute
**

syms m; f = (2*m−1)^3; expand(f)

**and we obtain ans = 8*m^3-12*m^2+6*m-1 Thus
**

p ( m ) = ( 2m – 1 ) = 8m – 12m + 6m – 1

3 3 2

(7.42)

**Following the procedure of Example 7.3, we find p n ( m ) with MATLAB as
**

pm=[8 −12 6 −1]; d0=[1 0]; [q0,r0]=deconv(pm,d0) d1=[1 −1]; [q1,r1]=deconv(q0,d1) d2=[1 −2]; [q2,r2]=deconv(q1,d2) d3=[1 −3]; [q3,r3]=deconv(q2,d3)

q0 = 8 -12 6

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

7−13

**Chapter 7 Finite Differences and Interpolation
**

r0 = 0 q1 = 8 r1 = 0 q2 = 8 r2 = 0 q3 = 0 r3 = 8 Therefore,

pn ( m ) = 8 ( m )

(3)

0 -4 0

0

-1

2

12

+ 12 ( m )

(2)

+ 2(m)

(1)

–1

(7.43)

**Taking the antidifference of (7.43) we obtain
**

–1 (m) + 2 ( m ) – ( m )( 1 ) ( m ) + 12 Δ pn ( m ) = 8 ------------------------------------------------------4 3 2 (4) (3) (2)

(7.44)

= 2(m)

(4)

+ 4(m)

(3)

+ (m)

(2)

– (m)

(1)

and with (7.40)

∑

Since

cubes = 2 ( m )

(4)

+ 4(m)

(3)

+ (m)

(2)

– (m)

(1)

n+1 m=1

= 2( n + 1)n(n – 1)(n – 2) + 4(n + 1)n(n – 1) + (n + 1)n – (n + 1) –2 ( 1 )

(4)

(7.45)

–4 ( 1 )

(4) (3) (2) (1)

(3)

– (1)

(2)

+ (1)

(1)

(1) (1) (1) (1)

= 1(1 – 1)(1 – 2 )(1 – 3 ) = 0 = 1(1 – 1)(1 – 2 ) = 0 = 1(1 – 1) = 0 = 1

(7.46)

relation (7.45) reduces to

∑ cubes

7−14

= 2(n + 1 )n(n – 1 )( n – 2 ) + 4( n + 1)n( n – 1) + ( n + 1 )n – (n + 1 ) + 1

(7.47)

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Newton’s Divided Difference Interpolation Method
**

and this can be simplified with the MATLAB collect(f) function as follows.

syms n; sum=collect(2*(n+1)*n*(n−1)*(n−2)+4*(n+1)*n*(n−1)+(n+1)*n−(n+1)+1)

sum = 2*n^4-n^2 that is,

∑ cubes

= 2n – n = n ( 2n – 1 )

4

2

2

2

(7.48)

We can verify that this is the correct expression by considering the first 4 odd integers 1, 3, 5, and 7 . The sum of their cubes is

1 + 27 + 125 + 343 = 496

**This is verified with (7.48) since
**

n ( 2n – 1 ) = 4 ( 2 ⋅ 4 – 1 ) = 16 ⋅ 31 = 496

2 2 2 2

One important application of finite differences is interpolation. Newton’s divided−difference interpolation method, Lagrange’s interpolation method, Gregory−Newton forward, and Gregory− Newton backward interpolation methods are discussed in Sections 7.4 through 7.7 below. We will use spreadsheets to facilitate the computations. Interpolation using MATLAB is discussed in Section 7.8 below.

**7.4 Newton’s Divided Difference Interpolation Method
**

This method, has the advantage that the values x 0, x 1, x 2, …, x n need not be equally spaced, or taken in consecutive order. It uses the formula

f ( x ) = f ( x 0 ) + ( x – x 0 ) f ( x 0, x 1 ) + ( x – x 0 ) ( x – x 1 ) f ( x 0, x 1, x 2 ) + ( x – x 0 ) ( x – x 1 ) ( x – x 2 ) f ( x 0, x 1, x 2, x 3 )

(7.49)

where f ( x 0, x 1 ) , f ( x 0, x 1, x 2 ) , and f ( x 0, x 1, x 2, x 3 ) are the first, second, and third divided differences respectively. Example 7.6 Use Newton’s divided−difference method to compute f ( 2 ) from the experimental data shown in Table 7.10.

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

7−15

**Chapter 7 Finite Differences and Interpolation
**

TABLE 7.10 Data for Example 7.6 x y = f (x) −1.0 3.0 0.0 −2.0 0.5 −0.375 1.0 3.0 2.5 16.125 3.0 19.0

Solution: We must compute the first, second, and third divided differences as required by (7.49). The first divided differences are:

– 2.000 – 3.000 ------------------------------------ = – 5.000 0 – ( – 1.0 ) – 0.375 – ( – 2.000 ) -------------------------------------------- = 3.250 0.5 – 0.0 3.000 – ( – 0.375 ) ---------------------------------------- = 6.750 1.0 – 0.5 16.125 – 3.000 ----------------------------------- = 8.750 2.5 – 1.0 19.000 – 16.125 -------------------------------------- = 5.750 3.0 – 2.5

(7.50)

**The second divided differences are:
**

3.250 – ( – 5.000 ) ---------------------------------------- = 5.500 0.5 – ( – 1.0 ) 6.750 – 3.250 -------------------------------- = 3.500 1.0 – 0.0 8.750 – 6.750 -------------------------------- = 1.000 2.5 – 0.5 5.750 – 8.750 -------------------------------- = – 1.500 3.0 – 1.0

(7.51)

**and the third divided differences are:
**

3.500 – 5.500 -------------------------------- = – 1.000 1.0 – ( – 1.0 ) 1.000 – 3.500 -------------------------------- = – 1.000 2.5 – 0.0 – 1.500 – 1.000 ------------------------------------ = – 1.000 3.0 – 0.5

(7.52)

With these values, we construct the difference Table 7.11.

7−16

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Lagrange’s Interpolation Method
**

TABLE 7.11 Difference table for Example 7.6

1st Divided Difference 2nd Divided Difference 3rd Divided Difference

x −1.0 0.0 0.5 1.0 2.5 3.0

f (x) 3.000 −2.000

f ( x 0, x 1 ) −5.000

f ( x 0, x 1, x 2 )

f ( x 0, x 1, x 2, x 3 )

5.500 3.250 −1.000 3.500 6.750 −1.000 1.000 8.750 −1.000 −1.500 5.750

−0.375 3.000 16.125 19.000

Now, we have all the data that we need to find f ( 2 ) . We start with x 0 = 0.00 ,* and for x in (7.49), we use x = 2 . Then,

f ( 2 ) = – 2.0 + ( 2 – 0 ) ( 3.250 ) + ( 2 – 0 ) ( 2 – 0.5 ) ( 3.500 ) + ( 2 – 0 ) ( 2 – 0.5 ) ( 2 – 1 ) ( – 1.000 ) = – 2.0 + 6.5 + 10.5 – 3 = 12

This, and other interpolation problems, can also be solved with a spreadsheet. The Excel spreadsheet for this example is shown in Figure 7.1.

**7.5 Lagrange’s Interpolation Method
**

Lagrange’s interpolation method uses the formula

( x – x 1 ) ( x – x 2 )… ( x – x n ) ( x – x 0 ) ( x – x 2 )… ( x – x n ) - f ( x 0 ) + -----------------------------------------------------------------------f(x ) f ( x ) = ----------------------------------------------------------------------( x 0 – x 1 ) ( x 0 – x 2 )… ( x 0 – x n ) ( x 1 – x 0 ) ( x 1 – x 2 )… ( x 1 – x n ) 1 ( x – x 0 ) ( x – x 1 )… ( x – x n – 1 ) - f(x ) + ----------------------------------------------------------------------------( x n – x 0 ) ( x n – x 2 )… ( x n – x n – 1 ) n

(7.53)

and, like Newton’s divided difference method, has the advantage that the values x 0, x 1, x 2, …, x n need not be equally spaced or taken in consecutive order.

* We chose this as our starting value so that f ( 2 ) will be between f ( 1 ) and f ( 2.5 )

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

7−17

Chapter 7 Finite Differences and Interpolation

Interpolation with Newton's Divided Difference Formula f(x) = f(x0)+(x-x0)f(x0,x1)+(x-x0)(x-x1)f(x0,x1,x2)+(x-x0)(x-x1)(x-x2)f(x0,x1,x2,x3) In this example, w e w ant to evaluate f(x) at x= 2 1st divided difference x -1.00 f(x) 3.000 -5.000 0.00 -2.000 3.250 0.50 -0.375 6.750 1.00 3.000 8.750 2.50 16.125 5.750 3.00 19.000 We use the above formula w ith starting value x0.00 f(2)=B12+(E3-E18)*C13+(E3-E18)*(E3-A14)*D14+(E3-E18)*(E3-A14)*(E3-A16)*E15 or f(2)= 12.00 The plot below verifies that our answ er is correct -1.000 3.000

20 15

2nd divided 3rd divided difference f(x0, x1, x2) difference f(x0,x1,x2,x3)

f(x0, x1)

5.500 -1.000 3.500 -1.000 1.000 -1.000 -1.500

0.000 -2.000 f(x) 0.500 -0.375 1.000 3.000 2.500 16.125 3.000 19.000

10 5 0 -5 -1.0 0.0 1.0 2.0 3.0

x

Figure 7.1. Spreadsheet for Example 7.6

Example 7.7 Repeat Example 7.6 using Lagrange’s interpolation formula. Solution: All computations appear in the spreadsheet of Figure 7.2 where we have used relation (7.53).

7−18

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Gregory−Newton Forward Interpolation Method
**

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 A B C

2 f(x) 3.000 -2.000 -0.375 3.000 16.125 19.000 x-x1 2.000 x0-x1 -1.000 x-x0 3.000 x1-x0 1.000 x-x0 3.000 x2-x0 1.500 x-x0 3.000 x3-x0 2.000 x-x0 3.000 x4-x0 3.500 x-x0 3.000 x5-x0 4.000 x-x2 1.500 x0-x2 -1.500 x-x2 1.500 x1-x2 -0.500 x-x1 2.000 x2-x1 0.500 x-x1 2.000 x3-x1 1.000 x-x1 2.000 x4-x1 2.500 x-x1 2.000 x5-x1 3.000 x-x3 1.000 x0-x3 -2.000 x-x3 1.000 x1-x3 -1.000 x-x3 1.000 x2-x3 -0.500 x-x2 1.500 x3-x2 0.500 x-x2 1.500 x4-x2 2.000 x-x2 1.500 x5-x2 2.500 x-x4 -0.500 x0-x4 -3.500 x-x4 -0.500 x1-x4 -2.500 x-x4 -0.500 x2-x4 -2.000 x-x4 -0.500 x3-x4 -1.500 x-x3 1.000 x4-x3 1.500 x-x3 1.000 x5-x3 2.000 x-x5 -1.000 x0-x5 -4.000 x-x5 -1.000 x1-x5 -3.000 x-x5 -1.000 x2-x5 -2.500 x-x5 -1.000 x3-x5 -2.000 x-x5 -1.000 x4-x5 -0.500 x-x4 -0.500 x5-x4 0.500 f(x0) 3.000

D

E

F

G

H

I

J

Numer. Partial Prods 4.500

K

Denom. Partial Prods

L

Division of Partial Prods

Lagrange's Interpolation Method Interpol. at x= x -1.00 0.00 0.50 1.00 2.50 3.00

x0 x1 x2 x3 x4 x5

-0.107 -42.000 f(x1) -2.000 -4.500 -1.200 3.750 f(x2) -0.375 -1.125 0.600 -1.875 f(x3) 3.000 13.500 4.500 3.000 f(x4) 16.125 -145.125 11.057 -13.125 f(x5) 19.000 -85.500 -2.850 30.000 f(2)= Sum= 12

Figure 7.2. Spreadsheet for Example 7.7

**7.6 Gregory−Newton Forward Interpolation Method
**

This method uses the formula

r(r – 1) 2 r( r – 1 )( r – 2) 3 - Δ f 0 + --------------------------------- Δ f0 + … f ( x ) = f 0 + r Δ f 0 + ----------------2! 3!

2 3

(7.54)

where f 0 is the first value of the data set, Δ f 0 , Δ f 0 , and Δ f 0 are the first, second, and third forward* differences respectively. The variable r is the difference between an unknown point x and a known point x 1 divided by the interval h , that is,

( x – x1 ) r = -----------------h

(7.55)

* This is an expression to indicate that we use the differences in a forward sequence, that is, the first entries on the columns

where the differences appear.

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

7−19

**Chapter 7 Finite Differences and Interpolation
**

The formula of (7.54) is valid only when the values x 0, x 1, x 2, …, x n are equally spaced with interval h . It is used to interpolate values near the smaller values of x , that is, the values near the beginning of the given data set. The formula that we will study on the next section, is used to interpolate values near the larger values of x , that is, the values near the end of the given data set. Example 7.8 Use the Gregory−Newton forward interpolation formula to compute f ( 1.03 ) from the following data.

TABLE 7.12 Table for Example 7.8 x y = f (x) 1.00 1.000000 1.05 1.257625 1.10 1.531000 1.15 1.820875 1.20 2.128000 1.25 2.453125

Solution: We enter the given x and f ( x ) values in a difference table; then, we compute the first, second, and third differences. These are not divided differences and therefore, we simply subtract the second value of f ( x ) from the first, the third from the second, and so on, as shown in Table 7.13. For this example,

f 0 = f ( 1.00 ) = 1.000000 h = x 1 – x 0 = 1.05 – 1.00 = 0.05 x – x1 1.03 – 1.00 - = -------------------------- = 0.60 r = ------------h 0.05

(7.56)

**and with the values shown in Table 7.13 and using (7.54), we obtain
**

( 0.60 ) ⋅ ( 0.60 – 1 ) f ( 1.03 ) = 1.000000 + ( 0.60 ) ⋅ ( 0.257625 ) + ------------------------------------------2! ( 0.60 ) ⋅ ( 0.60 – 1 ) ( 0.60 – 2 ) + -------------------------------------------------------------------- ⋅ ( 0.000750 ) = 1.152727 3!

(7.57)

The spreadsheet of Figure 7.3 shows the layout and computations for this example.

7−20

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Gregory−Newton Backward Interpolation Method
**

TABLE 7.13 Difference table for Example 7.8 1st Difference x 1.00 1.05 1.10 1.15 1.20 1.25 f (x) 1.000000 1.257625 0.273375 1.531000 0.289875 1.820875 0.307125 2.128000 0.325125 2.453125 0.018000 0.017250 0.000750 0.016500 0.000750 f ( x 0, x 1 ) 0.257625 0.015750 0.000750 2nd Difference f ( x 0, x 1, x 2 ) 3rd Difference f ( x 0, x 1, x 2, x 3 )

**7.7 Gregory−Newton Backward Interpolation Method
**

This method uses the formula

r(r + 1) 2 r(r + 1 )(r + 2 ) 3 - Δ f – 2 + ---------------------------------- Δ f–3 + … f ( x ) = f 0 + r Δ f – 1 + ----------------2! 3!

2 3

(7.58)

where f 0 is the first value of the data set, Δ f –1 , Δ f –2 , and Δ f –3 are the first, second and third backward differences, and

( x – x1 ) r = -----------------h

**Expression (7.58) is valid only when the values x 0, x 1, x 2, …, x n are equally spaced with interval
**

h . It is used to interpolate values near the end of the data set, that is, the larger values of x .

Backward interpolation is an expression to indicate that we use the differences in a backward sequence, that is, the last entries on the columns where the differences appear. Example 7.9 Use the Gregory−Newton backward interpolation formula to compute f ( 1.18 ) from the data set of Table 7.14.

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

7−21

**Chapter 7 Finite Differences and Interpolation
**

Gregory-Newton Forward Interpolation Method See expressions (7.54) and (7.55) Interpolate f(x) at x= x f(x) Δf 0.257625 1.05 1.257625 0.273375 1.10 1.531000 0.289875 1.15 1.820875 0.307125 1.20 2.128000 0.325125 1.25 2.453125 h= A10-A8= 0.05 0.6 r= (D5-A8)/C20= 0.018000 0.017250 0.000750 0.016500 0.000750 0.015750 0.000750 1.03 Δ 2f Δ 3f

1.00 1.000000

f(1.12)= B8+F20*C9+(F20*(F20-1)*D10)/FACT(2)+(F20*(F20-1)*(F20-2)*E11)/FACT(3) = 1.152727 1.00 1.000000 1.05 1.257625 1.10 1.531000 1.15 1.820875 1.20 2.128000 1.25 2.453125 3.00 2.50 2.00 1.50 1.00 0.50 0.00 1.00 1.05 1.10 1.15 1.20 1.25

Figure 7.3. Spreadsheet for Example 7.8 TABLE 7.14 Data for Example 7.9 x y = f (x) 1.00 1.000000 1.05 1.257625 1.10 1.531000 1.15 1.820875 1.20 2.128000 1.25 2.453125

7−22

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Gregory−Newton Backward Interpolation Method
**

Solution: We arbitrarily choose f 0 = 2.128000 as our starting point since f ( 1.18 ) lies between f ( 1.15 ) and

f ( 1.20 ) . Then, h = 1.20 – 1.15 = 0.05

and

r = ( x – x 1 ) ⁄ h = ( 1.18 – 1.20 ) ⁄ 0.05 = – 0.4

Now, by (7.58) we have:

( – 0.4 ) ( – 0.4 + 1 ) - ( 0.01725 ) f ( 1.18 ) = 2.128 + ( – 0.4 ) ( 0.307125 ) + ----------------------------------------2! ( – 0.4 ) ( – 0.4 + 1 ) ( – 0.4 + 2 ) + -------------------------------------------------------------------- ( 0.00075 ) = 2.003032 3!

The computations were made with the spreadsheet of Figure 7.4. If the increments in x values are small, we can use the Excel VLOOKUP function to perform interpolation. The syntax of this function is as follows.

VLOOKUP(lookup_value, table_array, col_index_num, range lookup)

where:

lookup_value is the value being searched in the first column of the lookup table table_array are the columns forming a rectangular range or array col_index_num is the column where the answer will be found range lookup is a logical value (TRUE or FALSE) that specifies whether we require VLOOKUP to find an exact or an approximate match. If TRUE is omitted, an approximate match is returned.

In other words, if an exact match is not found, the next largest value that is less than the lookup_value is returned. If FALSE is specified, VLOOKUP will attempt to find an exact match, and if one is not found, the error value #N/A will be returned.

A sample spreadsheet is shown in Figure 7.5 where the values of x extend from −5 to +5 volts. Only a partial table is shown.

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

7−23

**Chapter 7 Finite Differences and Interpolation
**

Gregory-New ton Backw ard Interpolation Method See form ula 7.58 Interpolate f(x) at x= 1.18 x 1.00 1.05 1.10 1.15 1.20 f(x) 1.000000 0.257625 1.257625 0.273375 1.531000 0.289875 1.820875 0.307125 2.128000 0.05 r= (C5-A16)/C18= -0.4 0.017250 0.016500 0.000750 0.015750 0.000750 Δf Δ 2f Δ 3f

h= A16-A14=

f(1.18)= B16+F18*C15+(F18*(F18+1)*D14)/FACT(2)+(F18*(F18+1)*(F18+2)*E13)/FACT(3) = 1.00 1.05 1.10 1.15 1.20 2.003032 2.500 1.000000 2.000 1.257625 1.531000 1.820875 2.128000 1.500 1.000 0.500 0.000 1.00

1.05

1.10

1.15

1.20

Figure 7.4. Spreadsheet for Example 7.9

**7.8 Interpolation with MATLAB
**

MATLAB has several functions that perform interpolation of data. We will study the following: 1. interp1(x,y,xi) performs one dimensional interpolation where x and y are related as y = f(x) and xi is some value for which we want to find y(xi) by linear interpolation, i.e., “table lookup”. This command will search the x vector to find two consecutive entries between which the desired value falls. It then performs linear interpolation to find the corresponding value of y. To obtain a correct result, the components of the x vector must be monotonic, that is, either in ascending or descending order.

7−24

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

0302 -1.725 -0.0295 -1.900 -0.875 -0.0274 -1.8375.750 -0.030927 -0.2) = =VLOOKUP(-1. Third Edition Copyright © Orchard Publications 7−25 . Using the Excel VLOOKUP function for interpolation 2.1496775 =VLOOKUP(4.0305803 -0.xi.2) = 0.A2:B282.0309266 -1.0285 -1.y.y.0323 -1.625 -0.0277 -1.05 0.8500 -0.0288 -1.0313 -1.5750 B 0.675 -0. The default is linear.925 -0. Numerical Analysis Using MATLAB® and Excel®.xi) where the string method allows us to specify one of the methods listed below. interp1(x.z.y.0281 -1.0267 -1. this requires equidistant values of x 3.950 -0.A2:B282.8375.z.xi) but performs two dimensional interpolation.800 -0.1484323 Figure 7.0306 -1.550 -0.0326 -1.0270 -1.8250 -0.10 0.775 -0.y. interp2(x. interp2(x. this is the default interpolation spline − cubic spline interpolation.0320 -1.yi) is similar to interp1(x.05 -2 -1 0 1 2 3 4 5 -2.’method’) is similar to interp1(x.850 -0.’method’) performs the same operation as interp1(x.’method’) but performs two dimensional interpolation.0330 -1.0309 -1.xi.030927 #N/A =VLOOKUP(-1.y. We will illustrate the applications of these functions with the examples below.1484323 0.A2:B282.2.525 -0.700 -0.00 -0.yi.650 -0.5535. The spline method does not apply to two dimensional interpolation.5.xi. nearest − nearest neighbor interpolation linear − linear interpolation.20 0. 4.975 -0.A2:B282.0316 -1.575 -0.5500 4.000 -0. this does also extrapolation cubic − cubic interpolation.0263 A 8 9 B -1.8375.xi.TRUE) = =VLOOKUP(-1.FALSE) = A 264 265 4.15 0.y.0299 -1.600 -0.825 -0.Interpolation with MATLAB V I 0.0292 -1.2.

1. plot(v. the corresponding values in milliamperes.265 is %2. % Define "v_a" as a two−column matrix to display volts % and amperes side−by−side.xi) command to interpolate at the desired value.0297 -0.59) where v is in volts and i in milliamperes.a.0000 -1.2 .1(exp(0. % Display values of volts and amps below the heading ma=interp1(v. xlabel('voltage (volts)'). % v=linspace(−2. % We can use the MATLAB function 'interp1' to linearly interpolate % the value of milliamps for any value of v within the specified interval. accordingly. plot i versus v in this range. % Linear (default) interpolation fprintf('current (in milliamps) @ v=1.a]'.* v)−1).265 volts. Third Edition Copyright © Orchard Publications .2v)−1).0262 -0. and plots the data for this range.1 ( e 0.* (exp(0. number_of_values) command.2v ( t ) – 1) (7. 30).a).5172 -1. title('volt−ampere characteristics for a junction diode'). % Heading of the two−column matrix fprintf(' \n'). Compute i for 30 data points of v within the interval – ( 2 ≤ v ≤ 5 ) .0225 7−26 Numerical Analysis Using MATLAB® and Excel®. fprintf(' volts milliamps \n'). % This script is for Example_7_10. and using linear interpolation compute i when v = 1. Solution: We are required to use 30 data points within the given range.7586 -1.m % It computes the values of current (in milliamps) vs. we will use the MATLAB linspace(first_value.Chapter 7 Finite Differences and Interpolation Example 7.265). voltage (volts) % for a diode whose v−i characteristics are i=0. % Specify 30 intervals in the −2<=v<=5 interval a=0. ylabel('current (milliamps)').y. we use the interp1(x. grid. % We use "a" for current instead of "i" to avoid conflict % with imaginary numbers v_a=[v.2759 milliamps -0.0330 -0. 5. The script below produces 30 values in volts.4f \n'.1. volts -2. Then. last_value. disp(v_a). ma) The data and the value obtained by interpolation are shown below.10 The i – v (current−voltage) relation of a non−linear electrical device is given by i ( t ) = 0.

2759 4. Third Edition Copyright © Orchard Publications 7−27 .6552 0.1034 2.0345 4.0598 0.0847 0.3448 2.0140 0.3103 3.0690 0.0187 -0.1035 0.1135 0. Numerical Analysis Using MATLAB® and Excel®.0104 -0.0196 0.1241 0.5172 4.0318 0.0690 3.7931 -0.0383 0.3103 -0.8966 1.0288 The plot for this example is shown in Figure 7.8276 3.4138 0.Interpolation with MATLAB -1.0000 -0.1724 0.1590 0.5517 -0.7586 5.7931 4.0147 -0.1468 0.0035 0.1379 1.0451 0.0760 0.5862 2.0345 -0.0523 0.6207 1.1718 current (in milliamps) @ v=1.0014 0.3793 1.0677 0.1352 0.5517 3.0256 0.6.8621 2.0060 -0.265 is 0.0939 0.0086 0.

60) in the interval 0 ≤ x ≤ 2 π with 120 inte rme diate values. 'cubic'). Compare the values obtained with the linear.^ 5. 'linear').y. % We need these two y=(cos(x)) .05 0 -0. for comparison with the linear. % plot(x. cubic. % linear_int=interp1(x.2 volt-ampere characteristics for a junction diode 0. use the MATLAB interp1(x. xlabel('x').11 Plot the function y = f ( x ) = cos 5x (7. cubic. title('y=cos^5(x)').6.60) and produces the values of analytical values.y. with the analytical values. grid. % This is the script for Example_7_11 % x=linspace(0.1 0. Plot for Example 7.’method’) function to interpolate at π ⁄ 8 .15 current (milliamps) 0.[pi/8 pi/4 3*pi/5 3*pi/7]'. Third Edition Copyright © Orchard Publications .y. 3 π ⁄ 5 . 7−28 Numerical Analysis Using MATLAB® and Excel®.120). % statements for the plot % analytic=(cos([pi/8 pi/4 3*pi/5 3*pi/7]'). and spline methods.Chapter 7 Finite Differences and Interpolation 0.[pi/8 pi/4 3*pi/5 3*pi/7]'.^ 5). Solution: The script below plots (7. and spline interpolation methods. π ⁄ 4 . ylabel('y'). and 3 π ⁄ 7 .xi.2*pi.y).10 Example 7.05 -2 0 2 voltage (volts) 4 6 Figure 7. Then. % The label 'linear' on the right side of the above statement % could be have been omitted since the default is linear % cubic_int=interp1(x.

'spline')./ analytic.2)=linear_int.5f\t %8.4).5f\t %8. % 2nd column of matrix z(:. % % The statements below compute the percent error for the three % interpolation methods as compared with the exact (analytic) values % error1=(linear_int−analytic).y') fprintf(' \n').3)=cubic_int.y. fprintf('%8. fprintf('Linear Int \t Cubic Int \t Spline Int \n') fprintf(' \n').7.*100 . error2=(cubic_int−analytic).*100 . fprintf('%8.3)=error3.4)=spline_int.3). error3=(spline_int−analytic).*100 .2)=error2.[pi/8 pi/4 3*pi/5 3*pi/7]'./ analytic.5f\n'. Third Edition Copyright © Orchard Publications 7−29 .1)=error1.1)=analytic. The plot for the function of this example is shown in Figure 7. % y=zeros(4. % Insert line fprintf('Analytic \t Linear Int \t Cubic Int \t Spline Int \n') fprintf(' \n').5f\t %8. % Insert line disp('The percent errors for each interpolation method are:') fprintf(' \n'). % 4th column of matrix fprintf(' \n'). % 3rd column of matrix % fprintf(' \n').5f\n'. % 2nd column of matrix y(:.5f\t %8./ analytic.% Construct a 4 x 4 matrix of zeros y(:.5f\t %8.z') fprintf(' \n'). % z=zeros(4. Numerical Analysis Using MATLAB® and Excel®. % 1st column of matrix y(:. % 3rd column of matrix y(:. % Construct a 4 x 3 matrix of zeros z(:.Interpolation with MATLAB % spline_int=interp1(x. % 1st column of matrix z(:.

4 -0.12 For the impedance example of Section 1.00012 -0.4 0.17678 -0.09681 13.6 -0.00184 -0.17678 -0.Chapter 7 Finite Differences and Interpolation 1 0.67274 0.00281 0.67310 0.00296 0. use the spline method of interpolation to find the magnitude of the impedance at ω = 792 rad ⁄ s .00282 0.67310 0. Solution: % The file is Example_7_12. Plot the function of Example 7.00062 Cubic Int 0.00002 0.00055 Linear Int 0.22707 5.67311 0.m % It calculates and plots the impedance Z(w) versus radian frequency w.7.11 The analytical and interpolated values are shown below for comparison.2 0 -0.64706 Spline Int 0.00011 -0.07445 7−30 Numerical Analysis Using MATLAB® and Excel®.17718 -0. Third Edition Copyright © Orchard Publications .7 in Chapter 1 whose script and plot are shown below.00055 The percent errors for each interpolation method are: Linear Int -0.17678 -0. Cubic Int 0.05211 0.00054 Spline Int 0.8 0.6 0.01027 -0.00282 0.40465 -0.8 -1 y y=cos 5(x) 0 1 2 3 x 4 5 6 Figure 7.27678 Example 7. Analytic 0.2 -0.

.*10.122.588 67.Interpolation with MATLAB % % Use the following five statements to obtain |Z| versus radian frequency w w=300:100:3000.^6. fprintf(' \n') fprintf('Magnitude of Z at w=792 rad/s is %6.088 73. Radian Frequency 1200 1000 800 |Z| in Ohms 600 400 200 0 2 10 10 w in rads/sec 3 10 4 Figure 7.8. z(:.111 90./w).2)=(10+(10.'spline').182 436. 469. Radian Frequency').437 222.286 44. % zi=interp1(w.3f Ohms \n'.513 62. 42.339 52.abs(z)') % w=[300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500.056 1014. 58.8. z=zeros(28.792. fprintf('%2..^4−j. title('Magnitude of Impedance vs.353. zi) fprintf(' \n') The plot for the function of this example is shown in Figure 7...z...032 187..432 38.751 120..789 71..*(0.104 97.12 Numerical Analysis Using MATLAB® and Excel®. ylabel('|Z| in Ohms').240 54. 2400 2500 2600 2700 2800 2900 3000].481.*w−10. grid.z).603 81. 103. Magnitude of Impedance vs. semilogx(w..^5. xlabel('w in rads/sec'). 1600 1700 1800 1900 2000 2100 2200 2300.830 266..665 140.717 46./(10+j. Plot for the function of Example 7.0f\t %10.3f\n'.1)=w'.611 51..052 145. z=[39.468 48. Third Edition Copyright © Orchard Publications 7−31 .182 40..2). z(:.845]./w)))'.1.938..

ylabel('y'). zlabel('z'). we will review the following commands which can be used for two−dimensional interpolation. % eps prevents division by zero mesh(X. % Create X and Y matrices R=sqrt(X. % y must have same number of points as x [X. This function is the equivalent of the function y = sin x ⁄ x in two dimensions. and the plot in the intervals – 2 π ≤ x ≤ 2 π and – 2 π ≤ y ≤ 2 π .Chapter 7 Finite Differences and Interpolation MATLAB interpolates the impedance at ω = 792 rad ⁄ s and displays the following message: Magnitude of Z at w=792 rad/s is 217.13 Generate the plot of the function sin R Z = ----------R (7. is R = X +Y 2 2 (7.Y]=meshgrid(x.034 Ohms Two−dimensional plots were briefly discussed in Chapter 1. y 0 ) = ( 0. % Define interval in increments of pi/24 y=x. y . Using these points. and z . Example 7. 1. % Generate mesh plot for Z=sin(R)/R xlabel('x'). we can form a function z = f ( x. For convenience.y) − Generates interpolation arrays which contain all combinations of the x and y points which we specify. mesh(Z) − Plots the values in the matrix Z as height values above a rectangular grid. Then.^ 2 + Y. % This is the script for Example_7_13 x=−2*pi: pi/24: 2*pi. Third Edition Copyright © Orchard Publications . 0 ) . Y ] matrices that form a rectangular grid of points in the x – y plane. and connects adjacent points to form a mesh surface. [X. we write and execute the following MATLAB script. X and Y comprise a pair of matrices representing a rectangular grid of points in the x – y plane. Y ] matrices.Y. Here.61) in three dimensions x .^ 2). y ) where z is a matrix.Y]=meshgrid(x.Z). 2. 7−32 Numerical Analysis Using MATLAB® and Excel®. R is a matrix that contains the distances from the origin to each point in the pair of [ X. Solution: The matrix R that contains the distances from the origin to each point in the pair of [ X./ (R+eps).0) to x−y points Z=sin(R).62) We let the origin be at ( x 0.y). % Compute distances from origin (0.

Figure 7. z_int=interp2(X. y . Then.14 Generate the plot of the function z = x + y – 3xy 3 3 (7. and the plot in the intervals – 10 ≤ x ≤ 10 and – 10 ≤ y ≤ 10 .14').2.^3+Y. Plot for Example 7.9. % Define interval in increments of 0. fprintf(' \n') Numerical Analysis Using MATLAB® and Excel®.Y. y 0 ) = ( 0.13 Example 7. title('Plot for the Function of Example 7.25: 10. % y must have same number of points as x [X. Solution: We let the origin be at ( x 0. 0 ) . % This is the script for Example_7_14 x=−10: 0. −1. ylabel('y').Interpolation with MATLAB title('Plot for the Three−dimensional sin(R) / R Function') The plot for the function of this example is shown in Figure 7.63) in three dimensions x . zlabel('z').Z.9. mesh(X.*Y. and z .Z). we write and execute the following script.^3−3. Use the cubic method to interpolate the value of z at x = – 1 and y = 2 . Third Edition Copyright © Orchard Publications 7−33 .25 y=x.y). % Generate mesh plot xlabel('x').*X.'cubic').Y. % Create X and Y matrices Z=X.Y]=meshgrid(x.

17 500.10.15 500.12 500.13 500.09 500.14 500.19 500.13 500.11 500.00 Example 7.14 500.08 500.12 500.12 500.02 500.17 500.17 500.13 500. Plot for Example 7.08 500.08 500.12 500.14 500.09 500.16 500.09 500.17 500.2f \n'.17 500.11 500.17 This rectangular land parcel is 175 meters wide and 275 meters deep.z_int) fprintf(' \n') The plot for the function of this example is shown in Figure 7.15 500.09 500.08 500.18 500.17 500.18 500.13 500.11 500.11 500.14 500.10.14 500.12 500.16 500.09 500.10 500.14 500.12 500.09 500.15 500.18 500.05 500.Chapter 7 Finite Differences and Interpolation fprintf('Interpolated Value of z at x = −1 and y = 2 is z = %4.15 500.19 500.02 500.12 500.13 500.15 500.18 500.16 500.08 500.17 500.01 500.09 500. The measurements shown above were made at points 25 meters apart.13 500.16 500.12 500.02 500.03 500.14 500.14 500.15 500.02 500.17 500.11 500.15 A land surveyor measured and recorded the data below for a rectangular undeveloped land which lies approximately 500 meters above sea level.12 500. 500.05 500.05 500.14 500.14 500. Third Edition Copyright © Orchard Publications .12 500.13 500.01 500.13 500.14 500.09 500.11 500.15 500.21 500.14 Interpolated Value of z at x = -1 and y = 2 is z = 13.09 500.15 500.15 500. Figure 7.10 500.11 500.13 500.01 500.14 500.15 500.11 500. 7−34 Numerical Analysis Using MATLAB® and Excel®.

ylabel('y−axis.09 500.13 500. Interpolate the value of z at x = 108 m. disp('Interpolated z is:'). 500.12.15 500.14 500.08 500.02.177.14 500.17]. 500.09 500. Denoting the width as the x – axis .yi). xi=0: 2.08 500.13 500.03 500.01 500. Third Edition Copyright © Orchard Publications 7−35 .13 500. plot the given data to form a rectangular grid.14 500.09 500. zlabel('Height.14 500.14 500.12 500.14 500.15.108.01 500.09.18 500. grid off. % Make x−axis finer % size(xi).21 500.y). % Make y−axis finer disp('size(yi)').11 500.13 500.12 500. title('Parcel map') % The pause command below stops execution of the program for 10 seconds % so that we can see the mesh plot pause(10).09 500. axis([0 175 0 275 500 502]). size(xi) = 1 71 disp('size(xi)'). % mesh(x.15. the depth as the y – axis and the height as the z – axis . % This script is for Example_7_15 % x=0: 25: 175.17 500.12.05 500.12 500.17 500.16 500.18 500. Solution: The MATLAB script and plot are shown below and explanations are provided with comment statements.08 500. box off xlabel('x−axis. 500.19 500.13 500.14 500.18 500.12 500.14 500. 500.yy]=meshgrid(x.01 500.16 500.11 500. z_int=interp2(x. meters above sea level').5: 275. 500.11 500.16 500. % y−axis varies down the columns of z z=[500.11.12 500.10. c.z.5: 175.13 500.z).14 500.02 500.11 500.17 500.05 500.17 500.yyi]=meshgrid(xi. % x−axis varies across the rows of z y=0: 25: 275.15 500.09 500.09 500. % Forms grid of all combinations of xi and yi Numerical Analysis Using MATLAB® and Excel®. 500.10 500.08 500.Interpolation with MATLAB a. m').'cubic').16 500.12 500. 500.15 500. 500. Compute the maximum height and its location on the x – y plane.12 500. m').17 500.11 500. size(xi) yi=0: 2.17 500.18 500. % Returns a row vector containing the size of xi where the % first element denotes the number of rows and the second is the number % of columns.y.15 500.13 500.08 500.y.09 500.15 500. 500.17 500.12 500.14.13 500.14 500.17 500. Here.11 500. z_int [xx.13 500. b.02 500.15.05 500.02 500.19 500. size(yi) [xxi.11. and y = 177 m. 500. 500.15 500.09 500.15 500.14 500.

Likewise max(zzi) returns a row vector which contains the % maxima of the columns in zzi. % Grid with original data plot3(xx.j]=find(A>2) % returns %i= % % 2 % 3 % 1 % 3 % % %j= % % 2 % 2 % 3 % 3 % That is.yyi.y.yyi. disp('size(yyi)').n]=find(zmax==zzi) %m= 7−36 Numerical Analysis Using MATLAB® and Excel®. meters above sea level'). disp('size(zzi)').yy]=meshgrid(x. size(zzi) size(xxi). % A=[a11 a12 a13. −2 5 6].z. axis([0 175 0 275 500 503]).yy.z. a21 a22 a23.xxi. title('Map of Rectangular Land Parcel') hold off.y). size(zzi) zzi=interp2(x. % max(x) returns the largest element of vector x % max(A) returns a row vector which contains the maxima of the columns % in matrix A. box off xlabel('x−axis. a13=3 and a33=6 % satisfy the condition A>2 % The == operator compares two variables and returns ones when they % are equal. % Plot smoothed data hold on. [xx. % [i. zlabel('Height. Third Edition Copyright © Orchard Publications . the elements a22=3.zzi). % Cubic interpolation − interpolates % all combinations of xxi and yyi and constructs the matrix zzi mesh(xxi.Chapter 7 Finite Differences and Interpolation % size(xxi) = size(yyi) = size(zzi) = 111 71 disp('size(xxi)'). and zeros when they are not equal % [m. size(yyi). m'). a32=5.'*k').'cubic'). Observe that size(max(zzi)) = 1 71 % and size(max(max(zzi))) = 1 1 zmax=max(max(zzi)) % Estimates the peak of the terrain % The 'find' function returns the subscripts where a relational expression % is true. 2 3 −4. For Example. size(yyi). grid off. size(xxi). ylabel('y−axis. a31 a32 a33] or % A=[−1 0 3. m').

size(xi) = 1 71 % and the y−coordinate is found from ymax=yi(m) % ymax = % % 3. size(yi) = 1 111 % Remember that i is the row index.Interpolation with MATLAB % % 65 % %n= % % 36 % % that is. % Cubic interpolation − interpolates % all combinations of xxi and yyi and constructs the matrix zzi Numerical Analysis Using MATLAB® and Excel®.7500 % Column 36. and x−axis % varies across the rows of z and y−axis varies down the columns of z Interpolated z is: z_int = 500.xxi. j is the column index.z.2000 % Row 65.y.1492 size(xi) ans = 1 size(yi) 71 ans = 1 size(xxi) 111 ans = 111 size(yyi) 71 ans = 111 71 zzi=interp2(x.yyi. zmax is located at zzi = Z(65)(36) % % the x−cordinate is found from xmax=xi(n) % xmax = % % 1.'cubic'). Third Edition Copyright © Orchard Publications 7−37 .

15 7−38 Numerical Analysis Using MATLAB® and Excel®.Chapter 7 Finite Differences and Interpolation size(zzi) ans = 111 zmax = 500. Third Edition Copyright © Orchard Publications .11. The plot is shown in Figure 7. Plot for Example 7. Figure 7.192 .21 where the x and y coordinates are x = 20 and y = 275 .2108 m = 111 n= 9 xmax = 20 ymax = 275 71 zmax=max(max(zzi)) % Estimates the peak of the terrain These values indicate that z max = 500.11. The interpolated value of z at x = 108 m and y = 177 m is z = 500.

9 Summary • The first divided difference is defined as: f ( xi ) – f ( xj ) f ( x i. 0. the typical value of x k is x k = x 0 + kh for k = … . 2. these values are referred to as the differences of the function. the second differences are expressed as Δ fk = Δ ( Δ fk ) = Δ fk + 1 – Δ fk 2 and. fourth. n = 1. the above relation reduces to Numerical Analysis Using MATLAB® and Excel®. the second divided difference is defined as: f ( x i. • Likewise. x j ) = ---------------------------xi – xj where x i and x j are any two. 3 and 4 . within an interval. – 1.fk + n – 2 + … + ( –1 ) Δ f k = f k + n – nf k + n – 1 + ------------------nf k + 1 + ( – 1 ) f k 2! For k = 0 . x k ) = --------------------------------------------xi – xk and the third. 2. and so on divided differences are defined similarly. x j ) – f ( x j. not necessarily consecutive values of x . … • We can now express the first differences are usually expressed in terms of the difference operator Δ as Δ fk = fk + 1 – fk • Likewise.– 2. for positive integer values of n Δ fk = Δ ( Δ n n–1 fk ) = Δ n–1 fk + 1 – Δ n–1 fk • The difference operator Δ obeys the law of exponents which states that Δ ( Δ fk ) = Δ n m n m+n fk • The nth differences Δ f k are found from the relation n n–1 n n(n – 1) . in general. x k ) f ( x i. • If the values of x are equally spaced and the denominators are all the same. • If the constant difference between successive values of x is h . Third Edition Copyright © Orchard Publications 7−39 . x j. 1.Summary 7.

Then.for j = 0. the factorial polynomial reduces to p n ( x ) = r 0 + xq 0 ( x ) 7−40 Numerical Analysis Using MATLAB® and Excel®. 1. n a j = -----------------j! j • Factorial polynomials provide an easier method of constructing a difference table. We divide p n ( x ) by x to obtain a quotient q 0 ( x ) and a remainder r 0 which turns out to be the constant term a 0 . • The factorial polynomials are defined as and (x) (n) = x ( x – 1 ) ( x – 2 )… ( x – n + 1 ) 1 = --------------------------------------------------------( x – 1 ) ( x – 2 )… ( x + n ) (x) –( n ) Using the difference operator Δ with the above relations we obtain and Δ(x) Δ(x) (n) = n(x) (n – 1) –( n ) = –n ( x ) n –( n – 1 ) These are very similar to differentiation of x and x . Third Edition Copyright © Orchard Publications . • We can express any algebraic polynomial f n ( x ) as a factorial polynomial p n ( x ) . the nth differences of a polynomial of degree n are constant. Then. we can express that polynomial as pn ( x ) = a0 + a1 ( x ) (1) –n + a2 ( x ) (2) + … + an ( x ) (n) where Δ pn ( 0 ) . …. in analogy with Maclaurin power series. The procedure is as follows: 1. 2.Chapter 7 Finite Differences and Interpolation Δ f0 = f2 – f1 Δ f 0 = f 2 – 2f 1 + f 0 Δ f 0 = f 3 – 3f 2 + 3f 1 – f 0 Δ f 0 = f 4 – 4f 3 + 6f 2 – 4 f 1 + f 0 4 3 2 • As with derivatives.

q0 ( x ) = r1 + ( x – 1 ) q1 ( x ) and by substitution we obtain pn ( x ) = r0 + x [ r1 + ( x – 1 ) q1 ( x ) ] = r0 + r1 ( x ) (1) + x ( x – 1 ) q1 ( x ) 3. • The definite sum of p n ( x ) for the interval a ≤ x ≤ a + ( n – 1 ) h is α + (n – 1)h x=α ∑ p n ( x ) = p n ( α ) + p n ( α + h ) + p n ( α + 2h ) + … + p n [ α + ( n – 1 ) h ] • In analogy with the fundamental theorem of integral calculus which states that ∫a f ( x ) d x b = f( b) – f(a ) Numerical Analysis Using MATLAB® and Excel®.Summary 2. Third Edition Copyright © Orchard Publications 7−41 . and thus q1 ( x ) = r2 + ( x – 2 ) q2 ( x ) and by substitution we obtain pn ( x ) = r0 + r1 ( x ) = r0 + r1 ( x ) (1) (1) + x ( x – 1 ) [ r2 + ( x – 2 ) q2 ( x ) ] + r2 ( x ) (2) (2) + x ( x – 1 ) ( x – 2 ) q2 ( x ) (n – 1) and in general. It is denoted as Δ p n ( x ) . pn ( x ) = r0 + r1 ( x ) (1) + r2 ( x ) + … + rn – 1 ( x ) j + rn ( x ) (n) where Δ pn ( 0 ) r j = a j = -----------------j! • The antidifference of a factorial polynomial is analogous to integration in elementary calculus. Then. We divide q 0 ( x ) by ( x – 1 ) . to obtain a quotient q 2 ( x ) and a remainder r 2 which turns out to be the constant term a 2 . and it is computed from Δ (x) –1 –1 (n) (x) = -------------------(n + 1) (n + 1) • Antidifferences are very useful in finding sums of series. to obtain a quotient q 1 ( x ) and a remainder r 1 which turns out to be the constant term a 1 . We divide q 1 ( x ) by ( x – 2 ) .

x 2 ) + ( x – x 0 ) ( x – x 1 ) ( x – x 2 ) f ( x 0. x 1. x 1. and third divided differences respectively. and f ( x 0. • Lagrange’s Interpolation Method uses the formula ( x – x 1 ) ( x – x 2 )… ( x – x n ) ( x – x 0 ) ( x – x 2 )… ( x – x n ) . x n need not be equally spaced. the values near the beginning of the given data set. x 2. or taken in consecutive order. x n are equally spaced with interval h .Δ f 0 + --------------------------------. x 1. x 1. hence the name forward interpolation. x 2. x 1 ) + ( x – x 0 ) ( x – x 1 ) f ( x 0.f(x ) + ----------------------------------------------------------------------------( x n – x 0 ) ( x n – x 2 )… ( x n – x n – 1 ) n and. x 2. x 3 ) where f ( x 0. …. x 1. It is used to interpolate values near the smaller values of x . The variable r is the difference between an unknown point x and a known point x 1 divided by the interval h . ( x – x1 ) r = -----------------h 2 3 This formula is valid only when the values x 0. x 1. ….f ( x 0 ) + -----------------------------------------------------------------------f(x ) f ( x ) = ----------------------------------------------------------------------( x 0 – x 1 ) ( x 0 – x 2 )… ( x 0 – x n ) ( x 1 – x 0 ) ( x 1 – x 2 )… ( x 1 – x n ) 1 ( x – x 0 ) ( x – x 1 )… ( x – x n – 1 ) . x 1. Δ f 0 . that is. that is. second. like Newton’s divided difference method.Chapter 7 Finite Differences and Interpolation we have the fundamental theorem of sum calculus which states that α + (n – 1)h x=α ∑ pn ( x ) = Δ pn ( x ) –1 α + nh α • One important application of finite differences is interpolation. • Newton’s Divided Difference Interpolation Method uses the formula f ( x ) = f ( x 0 ) + ( x – x 0 ) f ( x 0. …. x 2.Δ f0 + … f ( x ) = f 0 + r Δ f 0 + ----------------2! 3! where f 0 is the first value of the data set. has the advantage that the values x 0. second. Third Edition Copyright © Orchard Publications . x 1 ) . and third forward differences respectively. This method has the advantage that the values x 0. x 2. x 2 ) . x 3 ) are the first. • The Gregory−Newton Forward Interpolation method uses the formula r(r – 1) 2 r( r – 1)( r – 2) 3 . 7−42 Numerical Analysis Using MATLAB® and Excel®. Δ f 0 . x n need not be equally spaced or taken in consecutive order. f ( x 0. and Δ f 0 are the first.

Numerical Analysis Using MATLAB® and Excel®. that is. • If the increments in x values are small.y.’method’) where method allows us to specify nearest (nearest neighbor interpolation). Δ f –1 . It is used to interpolate values near the end of the data set. x n are equally spaced with interval h . interp1(x.y. cubic (cubic interpolation which requires equidistant values of x ). x 1. the last entries on the columns where the differences appear. and Δ f –3 are the first. and ( x – x1 ) r = -----------------h 2 3 This formula is valid only when the values x 0. Backward interpolation is an expression to indicate that we use the differences in a backward sequence.xi. that is.xi. the default interpolation). • We can perform interpolation to verify our results with the MATLAB functions interp1(x. spline (cubic spline interpolation which does also extrapolation).xi). Third Edition Copyright © Orchard Publications 7−43 .y. we can use the Excel VLOOKUP function to perform interpolation.Summary • The Gregory−Newton Backward Interpolation method uses the formula r(r + 1) 2 r(r + 1 )(r + 2) 3 Δ f – 2 + ---------------------------------Δ f–3 + … f ( x ) = f 0 + r Δ f –1 + ----------------2! 3! where f 0 is the first value of the data set.z. and interp2(x.y. the larger values of x . x 2. linear (linear interpolation. second and third backward differences.xi) but performs two dimensional interpolation.yi) which is similar to interp1(x. …. Δ f –2 .

Use the data of the table below and the appropriate (forward or backward) Gregory−Newton formula.323 2.112 1. to compute: a.219 1. Use the data of the table below and Newton’s divided difference formula to compute: a. and then construct the difference table with h = 1 . 50.5 1.636 1.3 ) b.054 1.280 54 7.10 Exercises 1. calculate the leading differences.071 51 7.8 2.9 x 50 7.483 x 3. f ( 1.011 7−44 Numerical Analysis Using MATLAB® and Excel®.0 3.141 52 7.211 53 7.95 ) x y=f(x) 1.348 55 7. Express the given polynomial f ( x ) below as a factorial polynomial p ( x ) . Third Edition Copyright © Orchard Publications .2 55.1 1.Chapter 7 Finite Differences and Interpolation 7. f ( 1. f ( x ) = x – 2x + 4x – x + 6 5 4 3 2.2 1.416 56 7.7 2. b.

q) function which divides the polynomial p by q.r3]=deconv(q2. r 4 and r 5 . i. i.e. x−4 % Computation of fifth quotient and remainder % Coefficients of sixth (last) divisor. [q3. we will use (7. to determine p n ( x ) . The MATLAB script is as follows: px=[1 −2 4 0 −1 6]. for convenience.e. r 3.d0) d1=[1 −1].e. i. r 1. f ( x ) = x – 2x + 4x – x + 6 5 4 3 The highest power of the given polynomial f ( x ) is 5 .e.11 Solutions to End−of−Chapter Exercises 1. [q0. r 2. x−5 % Computation of sixth (last) quotient and remainder q0 = 1 r0 = 0 q1 = 1 r1 = 0 q2 = 1 r2 = 0 0 0 13 1 5 0 0 0 2 -1 3 3 0 0 0 0 6 -2 4 0 -1 Numerical Analysis Using MATLAB® and Excel®.d4) d5=[1 −5]. Third Edition Copyright © Orchard Publications 7−45 .r0]=deconv(px.d2) d3=[1 −3]. pn ( x ) = r0 + r1 ( x ) (1) + r2 ( x ) (2) + … + rn – 1 ( x ) (n – 1) + rn ( x ) (n) We can compute the remainders by long division but.e. we must evaluate the remainders r 0.r1]=deconv(q0. x % Computation of first quotient and remainder % Coefficients of second divisor.r2]=deconv(q1. d0=[1 0]. [q4.d3) d4=[1 −4].d5) % Coefficients of given polynomial % Coefficients of first divisor. repeated below.e. then. [q5.Solutions to End−of−Chapter Exercises 7. [q1.d1) d2=[1 −2].r5]=deconv(q4. x−2 % Computation of third quotient and remainder % Coefficients of fourth divisor. i.28). i.r4]=deconv(q3. i. x−3 % Computation of fourth quotient and remainder % Coefficients of fifth divisor. we will use the MATLAB deconv(p. [q2. x−1 % Computation of second quotient and remainder % Coefficients of third divisor.

i. syms x.e.30). (13*x*(x−1))+2*x+6)) px = x^5-2*x^4+4*x^3-x+6 We observe that this is the same algebraic polynomial as f ( x ) . Third Edition Copyright © Orchard Publications . (x) (n) = x ( x – 1 ) ( x – 2 )… ( x – n + 1 ) with the MATLAB collect(‘s_expr’) function.e. Δ p n ( 0 ) = j! r j and p n ( x ) above Δ p ( 0 ) = 0! ⋅ 6 = 6 Δ p ( 0 ) = 3! ⋅ 17 = 102 Δ p ( 0 ) = 6! ⋅ 0 = 0 6 3 0 j Δ p ( 0 ) = 1! ⋅ 2 = 2 4 1 Δ p ( 0 ) = 2! ⋅ 13 = 26 Δ p ( 0 ) = 5! ⋅ 1 = 120 5 2 Δ p ( 0 ) = 4! ⋅ 8 = 192 We enter these values in the appropriate spaces as shown on the table below. We will now compute the leading entries for the difference table using (7. i. 7−46 Numerical Analysis Using MATLAB® and Excel®..Chapter 7 Finite Differences and Interpolation q3 = 1 r3 = 0 q4 = 1 r4 = 0 q5 = 0 r5 = 1 Therefore.12). with reference to (7.. the factorial polynomial is pn ( x ) = 6 + 2 ( x ) (1) 4 0 17 8 + 13 ( x ) (2) + 17 ( x ) (3) + 8(x) (4) + (x) (5) We will verify that p n ( x ) above is the same polynomial as the given f n ( x ) by expansion of the factorials using (7.28).. px=collect((x*(x−1)*(x−2)*(x−3)*(x-4)+(8*x*(x−1)*(x−2)*(x-3))+(17*x*(x−1)*(x−2))+.

2 should be in the interval 50 ≤ x ≤ 51 . as shown below.280 54 7. and so on. Numerical Analysis Using MATLAB® and Excel®. the third from the second. We enter the given x and f ( x ) values in a difference table. then.483 x a.Solutions to End−of−Chapter Exercises x p( x) 6 Δ 2 Δ 2 Δ 3 Δ 4 Δ 5 Δ 6 26 102 192 120 0 We obtain the remaining set of values by crisscross addition as shown on the table below.071 51 7. Third Edition Copyright © Orchard Publications 7−47 . We will use the differences in a forward sequence. This is because the value of 50. and third differences. the first entries on the columns where the differences appear.348 55 7. we compute the first. x 50 7. These are not divided differences and therefore.416 56 7.141 52 7.211 53 7. second. we simply subtract the second value of f ( x ) from the first. that is. x p( x) 6 Δ 2 Δ 2 Δ 3 Δ 4 Δ 5 Δ 6 8 28 36 156 192 578 770 1606 2376 3672 6048 26 102 128 294 422 606 1028 1038 2066 432 312 120 192 120 0 2.

211 7.2 ) = 7. fprintf('\n').20 – 2 ) + -------------------------------------------------------------------. we obtain ( 0.211 7.fx.001 0.[50. x 2.0849 7−48 Numerical Analysis Using MATLAB® and Excel®.068 -0.001 0..000 0.000 0. fx=[7. fprintf('spline interpolation yields f(50.070 -0.416 7. disp(spline_interp) spline interpolation yields f(50.20 . using (7.. x 1 ) 0. x 1.483]. Check with MATLAB: x =[50 51 52 53 54 55 56].20 ) ⋅ ( 0.20 – 1 ) ( 0.2) = 7.000 -0.070 7.20 – 1 ) f ( 50.001 0.348 7.⋅ ( 0.071 + ( 0.071 ) + ------------------------------------------2! ( 0.001 0.2 r = ------------1 h and with these values.'spline').2) = \n').067 7.071 7.20 ) ⋅ ( 7. Third Edition Copyright © Orchard Publications . spline_interp=interp1(x.280 7.483 f 0 = f ( 50 ) = 7. x 1.20 ) ⋅ ( 0.141 0.= 0.416 7.54).348 7.071 h = x 1 – x 0 = 51 – 50 = 1 x – x1 – 50.280 7.001 ) = 7.001 -0.069 0.0 -------------------------.Chapter 7 Finite Differences and Interpolation 1st Difference x 50 51 52 53 54 55 56 f (x) 7.001 -0.141 7..071 2nd Difference f ( x 0.2]'.= 50.085 3! The spreadsheet below shows the layout and computations for Part (a). x 3 ) f ( x 0. x 2 ) 3rd Difference f ( x 0.068 0.

2 Δ 2f Δ 3f Δ 4f Δ 5f Δ 6f h= A10-A8= 1. Third Edition Copyright © Orchard Publications 7−49 .003 Δf 50.001 0.483 7.3) = 7.002 -0.Solutions to End−of−Chapter Exercises Gregory-New ton Forw ard Interpolation Method for Exercise 7.20 7.071 0.0 52.0 f(x) 7.001 0.00 50 51 52 53 54 55 56 b.068 7. we will use the backward interpolation formula r(r + 1) 2 r(r + 1 )(r + 2) 3 Δ f – 2 + ---------------------------------Δ f–3 + … f ( x ) = f 0 + r Δ f – 1 + ----------------2! 3! Numerical Analysis Using MATLAB® and Excel®.55) Interpolate f(x) at x= x 50.001 -0.001 0.0 55.2 -0.141 7.54) and (7.085 50 51 52 53 54 55 56 7.141 0.001 0.001 0.348 0.0 51.000 -0.2(a) See expressions (7.0 56.50 7.483 r= (D5-A8)/C22= 0.416 0.40 7.2)= round(B8+F22*C9+(F20*(F20-1)*D10)/FACT(2)+(F20*(F20-1)*(F20-2)*E11)/FACT(3).348 7.0 54.211 0.071 7.000 0. Since the value of 55.30 7.0 53.416 7.068 7.10 7.9 is very close to the last value in the given range.001 -0.070 7.000 -0.211 7.070 7.069 7.280 7.067 7.000 -0.003 0.001 -0.280 0.00 f(50.

fx.500 2.112 1.9) = 7. disp(spline_interp) spline interpolation yields f(55.8 – 1. Third Edition Copyright © Orchard Publications .440 2.'spline').1 ) ( 0.219 – 1..9 – 56.070 1.000 1.348 7. and ( x – x1 ) r = -----------------h 2 3 We arbitrarily choose f 0 = 7.219 -------------------------------.1 ) ( – 0.5 2.390 – 1.054 -------------------------------.1 1.636 -------------------------------. x y=f(x) 1.323 -------------------------------.5 2.= 2. Δ f –2 .0 – 1.323 – 2.071 7.011 a.1 ) ( – 0.7 2.483].( – 0.070 -------------------------------. h = 56 – 55 = 1 and r = ( x – x 1 ) ⁄ h = ( 55.690 – 2.Chapter 7 Finite Differences and Interpolation where f 0 is the first value of the data set.483 as our starting point since f ( 55.090 1.[55.= 0.219 1. The first divided differences are: 1.440 – 2.58) we have: ( – 0.011 – 2.390 1.2 3.0 – 1.141 7.400 1.= 2.7 3. fx=[7..= 1.390 -------------------------------.1 1.690 1.7 – 1.070 ) + ----------------------------------------2! ( – 0.9) = \n').5 1.054 1.112 -------------------------------.5 – 1.001 ) = 7. second and third backward differences. Δ f –1 ..0 3.5 – 1.9 ) lies between f ( 55 ) and f ( 56 ) . by (7.1 2.0 ) ⁄ 1 = – 0.1 + 2 ) + -------------------------------------------------------------------.636 – 1.= 1. spline_interp=interp1(x.4764 3.1 Now. Then.211 7.9 ) = 7.( 0.2 – 1. and Δ f –3 are the first. fprintf('spline interpolation yields f(55.= 2.090 – 1.1 + 1 ) .416 7.1 + 1 ) ( – 0.8 – 1.280 7.7 – 1.090 -------------------------------.= 1.= 3.690 -------------------------------.800 1.= 2.2 2.2 1.054 – 1.636 1.9]'.8 2.323 2.000 ) f ( 55.8 The second divided differences are: 1.483 + ( – 0. fprintf('\n').7 7−50 Numerical Analysis Using MATLAB® and Excel®.476 3! Check with MATLAB: x =[50 51 52 53 54 55 56].

Third Edition Copyright © Orchard Publications 7−51 . x 3 ) f ( 1.000 2.1 and for x in (7.054 2.95 ) = 3.1 1.49).000 1.214 + 0.95 – 1.690 1.090 1.8 ) ( 1.636 2.= 1.95 ) we start with x 0 = 2.3 – 1. x 2 ) + ( x – x 0 ) ( x – x 1 ) ( x – x 2 ) f ( x 0.95 – 1.112 + 0.323 3. x 2.500 3.44 ) + ( 1.2 ) ( 1.440 1.= 1.011 + ( 1.000 2.112 + ( 1. 1st Divided Difference 2nd Divided Difference 3rd Divided Difference x 1. x 1 ) + ( x – x 0 ) ( x – x 1 ) f ( x 0. x 1.3 – 1.3 – 1.1 2.1 ) ( 1.000 1.7 – 1. x 2 ) f ( x 0.000 2.0 – 1.002 = 2. Then.95 – 1.3 – 1.5 ) + ( 1.7 1.800 -------------------------------.000 1.8 2.3 ) = 1.3 – 1.07 ) + ( 1. x 3 ) f ( 1. we construct the difference table below.5 With these values. x 3 ) 0. x 1 ) + ( x – x 0 ) ( x – x 1 ) f ( x 0. Numerical Analysis Using MATLAB® and Excel®.400 – 0.95 – 2 ) ( 3.Solutions to End−of−Chapter Exercises and the third divided differences are: 1. x 2.3 – 1.49).2 ) ( 1.011 To find f ( 1. we use x = 1. x 1.95 – 2 ) ( 1.818 The spreadsheet below verifies our calculated values.000 – 1.0 f (x) 1. We start with x 0 = 1.400 2.4 ) + ( 1.350 b. x 1. f ( x ) = f ( x 0 ) + ( x – x 0 ) f ( x 0.000 -------------------------------.390 1.To find f ( 1.1 ) ( 1.000 2. x 1.3 . x 1 ) 1.004 = 1.2 2.95 .011 – 0.0 and for x in (7.112 1. x 1.019 – 0.5 1.1 ) ( 1. x 2.3 ) . x 2 ) + ( x – x 0 ) ( x – x 1 ) ( x – x 2 ) f ( x 0.7 ) ( 1 ) = 3.172 – 0. x 1.219 f ( x 0.95 – 2 ) ( 1.8 – 1.500 – 2.2 1.400 -------------------------------.028 – 0. we use x = 1.5 ) ( 1 ) = 1.8 ) ( 2. Then.070 f ( x 0.800 1.= 1. f ( x ) = f ( x 0 ) + ( x – x 0 ) f ( x 0.

50 3.011 3.3 1.2 1.5 1.x3) x 1.fx. spline_interp=interp1(x. Third Edition Copyright © Orchard Publications .090 2. disp(spline_interp) spline interpolation value of f(1.011].011 2.95): 2. x1) 1.112 1.Chapter 7 Finite Differences and Interpolation 1st divided difference f(x0.00 1.50 1. x1.323 3.'spline').8 2.x1.1 1. fprintf('spline interpolation value of f(1.8 2.00 f(x) 2.3): 1.070 1.[1.8 2..3): \n\n').0 1.440 3.95): \n\n').636 2.690 2.000 0.5 x 1..054 2.8 2 1..3]'.x2.636 2.112 1.7 1.5 1.636 2.390 1.50 2.95]'..[1.800 1.219 1. fprintf('\n').054 2.2 1.323 3.7 1.112 1.000 1.0 Check with MATLAB: x=[ 1.500 2.8184 7−52 Numerical Analysis Using MATLAB® and Excel®. fprintf('spline interpolation value of f(1.5 1...00 1. disp(spline_interp) spline interpolation value of f(1.3380 spline_interp=interp1(x.400 1. x2) f(x0.000 2nd divided 3rd divided difference difference f(x0. fx=[ 1.323 3.1 1. fprintf('\n').fx.054 2.219 1.1 f(x) 1.7 1.'spline').000 1.0].2 1.219 1.0 1.

1.2) x Figure 8. that is.2. Third Edition Copyright © Orchard Publications 8−1 . we normally assume that the data.1 Curve Fitting Curve fitting is the process of finding equations to approximate straight lines and curves that best fit given sets of data. for the data of Figure 8.2.1) y = mx + b y x Figure 8. (8. For example. and our task is to find the dependent variable y . Numerical Analysis Using MATLAB® and Excel®. Straight line approximation.2. We will illustrate their application with several examples. represent the independent variable x .Chapter 8 Linear and Parabolic Regression T his chapter is an introduction to regression and procedures for finding the best curve to fit a set of data. 8. Parabolic line approximation In finding the best line. and regression with power series approximations.1 and 8. we can use the equation for the quadratic or parabolic curve of the form y = ax + bx + c y 2 (8. shown by the small circles in Figures 8. We will discuss linear and parabolic regression. This process is called regression. we can use the equation of a straight line. For Figure 8.1.

we call it a least-squares parabola. Financial analysts in large corporations use regression to forecast future costs. y n ) approximate a straight line. and assume that corresponding to the values x 1. ( x 3. a straight line that satisfies (8. Now.3) and it is referred to as the least-squares curve. we can find more than one straight line or curve to fit a set of given data.2 Linear Regression We perform linear regression with the method of least squares. y 3. We derive the values of m and b . We denote the straight line equations passing through these points as y 1 = mx 1 + b y 2 = mx 2 + b y 3 = mx 3 + b … y n = mx n + b (8. and so on. but we interested in finding the most suitable. we need to determine the values of the unknowns m and b from all n equations. y n . Investment corporations use regression analysis to compare a portfolio’s past performance versus index figures. the distance of data point x 2 from the same line as d 2 . that will make the equation of the straight line to best fit the observed data. If it is a parabola. With this method. The best fitting straight line or curve has the property that d 1 + d 2 + … + d 3 = minimum 2 2 2 (8. and we have observed that the points ( x 1. Let the distance of data point x 1 from the line be denoted as d 1 . and the Census Bureau use it for population forecasting.5).Chapter 8 Linear and Parabolic Regression Regression can be linear (straight line) or curved (quadratic. ( x 2. …. Thus. y 3 ). cubic.4) such that the sum of the squares of the errors will be minimum. x 2. we have observed the values y 1. the slope m and y-intercept b are the same in all equations since we have assumed that all points lie close to one straight line. y 2. However. 8. we compute the coefficients m (slope) and b (y-intercept) of the straight line equation y = mx + b (8. …. y 2 ). ( x n. etc.5) In (8. x 3. ….3) is called a least squares line. we will not obtain valid values for all points if we solve just two 8−2 Numerical Analysis Using MATLAB® and Excel®. Obviously.) and it is not restricted to engineering applications. Third Edition Copyright © Orchard Publications . as follows: Let x and y be two related variables. x n . let us suppose that we have plotted the values of y versus the corresponding values of x . y 1 ).

7) and (8. the error between the observed value y 2 and the value that lies on the straight line is y 2 – ( mx 2 + b ) and so on. Numerical Analysis Using MATLAB® and Excel®. ∑ squares = – 2x 1 [ y 1 – ( mx 1 + b ) ] – 2x 2 [ y 2 – ( mx 2 + b ) ] – … – 2x n [ y n – ( mx n + b ) ] = 0 (8.7) and ∂ ∂b ∑ squares = – 2 [ y 1 – ( mx 1 + b ) ] – 2 [ y 2 – ( mx 2 + b ) ] – … – 2 [ y n – ( mx n + b ) ] = 0 (8. and the value that lies on the straight line. a system that has more unknowns than equations is said to be underdetermined and these systems have infinite solutions.6) we must equate to zero its two partial derivatives with respect to m and b . Accordingly. depending on the position of the observed value. to minimize (8.Linear Regression equations with two unknowns. will be minimum. Third Edition Copyright © Orchard Publications 8−3 . is y 1 – ( mx 1 + b ) . some may cancel each other and give us an erroneous sum of the distances. This will be achieved if we use the magnitudes (absolute values) of the distances. and simplifying (8. this method is referred to as the method of least squares. Let the sum of the squares of the errors be ∑ squares Since = [ y 1 – ( mx 1 + b ) ] + [ y 2 – ( mx 2 + b ) ] + … + [ y n – ( mx n + b ) ] 2 2 2 (8. For this reason. and the value at the point on the straight line.8) are positive and thus value. Collecting like terms. On the contrary. This difference could be positive or negative.8) we obtain ∑ squares will have its minimum * A linear system of independent equations that has more equations than unknowns is said to be overdetermined and no exact solution exists. The straight line that we choose must be a straight line such that the distances between the observed values. we find the sum of the squared distances between observed points and the points on the straight line.6) ∑ squares ∂ ∂m is a function of two variables m and b . and the corresponding values on the straight line.8) The second derivatives of (8. Likewise.7) and (8. Then. if we were to combine positive and negative values. * The error (difference) between the observed value y 1 .

1 In a typical resistor. m and b are computed from D1 m = ----Δ D2 b = ----Δ Σ xy Σ x Σy n 2 (8.9) simultaneously by Cramer’s rule.9) where Σ x = sum of the numbers x Σ y = sum of the numbers y Σ xy = sum of the numbers of the product xy Σ x = sum of the numbers x squared n = number of data x 2 We can solve the equations of (8.3 80 51. or with MATLAB using matrices. we use the spreadsheet of Figure 8. TABLE 8.5 70 48.1 Data for Example 8.3 where we enter the given values and we perform the computations using spreadsheet formulas.1 90 54 100 56. or with Excel. For convenience.0 30 37 40 40 50 42. With Cramer’s rule.0 20 34.7 R (Ω) Solution: There are 11 sets of data and thus n = 11 .6 60 45.Chapter 8 Linear and Parabolic Regression ( Σ x ) m + ( Σ x ) b = Σ xy ( Σ x ) m + nb = Σ y 2 (8. Third Edition Copyright © Orchard Publications .1 .Resistance versus Temperature T ( C) ° x y 0 27. 8−4 Numerical Analysis Using MATLAB® and Excel®. Compute the straight line equation that best fits the observed data. The temperature increments and the observed resistance values are shown in Table 8-1.10) where Δ = Σx Σx Σx n 2 D1 = D2 = Σ x Σ xy Σx Σy (8. the resistance R in Ω (denoted as y in the equations above) increases with an increase in temperature T in ° C ( denoted as x ) .11) Example 8.6 10 31.

Linear Regression Spreadsheet for Example 8.0 34.3 51.0 30. Then. All work is shown on the spreadsheet of Figure 8. Spreadsheet for Example 8.3.0 Resistance 50.8 Σx x2 0 100 400 900 1600 2500 3600 4900 6400 8100 10000 38500 xy 0 310 680 1110 1600 2130 2730 3381 4088 4860 5670 26559 38500 = 550 26559 = 467.12) We can use Excel’s Add Trendline feature to produce quick answers to regression problems.5 48.1 x (0C) 0 10 20 30 40 50 60 70 80 90 100 550 Σx 2 y(Ω ) 27. the straight line equation that best fits the given data is y = mx + b = 0.0 42.10) and (8.0 56. we compute the sums so they can be used with (8.0 40.1 54.7 467.6 31. The values of m and b are shown in cells I20 and I24 respectively. Third Edition Copyright © Orchard Publications 8−5 .288 550 = 11 b=D2/Δ = 28.3.0 20.288x + 28. We will illustrate the procedure with the following example. we enter the x (temperature) values in Column A.0 0 20 40 60 80 100 Temperature 550 = 11 m=D1/Δ = 0.0 37.6 45. Numerical Analysis Using MATLAB® and Excel®. and y (the measured resistance corresponding to each temperature value) in Column B. Thus.1 Accordingly.8 38500 = 550 Resistance versus Tem perature 60.0 40. Columns C and D show the x 2 and xy products.123 26559 = 467.123 (8.11).8 3402850 34859 121000 Σx Σ xy Σy Σx 2 n Σx n Σ xy Σy Σx Figure 8.

5 48.2 To produce the plot of Figure 8. we choose Plot Area from the taskbar below the main taskbar. we type Resistance (Ohms). We click on the Chart Wizard icon.6 31. we click on the top (scatter) sub-type.2. and on the Chart title box.7 60 50 40 30 20 0 20 40 60 80 100 Temperature (degrees Celsius) Straight line for Example 8.0 42. and we observe that the plot appears next to the data. We click anywhere on the y-axis to select it. and on the Gridlines tab. we perform the following steps: 1. Third Edition Copyright © Orchard Publications Resistance (Ohms) .6 45. we click on the Major gridlines box to deselect it. We click on Chart on the main taskbar. we click on Next>Next> Next>Finish. We click on the Titles tab.2 Repeat Example 8. We click on XY (Scatter) Type. Plot of the straight line for Example 8.1 54. We click on the Series 1 block inside the Chart box.0 37. on the Value X-axis. and we press the Delete key to delete it. The displayed chart types appear on the Standard Types tab. on the Patterns tab we click on the white box (below the selected gray box).3 51. we change the maximum value from 150 to 100.0 56. and we click on the small (with the hand) box.4.Chapter 8 Linear and Parabolic Regression Example 8.0 34. and we click on OK. We click anywhere on the x-axis to select it. Then. Next. 2.4. 3. we type Straight line for Example 8.0 40. and on the Value Y-axis.1 using Excel’s Add Trendline feature.4. To change the plot area from gray to white. and we click OK. we click on the small (with the hand) box. Solution: We first enter the given data in columns A and B as shown on the spreadsheet of Figure 8. we change the Major Unit to 10. 8−6 Numerical Analysis Using MATLAB® and Excel®. and we click on OK. we click anywhere on the perimeter of the Chart area. We click on the Scale tab. On the Chart sub-types options. we type Temperature (degrees Celsius). We observe now that the plot area is white.2 Figure 8. x (0C) 0 10 20 30 40 50 60 70 80 90 100 y(Ω ) 27. and observe six square handles (small black squares) around it. and we click on the small (with the hand) box. Under the Value (Y) axis. we change the minimum value from 0 to 20. We click on the Scale tab.

Third Edition Copyright © Orchard Publications 8−7 . We can also use Excel to compute and display the equation of the straight line. To make the plot more presentable.2. and when the two-directional arrow appears. We can also stretch (or shrink) the height of the Chart area by placing the cursor near the center handle of the lower side of the graph. and c are found from ( Σ x ) a + ( Σ x ) b + nc = Σ y ( Σ x ) a + ( Σ x ) b + ( Σ x ) c = Σ xy (Σx )a + (Σx )b + (Σx )c= Σx y 4 3 2 2 3 2 2 (8. It provides a wealth of information for statisticians. We now observe that the points on the plot have been connected by a straight line.3 Parabolic Regression We find the least-squares parabola that fits a set of sample points with y = ax + b + c 2 (8. 5. we can stretch or shrink the width of the plot to the left or to the right. we click anywhere on the perimeter of the Chart area. and move it downwards with the mouse.3 Find the least−squares parabola for the data shown in Table 8.14) where n = number of data points. b. This feature will be illustrated in Example 8.13) where the coefficients a. We place the cursor near the center handle of the upper side of the graph. 8. and we observe the six handles around it. On the pull-down menu. and we click on OK. we click on Add Trendline. and contains several terms used in probability and statistics. The Data Analysis Toolpack in Excel includes the Regression Analysis tool which performs linear regression using the least squares method. Numerical Analysis Using MATLAB® and Excel®. we click on the first (Linear). and we click on Chart above the main taskbar.4. Example 8. Similarly. we move it upwards by moving the mouse in that direction. We click anywhere on the perimeter of the Chart area to select it.Parabolic Regression 4. by placing the cursor near the center handle of the left or right side of the Chart area. On the Type tab.

29 6 3.05 39.58 45.96 262.2 4.48 9 5.15) with matrix inversion and multiplication.98 8 4.35 21.93 31.1 7.79 5 2.1 2. The procedure was presented in Chapter 4.93 11 6.8 7.39 134.9 40.71 13 7.6 6.83 10.9 7. b.62 12 7.91 2541.14 1677.06 7.7 6.88 31. 8−8 Numerical Analysis Using MATLAB® and Excel®.60 41.5 4.49 437.64 778.46 259.6 9.72 2698.3 7.27 7 4.76 294.3 x y x y 1.7 5.15a + 79.1 7.2 3.7 96.72 32331.48 4 1.6 6.15 4004.31 17 Σx= Σy= Σx2= Σx3= Σx4= Σxy= Σx2y= 18 79.7 7.76 17.6 5.73 2.01 117.25 3.7 7.50 32331.65 576.87 10 5.1 87.3 7.5 73.72 44.15c = 2698.16 282.15b + 79.07 5.5 1.65 11.1 6.88 789.1 3.6 2.51 341. as shown in Figure 8.8 4004.37 (8. we compute the values shown in Columns C through G.0 8.4 4.1c = 437.09 148.7 7.5.19 1055.4 84.8 5. 530.9 1.96 636.5 5.06 5470.6 4.0 9.1 6.14).8 3.42 45.8 530.2 4. and from the data of Columns A and B.48 37.1 6.24 5.2 Data for Example 8.1 57.8 5.3 7.76 438.44 1.04 233.44 18.38 5.6 5.22 207.3 18.49 185.41 357. The sum values are shown in Row 18.98 3336.82 15 9.5 5.6 4.1b + 15c = 87.8 2.Chapter 8 Linear and Parabolic Regression TABLE 8. A B C 2 D 3 E 4 F G 2 x x x xy 1 x y xy 2 1.70 17.24 182.8 2.58 14 8.6.2 3.70 332.4 28.5 5.22 38.6 50.2 32.3 By substitution into (8. Spreadsheet for Example 8.48 3 1.68 26.50 10. and c .04 941.3 7.4 6.4 Solution: We construct the spreadsheet of Figure 8.4 6.86 332.37 Figure 8.50a + 530.50b + 530.9 7.2 1. and from these we form the coefficients of the unknown a.19 9223.78 16 9.69 7163. Third Edition Copyright © Orchard Publications .28 287.40 6.15) We solve the equations of (8.49a + 4004.3 9.17 46.61 29.49 79.79 92.6 24.08 38.70 67.5.

364 4.20x +1.15 5 6 7 0.1 0.72 4 2 Σx y= 2698.94x+2. Use Excel’s Add Trendline feature to derive a polynomial that best approximates the given data.7.002 a= -0.872 4.2 1.4 1.6.20x + 1.9 1.80 3 Σxy= A= 4004.15 79.104 5.94x + 2.385 0.032 -0.2 0.240 y = − 0.820 4.181 -0. Spreadsheet for the solution of the equations of (8. Third Edition Copyright © Orchard Publications y 8−9 .524 3.344 3.94 9 0.50 530.016 b= 1.00 87.3 Example 8.1 1.3 0. the least−squares parabola is y = – 0.5 y 2. for brevity.10 437.780 2.4 The voltages (volts) shown on Table 8.700 3.Parabolic Regression A B C D E F G 1 Matrix Inversion and Matrix Multiplication for Example 8.20 -1 A = 8 -0.979 -0. x 0.15 79.3 2 Σy= 530.7.0 0.49 4004.040 4.10 15.3 were applied across the terminal of a non−linear device and the current ma (milliamps) values were observed and recorded.520 4.016 0.160 3.8 where.7 0.204 4.672 4.37 32331.5 0.964 5.3 1.0 1.032 c= 2.78 Figure 8.4 0.15) Therefore.78 8 7 6 5 4 3 2 1 0 0 1 2 3 4 5 x 6 7 8 9 10 2 Figure 8.6 0.385 0.78 2 The plot for this parabola is shown in Figure 8. Solution: We enter the given data on the spreadsheet of Figure 8.8 0.972 3. only a partial list of Numerical Analysis Using MATLAB® and Excel®.50 530. Parabola for Example 8.

50 0.50 0.1275x .25 0. we clicked on Display R squared value on chart.25 1.00 2.25 0.34 Volts Amps 2. we clicked on Display equation on chart.00 1.05 3.63 2.00 0.50 2.4 Experimental Data Volts ma Volts ma 0. the smooth curve was chosen from the Add trendline feature. we need to enter all data in Columns A and B.50 0. On the Options tab.00 2. it is discussed in probability and statistics textbooks.50 1.9997 1 2 3 4 5 Figure 8.00 0.00 0. However. This is the Pearson correlation coefficient R . The quantity R is a measure of the goodness of fit for a straight line or.39 2.28 0.0.75 0.25 1.72 1.50 0.03 3.42 0.3 Data for Example 8. and Technology. as in this example.Chapter 8 Linear and Parabolic Regression the given data is shown.* 2 * It is also discussed in Mathematics for Business.03 0.18 4.75 0.50 0. but we clicked on the polynomial order 3 on the Add trendline Type tab.23 0.42 1.75 1.75 1.25 0.14 4.01 0.8.08 0.50 0.0403x 2 + 0.0182x 3 .00 1.0.75 0.50 1.11 0.75 1.60 0.75 0.25 0.25 0. Here.50 1. 8−10 Numerical Analysis Using MATLAB® and Excel®.14 0. Science. and on OK.28 5.05 0.0177 R2 = 0.2.4 Following the steps of Example 8.50 0.00 0.75 2. Third Edition Copyright © Orchard Publications .85 1.34 0.18 0.00 0 y = 0.00 0.23 4.01 3. to obtain the plot. TABLE 8.00 0. ISBN 0-9709511-0-8. for parabolic regression. we create the plot shown next to the data.11 4.75 0.18 1.00 1.25 0.25 1.00 0.08 3.91 2.75 0.25 2. Plot for the data of Example 8.00 1.00 0.

With MATLAB.y.'*b'). linear regression.5 48. Third Edition Copyright © Orchard Publications 8−11 .n) command. and compute the percent error (difference between the given values and the polynomial values).R. or higher order) approximation that is computed with the MATLAB polyfit(x.7]. that is. ylabel('R'). that is. Suppose we want to evaluate the polynomial p at one or more points. there is no relationship between the dependent y and independent x variables.6 45.x) command to evaluate the best line approximation p in the 0 ≤ T ≤ 100 range in ten degrees increments. Thus. % Fits a first degree polynomial (straight line since n =1) and returns % the coefficients m and b of the straight line equation y = mx + b 2 2 Numerical Analysis Using MATLAB® and Excel®. hold % Hold current plot so we can add other data p=polyfit(T.1 54 56.1).4 indicates that there is a strong relationship between the variables x and y . % Display experimental (given) points with asterisk % and smoothed data with blue line grid.5 % T= [ 0 10 20 30 40 50 60 70 80 90 100]. MATLAB returns the value of the polynomial at point x. and so on.5 Repeat Example 8. Let p denote the polynomial (linear. there is a nearly perfect relationship between these variables. cubic. if n = 1 .1 using the MATLAB’s polyfit(x. If x is a row vector. and returns the coefficients m and b . When R ≈ 1 . % Establishes desired x and y axes limits plot(T.x) function to evaluate the polynomial. Likewise.y. If n = 2 . When R ≈ 0 . regression is performed with the polyfit(x.n) function. it computes the best quadratic polynomial approximation and returns the coefficients of this polynomial. n=1'). the result of Example 8. Use also the polyval(p. MATLAB computes the best straight line approximation. Thus. % x−axis data R=[27. Plot resistance R versus temperature T in the range – 10 ≤ T ≤ 110 ° C . Example 8. title('R (Ohms) vs T (deg Celsius. quadratic.3 51.6 31 34 37 40 42. if n = 3 . xlabel('T'). % This is the script for Example 8.Parabolic Regression The correlation coefficient can vary from 0 to 1. We can use the polyval(p. Solution: The following MATLAB script will do the computations and plot the data.y. the polynomial is evaluated for all values of the vector x.n) function. and n is the degree of the polynomial. where x and y are the coordinates of the data points. Use n = 1 to compute the best straight line approximation. there is a nearly perfect fit between the cubic polynomial and the experimental data. it computes the best cubic polynomial approximation. If x is a scalar.R. % y−axis data axis([−10 110 20 60]).

Plot for Example 8.9. q=polyval(p.9.a).q) The plot for the data of this example is shown in Figure 8.2881 28. disp(p). format bank % Two decimal place display will be sufficient disp('Smoothed R values evaluated from straight line are:'). R_smoothed=polyval(p. fprintf('\n'). n=1 55 50 45 R 40 35 30 25 0 10 20 30 40 50 T 60 70 80 90 100 Figure 8.1227 Smoothed R values evaluated from straight line are: R_smoothed = 8−12 Numerical Analysis Using MATLAB® and Excel®./R_exper format short % Return to default format a=0: 10: 100. Third Edition Copyright © Orchard Publications .5 MATLAB also displays the following data: Coefficients m and b are: 0.*100. plot(a. 60 R (Ohms) vs T (deg Celsius.T) % Compute and display the values of the fitted % polynomial at same points as given % (experimental) values of R R_exper = R % Display the experimental values of R for comparison % The statement below computes the percent error between % the fitted polynomial and the experimental data disp('% Error at points of given values is:') % The percent error is computed with the following statement error=(R_smoothed−R_exper).Chapter 8 Linear and Parabolic Regression % Define range to plot the polynomial % Compute p for each value of a % Plot the polynomial % Display the coefficients m and b fprintf('\n') % Insert line disp('Coefficients m and b are:').

3)=R_smoothed'.10 40.00 Columns 6 through 10 42.00 51.34 Columns 6 through 10 -0.4f\t %5. % Insert line fprintf('%3.Parabolic Regression Columns 1 through 5 28.4f\n'.53 45. Third Edition Copyright © Orchard Publications 8−13 .00 33.88 0. % Insert line fprintf('Temp \t Exper R\t Smoothed R \t |Error| \n') fprintf(' \n').00 -0. % x−axis data R=[27.30 Column 11 56.01 -0. % 1st column of matrix y(:. % 4th column of matrix fprintf(' \n').*100.1). % Construct an 11 x 4 matrix of zeros y(:. R_smoothed=polyval(p.70 % Error at points of given values is: error = Columns 1 through 5 1.65 54.17 -0.77 51.93 R_exper = Columns 1 through 5 27. % y−axis data p=polyfit(T.6 45.50 48.60 45. y=zeros(11.41 We can make the displayed data more presentable by displaying the values in four columns.29 Column 11 56.88 Columns 6 through 10 42. % 2nd column of matrix y(:.0 56.00 34.R.60 31.63 0.2)=R_exper'.05 37.4f\t %5.0 40.20 -0.7].1 54. % 3rd column of matrix y(:.17 39. The following MATLAB script will do that and will display the error in absolute values.12 31. T= [0 10 20 30 40 50 60 70 80 90 100].00 54.0 34.5 48.09 Numerical Analysis Using MATLAB® and Excel®.y') 36./R_exper.0 42.41 48.0 37.4)=abs(error)'.89 0.02 Column 11 0. R_exper = R.6 31.4).14 -0.1)=T'. error=(R_smoothed−R_exper).T).3 51.0f\t %5.

25 0.50 0.7655 39.1227 31.6339 0. 8−14 Numerical Analysis Using MATLAB® and Excel®.75 0. Third Edition Copyright © Orchard Publications .0036 33.6 Experimental Data Volts ma Volts ma 0.14 4.1707 0.01 3.4 Data for Example 8.85 1.18 1.75 0.0226 0.1370 0.75 1.6000 31. Temp 0 10 20 30 40 50 60 70 80 90 100 Exper R Smoothed R |Error| 27.23 4.50 0.4.42 0.6 The voltages (volts) shown in Table 8.05 3. % Insert line When this script is executed.00 0.00 1.5000 48.75 0.75 0.50 0.3000 51.39 2.Chapter 8 Linear and Parabolic Regression fprintf(' \n').00 1. we can draw a smooth curve and compute the coefficients of a power series by approximating the derivatives di ⁄ dv with finite differences Δ i ⁄ Δ v .0000 40.1700 54.50 1.0117 0.60 0.10.0943 0.25 1.9318 1. Use the power series method to derive a polynomial that best approximates the given data.4082 48.63 2.8845 36. and the current ma (milliamps) values were observed and recorded.50 0.0000 34.0000 56. The following example illustrates the procedure. MATLAB displays the following where the error is in percent.4 Regression with Power Series Approximations In cases where the observed data deviate significantly from the points of a straight line.11 4.00 1. were applied across the terminal of a non−linear device.0000 37.8939 0.1000 54.08 3.00 0.91 2.6464 42.25 0.00 2.0509 56. Example 8.03 3. TABLE 8.34 Solution: We begin by plotting the given data and we draw a smooth curve as shown in spreadsheet of Figure 8.25 0.7000 28.2891 51.72 1.2018 0.18 4.3396 0.0000 42.00 0.25 0.6000 45.8841 0.50 0.5273 45.00 0.28 5.4089 8.

50 0.80 1.85 1.00 -0. TABLE 8.20 1.00 4.25 1.20 0 1 2 3 v 4 5 6 Smoothed Experimental Data From this plot.23 0.18 0.63 2.72 1.06 3.00 1.00 0. In cell E4 we enter the formula =(B5-B4)/(A5-A4) and we copy it down to E5:E23.75 0.40 0.20 0.39 1.63 1.18 4.6 Using the plot of Figure 8.00 0. Numerical Analysis Using MATLAB® and Excel®.25 3.75 1.00 1.25 1.75 0.00 ma 0.25 0.75 5.10 we read the voltmeter reading and the corresponding smoothed ma readings and enter the values in Table 8.22 4.50 2.00 3.02 Figure 8. i | v =0 = i(0) = −0.40 2.11 4.34 0.60 0. we enter these values in the spreadsheet of Figure 8.00 1.50 0.00 1.00 0.5 Data for the first derivative Smoothed Data for Computation of Δi / Δv Volts ma Volts ma 0.02 2.75 2.60 0.33 −0.25 2.40 1.04 3.60 1.50 1.00 0.75 3.00 0.03 0.10.25 0.05 0.91 ma 0.25 0.85 1.20 1.50 4.75 1.18 1.50 1.00 2.80 0.11 2.00 0. Spreadsheet for Example 8.09 3.14 4. … 20 To facilitate the computations.00 0.01 3.14 0.00 1. Third Edition Copyright © Orchard Publications 8−15 .49 0.25 4.25 0.75 4.50 0.50 0.Regression with Power Series Approximations Experim ental volts 0.01 0.01 1.28 0. we compute Δ i ⁄ Δ v for i = 1.41 Next.08 0.75 0.60 0.72 0. 2.75 0.5.50 3.89 2.25 0.11.50 0.27 5.42 0.50 0.

12 0.04-(0.25-0.00 3.02 0.00 B Smoothed ma -0.33 0.00)= 0.25 4.02 + 0.01))/(0.02 2 2 2 2 2 2 2 (8.64 0.50 1.12 0.63))/(0.16 0.75 5.75 1.48 0.32 0.08 0.5 ( – 0.89-(1.25-0.92 1.50 4.6 Next.08 0.20 0.75 3.25 2.25 0.25-0.25 1.12 0.02))/(0.16 0.00 2.11.50 3.04v + 0.76 0. Following the same procedure we can find higher order derivatives.27 0. Spreadsheet for computation of Δ i ⁄ Δ v in Example 8. for this example we will consider only the first three terms of the polynomial whose coefficients i .01 1.14 0.80 0.06 0.63 Δi20 / Δv20= (1. the polynomial that 1.75 2. However. Third Edition Copyright © Orchard Publications .04 … … … … … … … … … … … … … … … … … … … … … … … … … … … … … … … … … … 1.00 0. we plot Δ i ⁄ Δ v versus volts and again we smooth the data as shown in Figure 8. Therefore.12 0. Finally.11 0.08 ) v i ( v ) = i ( 0 ) + i′ ( 0 ) + ---2! = – 0.89 C D E Computed Δi / Δv Δi1 / Δv1= (0. The smoothed values of the plot of Figure 8.00)= Δi2 / Δv2= (0.00 4.41 0.75 4.40 1.01-(-0.12v + 0.49 0.00)= Figure 8.12v – 0.09 0.16) 8−16 Numerical Analysis Using MATLAB® and Excel®.12.52 0. and from these we compute Δ i ⁄ Δ v .01 0.24 0.85 1.13.′′ 2 i ( 0 ) + … = – 0.22 0.14.25 3.44 0.04 0.Chapter 8 Linear and Parabolic Regression A 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Volts 0.50 2.18 0.00 1. we plot the computed values of Δ i ⁄ Δ v versus v and again we smooth the data as shown in the spreadsheet of Figure 8.60 0.20 1.50 0.72 0.12 are shown in Figure 8.32 0. all three evaluated at v = 0 and are read from the plots. Δ i ⁄ Δ v and best fits the given data is Δ i ⁄ Δ v .

27 0.6 0. Use n = 3 to compute the best cubic polynomial approximation.00 3.64 0.80 0.20 0.25 2. Third Edition Copyright © Orchard Publications 8−17 .32 0.00 Δi / Δv 0. The MATLAB script below computes the smoothed line and produces the plot shown on Figure 8.01 0.25 0.50 4.0 0.n) function. (Δi / Δv) | v=0 = i'(0) = 0.y.75 1.12.16 0.04 Smoothed Δ i / Δ v 1.25 1.44 0.12 0.89 1. Plot to obtain smoothed data for Δ i ⁄ Δ v in Example 8.72 0.16 0.92 1.6 Example 8.63 1. In this example we will use n = 3 as we did with Excel.40 1.60 0.20 Figure 8.00 0.04 0.12 0.02 0.08 0.0 0 1 2 volts 3 4 5 1.12 0.08 0.24 0.n) function.2 1.25 3.75 5.52 0.75 2.50 2.75 4.09 0.12 0.2 0.y.Regression with Power Series Approximations Com puted Volts 0.12 Δi / Δv -0.50 3.76 0.00 2.00 1.75 3.50 1.48 0.01 0. Numerical Analysis Using MATLAB® and Excel®.15.8 0.18 0.49 0.32 0.4 using the MATLAB polyfit(x.41 0.14 0.25 4. Solution: With MATLAB.11 0.7 Repeat Example 8.22 0.85 1. where n ≥ 2 .06 0.00 4.50 0.33 0. higher degree polynomial regression is also performed with the polyfit(x.4 From this plot.

25-0.ma.25:5.32 0.75 4.50 4.a).25 0.00 0.10-011)/(0.75 5].32 0. % Establishes desired x and y axes limits plot(v.25 4.04 0.20 0.64 0.75 2 2.04 -0.12 0.08 0.72 0.50 0..25 1.00 0.36 0.04 0.3) % Fits a third degree polynomial to % the data and returns the coefficients % of the polynomial (cubic equation for % this example since n=3) a=0:0.82 0.20 0.5 0..25 1.12 0.75 5.25 3.11 0.50 3.35 0.28 0.14 0.91].36 0.25 2.12 0.75 1.75 3. 3.11 0.ma.24 0.'+r').28.5 1.23 0.34 0.00)= -0.00 4.29 0.00 1. % y−axis data axis([−1 6 −1 2]).05 0.10 0.11 0. Third Edition Copyright © Orchard Publications .92)/(0.5 3.75 4 4.00)= 1.00)= (0.25-0.03 0.24 0.50 1.5 2. % Define range to plot the polynomial q=polyval(p.25 3. grid % Indicate data points with + and straight line in red % hold % hold current plot so we can add other data disp('Polynomial coefficients in descending order are: ') % p=polyfit(v.12 0.56 0.5 4. Spreadsheet for computation of Δ i ⁄ Δ v in Example 8.25 2..16 0.00 3.11-0.14 0.85 1.18 1.03 Δi220 / Δv220= Figure 8.10 0.12)/(0.40 0.04 0..6 v=[0 0.50 0.75 2.25 4..01 0.00 1.Chapter 8 Linear and Parabolic Regression A 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 B C D E Computed 2 2 Δi / Δv 2 1 2 Δi 2 Smoothed Δi/Δv Volts 0.42 0.03-0.41 0.92 Δi / Δv21= / 2 Δv 2= (0.08 0.75 1 1.00 2.75 3.25 0.60 0.44 … … … … … … … … … … … … … … … … … … … … … … … … … … … … … … … … … … (1. 0.18 0. % x−axis data ma=[0 0.63 1.% Calculate p at each value of a % continued on the next page 8−18 Numerical Analysis Using MATLAB® and Excel®.24 0.50 2.25-0.73 0.00 2 2 0.39 1..48 0.44 0.17 0.13.

35 0.*100.48 0.2)=ma_exper'.12 0.64 0.11 0.00 0.75 3.08 v 2 2 0.92 4 5 0.25 2.14 0.14.10 0.3)=ma_smooth'..20 0.10 0.11 0.75 2. n=3').2 0.25 4. Third Edition Copyright © Orchard Publications 8−19 .24 0. % 4th column of matrix fprintf(' \n'). % 1st column of matrix y(:.16 0..v). % Insert line % continued on the next page Numerical Analysis Using MATLAB® and Excel®.04 0.56 0.20 0.40 0. % 3rd column of matrix y(:.75 4.25 1. (Δi21 3 From this / Δv2) | v=0 =2 i''(0) = − 0.04 0.75 1.q).12 0. % y=zeros(21. ylabel('ma') % Plot the polynomial % Display actual.50 2.00 1. Plot to obtain smoothed data of Δ i ⁄ Δ v in Example 8.3 0.4).28 0.29 0.00 0.1 0 plot.08 0.41 0./(ma_exper+eps). smoothed and % error values ma_smooth=polyval(p.50 0. % Construct a 21 x 4 matrix of zeros y(:.00 3.00 2. % The following statement computes the percent error between the % smoothed polynomial and the experimental (given) data error=(ma_smooth−ma_exper)..12 0.36 0.32 0.73 Figure 8. xlabel('v').1 0.24 0.Regression with Power Series Approximations Com puted Volts 0.25 0.04 -0.44 2 2 Smoothed Δ i / Δ v 0.82 0.25 3.32 0.00 4.0 -0.50 1.4)=abs(error)'. % Calculate the values of the fitted polynomial ma_exper = ma.36 0.12 0.1)=v'.5 0. title('milliamps vs volts.50 Δ i2 / Δ v2 -0.6 % plot(a. % 2nd column of matrix y(:.04 0.50 3.17 0.4 0.24 0.

25 1.18000 0.00 2.50 2.50000 0.60000 -0.14441 0.16677 3.2f\t %7.5 v 3 3.50 1.5f\t %7.06298 0.5f\n'. fprintf(' \n').03828 0.34000 0.61614 25. fprintf('%3.27275 0.50 0.58785 1.5 2 2.51632 2.75 2.01000 0.57123 8−20 Numerical Analysis Using MATLAB® and Excel®.5 1 ma 0.75 1.0177 volts 0.28000 0.25 2.42000 0.01766 0.08775 0.25 Exper ma Smoothed ma 0.05000 0.y') fprintf(' \n').25 0.03000 0.00 0.00000 0.1275 -0.5 0 -0.11000 0. MATLAB computes and displays the following data.69226 3.Chapter 8 Linear and Parabolic Regression 2 milliamps vs volts.5 0 0.7 fprintf('volts \t Exper ma\t Smoothed ma \t |%%Error| \n').01197 0.5 1 1. Plot for Example 8.59657 |%Error| 7955257388080461.33393 0. Third Edition Copyright © Orchard Publications .00000 19.5 4 4.5f\t %7.78451 3.93513 3.49413 0.14852 0.0182 -0.17970 0.15.05797 1.74402 27. Polynomial coefficients in descending order are: p = 0.08000 0.95052 9. n=3 1.22191 0.11433 0.17324 0.14000 0.5 5 Figure 8.23000 0.0403 0.40716 0.75 3.00 1.00 3.

x=0:2:16.85467 1. y ( 10 ). plot the function y = sin x ⁄ x (8.50 3.75 4. title('(sinx)/x curve for x > 0') The plot for the function of (8.00 0.00000 1.91000 0. y ( 14 ). seventh. p9=polyfit(x.7).x) functions to find a suitable polynomial that best fits the x and y data. y ( 8 ).09538 1..85000 1.16. Third Edition Copyright © Orchard Publications 8−21 .8 Use MATLAB to a.x) functions.89005 0.n) and polyval(p.40048 1.. compute y ( 0 ).75362 0. % and ninth degree polynomials % continued on the next page Numerical Analysis Using MATLAB® and Excel®.75 5.71618 0. y=sin(x).17) in the interval 0 ≤ x ≤ 16 radians.9). b. y ( 4 ). plot y versus x for these values and use the MATLAB polyfit(x..04436 We will conclude this chapter with one more example to illustrate the uses of the MATLAB polyfit(x.00 4.y.y. grid.17) is shown in Figure 8./(x+eps)'.39000 1.y. y ( 16 ) c. y ( 6 ).63155 1. The fplot function below plots y = sin x ⁄ x . We added eps to avoid division by zero.Regression with Power Series Approximations 3.54911 1.01374 1. We use the script below to evaluate y at the specified points. y ( 2 ).50 4.n) and polyval(p.25 4.y./(x+eps) p7=polyfit(x. % of x and y with fifth. Solution: a. fplot('sin(x).63000 1.19511 1.28020 0.37399 1.[0 16 −0.18000 1. Example 8. b.72000 0. y ( 12 ).5 1]).53040 0.

0708 −0.y.1. % Compute the polynomials for this range of x values. p5=polyfit(x.y_ref−0.16.5.1.x_ref.p5_pol.0544 −0.1237 −0.'−'.Chapter 8 Linear and Parabolic Regression 1 (sinx)/x curve for x > 0 0.5 0 2 4 6 8 10 12 14 16 Figure 8.'−. % Fits the polynomial to the data x_span=0: 0.3].3.x. x_span). p7_pol=polyval(p7. grid.8 c.y.y_ref.'−.0180].5.4. '7th degree polynomial'). x_span). '9th degree polynomial').0.1892 −0.5 0 -0.5. y_ref=[1. x_span.7) disp('The coefficients of 9th order polynomial in descending order are:') p9_coef=polyfit(x. hold off format short e % Exponential short format disp('The coefficients of 5th order polynomial in descending order are:') p5_coef=polyfit(x. text(5. plot(x_span. x_span).y. '5th degree polynomial').1.p7_pol.3]. The script for finding a suitable polynomial is listed below. x_span.1: 16.'. text(5. % The following are line legends for each curve plot(x_ref. hold on.'−−'.5) disp('The coefficients of 7th order polynomial in descending order are:') p7_coef=polyfit(x.9) format short % We could just type format only since it is the default 8−22 Numerical Analysis Using MATLAB® and Excel®. % The following two statements establish coordinates for three legends % in x and y directions to indicate degree of polynomials x_ref=[2 5.'*').p9_pol.y_ref−0. % Specifies values for x−axis p5_pol=polyval(p5. Third Edition Copyright © Orchard Publications . % The following are text legends for each curve text(5. x=[0 2 4 6 8 10 12 14 16]. Plot for Example 8.1. y=[1 0.x_ref.2.9.5).y.'−−'.3.4546 −0. p9_pol=polyval(p9.'.0447 0.y.0466 0.'−').

4318e-004 -4.3560e-001 Columns 7 through 8 7.Regression with Power Series Approximations The 5th.2508e-001 9. 7th.6 0.4 0.8 The coefficients of the 5th.4 0 2 4 6 8 10 12 14 16 Figure 8.9437e-002 1. Third Edition Copyright © Orchard Publications 8−23 .6672e-004 -5.17.1923e-006 9.9165e-001 The coefficients of 9th order polynomial in descending order are: -4. 7th.4572e-002 -1.5032e-004 -9.4 5th degree polynomial 1.2 0 -0.0000e+000 Numerical Analysis Using MATLAB® and Excel®.17.8 0. and 9th order polynomials are shown in Figure 8.9965e-001 p9_coef = Columns 1 through 6 -1.0293e+000 The coefficients of 7th order polynomial in descending order are: p7_coef = Columns 1 through 6 2.1644e-003 -9.0057e-001 0 4.8340e-005 4. 1.0067e-002 -1. The coefficients of 5th order polynomial in descending order are: p5_coef = 6.5865e-006 6.7650e-003 Columns 7 through 10 -8. and 9th order polynomials are shown below.2 -0.6529e-001 -1.5825e-003 1.2092e-002 3.0444e-008 1.2 7th degree polynomial 1 9th degree polynomial 0. Polynomials for Example 8.6483e-006 -1.

cubic.) • The best fitting straight line or curve has the property that d 1 + d 2 + … + d 3 = minimum and it is referred to as the least−squares curve. we compute the coefficients m (slope) and b (y-intercept) of the straight line equation y = mx + b such that the sum of the squares of the errors will be minimum. and c are found from ( Σ x ) a + ( Σ x ) b + nc = Σ y ( Σ x ) a + ( Σ x ) b + ( Σ x ) c = Σ xy (Σx )a + (Σx )b + (Σx )c= Σx y 4 3 2 2 3 2 2 where n = number of data points. If it is a parabola. Σ x = sum of the numbers x squared n = number of data x 2 The values of m and b are computed from D1 m = ----Δ Σx Σx Σx n 2 where Δ = D2 b = ----Δ Σ xy Σ x Σy n Σ x Σ xy Σx Σy 2 2 D1 = D2 = • We find the least−squares parabola that fits a set of sample points with y = ax + b + c where the coefficients a. The values of m and b can be found from the relations ( Σ x ) m + ( Σ x ) b = Σ xy 2 2 2 2 where ( Σ x ) m + nb = Σ y Σ x = sum of the numbers x . 8−24 Numerical Analysis Using MATLAB® and Excel®.Chapter 8 Linear and Parabolic Regression 8. we call it a least−squares parabola. Third Edition Copyright © Orchard Publications . b. With this method. Σ y = sum of the numbers y Σ xy = sum of the numbers of the product xy . Regression can be linear (straight line) or curved (quadratic. etc.5 Summary • Curve fitting is the process of finding equations to approximate straight lines and curves that best fit given sets of data. • Regression is the process of finding the dependent variable y from some data of the independent variable x . A straight line that satisfies this property is called a least squares line. • We perform linear regression with the method of least squares.

• In cases where the observed data deviate significantly from the points of a straight line. if n = 3 .Summary • With MATLAB. Thus. that is. and returns the coefficients m and b . MATLAB computes the best straight line approximation. it computes the best cubic polynomial approximation. where x and y are the coordinates of the data points. Likewise.y. Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications 8−25 . we can draw a smooth curve and compute the coefficients of a power series by approximating the derivatives dy ⁄ dx with finite differences Δ y ⁄ Δ x . and so on.n) command. linear regression. it computes the best quadratic polynomial approximation and returns the coefficients of this polynomial. if n = 1 . If n = 2 . regression is performed with the polyfit(x. and n is the degree of the polynomial.

Chapter 8 Linear and Parabolic Regression 8.80 200 0.70 180 0.11).55 140 0. A sales manager wishes to forecast sales for the next three years for a company that has been in business for the past 15 years.1 to compute the straight line equation that best fits the given data.y. 3. 5. The sales during these years are shown on the next page.n) and polyval(p.x) functions. Repeat Exercise 2 above using Excel’s Trendline feature. Consider the system of equations below derived from some experimental data. Repeat Exercise 2 above using the MATLAB’s polyfit(x.10) and (8. Third Edition Copyright © Orchard Publications .45 120 0.6 Exercises 1. 2. measurements yielded the following sets of values: millivolts milliamps 100 0. 2x + y = – 1 x – 3y = – 4 x + 4y = 3 3x – 2y = – 6 – x + 2y = 3 x + 3y = 2 Using the relations (8.85 Use the procedure of Example 8. In a non−linear device. 4. find the values of x and y that best fit this system of equations.60 160 0. 8−26 Numerical Analysis Using MATLAB® and Excel®.

837 207.127 39.456.547.048.Exercises Year 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Sales $9. 6.545.547.678.574 130. compute the sales for the next three years.149.456.687 28. Numerical Analysis Using MATLAB® and Excel®.632 206.216 112.642.745 19.784 54. Repeat Exercise 5 above using the MATLAB polyfit(x.548 13.321 148.456.987 Using Excel’s Trendline feature.983 176.147.453.n) and polyval(p.873.147. You may round the sales to the nearest thousand.782 89. Third Edition Copyright © Orchard Publications 8−27 .352 204.163.x) functions. choose an appropriate polynomial to smooth the given data and using the polynomial found.369 72.y.

x = – 1. A B C D E F G H I J 1 Spreadsheet for Exercise 8.1 2 2 2 a b c a ab b ac bc 3 4 5 2 1 -1 4 2 1 -2 -1 6 1 -3 -4 1 -3 9 -4 12 7 1 4 3 1 4 16 3 12 8 3 -2 -6 9 -6 4 -18 12 9 -1 2 3 1 -2 4 -3 6 10 1 3 2 1 3 9 2 6 11 7 5 -3 17 -2 43 -22 47 12 Σ 13 2 Σ a Σ ab 14 17 -2 15 Δ = = 727 2 Σ ab Σ b 16 -2 43 x=D1/Δ= -1.7 Solutions to End−of−Chapter Exercises 1. Third Edition Copyright © Orchard Publications . We construct the spreadsheet below by entering the given values and computing the values from the formulas given.Chapter 8 Linear and Parabolic Regression 8.039 8−28 Numerical Analysis Using MATLAB® and Excel®.039 21 2 Σ a Σ ac 22 17 -22 23 D2 = = 755 Σ ab Σ bc 24 -2 47 Thus.172 and y = 1.172 17 Σ ac Σ ab 18 -22 -2 19 D1 = = -852 2 Σ bc Σ b 20 47 43 y=D2/Δ= 1.

95 142000 120 140 160 180 200 Millivolts Σx 2 Σx = n Σx = n Σ xy = Σy 142000 900 621 4.0476 621 = 4.80 0.0 2000 171 42000 Σx Σ xy Σy 2 Σx Σx Thus.55 14400 0.80 32400 0. Spreadsheet for Exercise 8.004x + 0.40 100 0. Third Edition Copyright © Orchard Publications 8−29 .60 19600 0. We construct the spreadsheet below by entering the given values and computing the values from the given formulas.0476 Numerical Analysis Using MATLAB® and Excel®. y = mx + b = 0.00 Milliamps 0.0 142000 900 900 = 6 m=D1/Δ = 0.004 900 = 6 b=D2/Δ = 0.60 0.85 40000 3.70 25600 0.Solutions to End−of−Chapter Exercises 2.2 x (mV) y(mA) 100 120 140 160 180 200 900 x2 xy 45 66 84 112 144 170 621 Milliam ps versus Millivolts 1.45 10000 0.

ma=[0.95 142000 120 140 160 180 200 Millivolts Σx 2 Σx = n Σx = n Σ xy = Σy 142000 900 621 4. plot(mv. n=1'). hold % Hold current plot so we can add other data p=polyfit(mv. % Compute p for each value of a plot(a.a).0 142000 900 900 = 6 m=D1/Δ = 0.1).70 0.ma.0476 621 = 4.'*b').45 10000 0. % x-axis data % y-axis data % Establishes desired x and y axes limits % Display experimental (given) points with % asterisk and smoothed data with blue line grid.q) % Plot the polynomial % Display the coefficients m and b fprintf('\n') % Insert line mv= [100 120 140 160 180 200].00 Milliamps 0.80 32400 0. Trendline for Exercise 8.85 40000 3. we obtain the trendline shown below.80 0.ma.85].70 25600 0.55 0. title('ma (milliamps) vs mv (millivolts.2.40 100 0.55 14400 0.80 0.004 900 = 6 b=D2/Δ = 0. Third Edition Copyright © Orchard Publications . axis([100 200 0 1]).60 19600 0. ylabel('ma'). xlabel('mv'). % Fits a first degree polynomial (straight line since n =1) and returns % the coefficients m and b of the straight line equation y = mx + b a=0: 10: 200.0 2000 171 42000 Σx Σ xy Σy Σx 2 Σx 4. Following the procedure of Example 8.60 0. 8−30 Numerical Analysis Using MATLAB® and Excel®.Chapter 8 Linear and Parabolic Regression 3. % Define range to plot the polynomial q=polyval(p.60 0.45 0.3 x (mV) y(mA) 100 120 140 160 180 200 900 x2 xy 45 66 84 112 144 170 621 Milliam ps versus Millivolts 1.

4)=abs(error)'.1361 2. % Display the experimental values of ma for comparison % The statement below computes the percent error between % the fitted polynomial and the experimental data % disp('% Error at points of given values is:').4500 0.4548 0.y') fprintf(' \n').4 0.8000 0.4f\n'.mv).4).6000 0.4405 1. % 2nd column of matrix y(:. % 1st column of matrix y(:.5500 0.9365 0. % Construct an 6 x 4 matrix of zeros y(:.3 0. % Compute the values of the fitted polynomial at % same points as given (experimental) values of ma ma_exper = ma.1 0 0 20 40 60 80 100 mv 120 140 160 180 200 0.1)=mv'.Solutions to End−of−Chapter Exercises disp('Coefficients m and b are:'). % The percent error is computed with the following statement error=(ma_smoothed-ma_exper).2 0.0041 mv 100 120 140 160 180 200 0. % 3rd column of matrix y(:.2)=ma_exper'.0582 2. % 4th column of matrix fprintf(' \n').9 0.6990 0.8500 0.8 0.5362 0.6 ma 0. % Insert line Coefficients m and b are: 0.7 0. % Insert line fprintf('mv \t Exper ma\t Smoothed ma \t |Error| percent \n') fprintf(' \n').8619 Numerical Analysis Using MATLAB® and Excel®.3)=ma_smoothed'. n=1 Exper ma Smoothed ma 0./ma_exper.7000 0.0f\t %5.*100. fprintf('\n'). % Insert line fprintf('%3.4006 ma (milliamps) vs mv (millivolts.5 0.7805 0.6176 0. Third Edition Copyright © Orchard Publications 8−31 .0476 |Error| percent 1.5108 2.4f\t %5. disp(p). format bank % Two decimal place display will be sufficient ma_smoothed=polyval(p. format short % Return to default format y=zeros(6.4f\t %5.

p=polyfit(year.2E+06x 2 + 1E+07x . Third Edition Copyright © Orchard Publications .Chapter 8 Linear and Parabolic Regression 5. 15. disp(p). These results indicate that non-linear interpolation is. % Define range to plot the polynomial q=polyval(p. title('Yearly Sales vs Years. 8−32 Numerical Analysis Using MATLAB® and Excel®. 15). xlabel('Years'). We will compare these values with the results of Exercise 6.4. fprintf('\n').sales. 148678983 176453837 207547632 206147352 204456987].. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 9149548 13048745 19147687 28873127 39163784 54545369 72456782 89547216 112642574 130456321 148678983 176453837 207547632 206147352 y = -17797x 4 + 436354x 3 . n=4').. in most cases. % Display experimental (given) points with % asterisk and smoothed data with blue line hold % Hold current plot so we can add other data grid. % Compute p for each value of a plot(a..4)..9966 258000000 208000000 158000000 108000000 58000000 8000000 0 5 10 15 20 The sales for the next 3 years are from the equation above produced by Excel. year= [1 2 3 4 5 6 7 8 9 10 11 12 13 14 15].a). % x-axis data sales=[9149548 13048745 19147687 28873127 39163784 . % y-axis data plot(year.'*b'). ylabel('Yearly Sales'). unreliable.q) % Plot the polynomial % Display coefficients ofpolynomial fprintf('\n') % Insert line disp('Coefficients are:'). we choose Polynomial 4 and we obtain the trendline shown below. y 16 = – 17797x + 436354x – 2 × 10 x + 10 x – 2 × 10 y 17 = – 17797x + 436354x – 2 × 10 x + 10 x – 2 × 10 y 18 = – 17797x + 436354x – 2 × 10 x + 10 x – 2 × 10 4 3 6 2 7 4 3 6 2 7 4 3 6 2 7 6 x = 16 6 x = 17 6 x = 18 = 266961792 = 247383965 = 206558656 6. Following the procedure of Example 8. % Fits a first degree polynomial (n=4) and returns % the coefficients of the polynomial a=linspace(0.2E+06 R2 = 0. 54545369 72456782 89547216 112642574 130456321 .sales.

Coefficients are: 1.5 2 1.Solutions to End−of−Chapter Exercises % Compute the values of the fitted polynomial at % same points as given (experimental) values of ma sales_exper = sales.67 11. y=zeros(15. % Construct an 15 x 4 matrix of zeros y(:.4).0f\t %9. fprintf('%2.3)=sales_smoothed'. % 1st column of matrix y(:.4)=abs(error)'.5 0 -0.year). % Display the experimental values of ma for comparison % The statement below computes the percent error between % the fitted polynomial and the experimental data % The percent error is computed with the following statement error=(sales_smoothed-sales_exper).63 Numerical Analysis Using MATLAB® and Excel®.5 Yearly Sales 1 0. % 4th column of matrix fprintf(' \n'). n=4 0 5 Years 10 15 year Exper sales Smoothed sales 1 2 3 4 9149548 13048745 19147687 28873127 7258461 14529217 21374599 29344934 |Error| percent 20. Third Edition Copyright © Orchard Publications 8−33 .*100.1)=year'.63 1.y') fprintf(' \n').5 x 10 8 0.2)=sales_exper'.0436 -0.2f\n'./sales_exper. % 3rd column of matrix y(:. sales_smoothed=polyval(p. fprintf('year\t Exper sales\t Smoothed sales \t |Error| percent \n') fprintf(' \n').0f\t %9.2386 1.1641 -0.0018 2.35 11. % 2nd column of matrix y(:.0e+007 * -0.2415 Yearly Sales vs Years.0f\t %5.

36*10^5*x^3−2. y18 y16 = 1.386*10^6*x^2+1.56 5.8 × 10 x + 4.13 1. 8−34 Numerical Analysis Using MATLAB® and Excel®.7243e+007 These values vary significantly from those of Exercise 5.1641*10^7*x−2. y17=−1.30 From the coefficients produced by MATLAB. y17. shown on the previous page.02 3.27 5.34 4.96 2. For extrapolation.386 × 10 x + 1.36 × 10 x – 2. linear regression gives the best approximations.386*10^6*x^2+1. We should remember that the equations produced by both Excel and MATLAB represent the equations that best fit the experimental values.22 1. y16=−1.1641 × 10 x – 2.1641*10^7*x−2.415*10^6. non-linear interpolation especially for polynomials of fourth degree and higher give inaccurate results. x=18.4462e+008 y18 = 8. Third Edition Copyright © Orchard Publications . x=17. we form the polynomial y = – 1.Chapter 8 Linear and Parabolic Regression 5 6 7 8 9 10 11 12 13 14 15 39163784 54545369 72456782 89547216 112642574 130456321 148678983 176453837 207547632 206147352 204456987 39563426 52726163 69102111 88533118 110433913 133792104 157168183 178695519 196080363 206601848 207111986 1.53 0.36*10^5*x^3−2.386*10^6*x^2+1.8*10^4*x^4+4.71 1. As stated above.36*10^5*x^3−2.63 1.1641*10^7*x−2.8*10^4*x^4+4.7923e+008 y17 = 1. y16.8*10^4*x^4+4.415*10^6.415*10^6. y18=−1.415 × 10 4 4 5 3 6 2 7 6 and from it we find the values of y (the yearly sales) as follows: x=16.

y''' 0 h 3 -y h + … y 1 = y 0 + y' 0 h 1 + ---1 + ---1 + ---2! 3! 4! 0 1 (9. close to x 1 . x 1 = 0.1 Taylor Series Method We recall from Chapter 6 that the Taylor series expansion about point a is n f'' ( a ) f (a) . x 3 = 0.y''' h 3 + ----y h y i + 1 = y i + y' i h i + 1 + ---+… 2! i i + 1 3! i i + 1 4! i i + 1 (9. (9. For another value x 2 > x 1 . Such approximations are necessary when no exact solution can be found.5 with the initial condition y ( 0 ) = 1 . Adams’. Runge−Kutta. The Taylor. 111.1). 1 1 1 (4) 4 .5) correct to four decimal places for values x 0 = 0.( x – a ) 2 + … + ----------------(x – a) y n = f ( x ) = f ( a ) + f' ( a ) ( x – a ) + ----------n! 2! (n) (9. 9.( 4 ) 4 2 3 y''1 h 2 + ---y''' y h +… y 2 = y 1 + y'1 h 2 + ---1 h 2 + ---2! 3! 4! 1 2 (9. we obtain: 1 1 1 (4) 4 . to minimize the error f ( x 1 ) – y 1 we need to keep h 1 sufficiently small.y'' h 2 + ---. and Milne’s methods are discussed.1) Now.2 .y''0 h 2 .1 . Third Edition Copyright © Orchard Publications 9−1 . Letting h 1 = x – a in (9.1 Use the Taylor series method to obtain a solution of y' = – xy x 5 = 0.0 . and Numerical Analysis Using MATLAB® and Excel®.Chapter 9 Solution of Differential Equations by Numerical Methods T his chapter is an introduction to several methods that can be used to obtain approximate solutions of differential equations. if x 1 > a is a value close to a . we repeat the procedure with h 2 = x 2 – x 1 .2) Obviously. x 4 = 0.4) Example 9.4 . then.3) In general. x 2 = 0. we can find the approximate value y 1 of f ( x 1 ) by using the first k + 1 terms in the Taylor expansion of f ( x 1 ) about x = a .3 .

0000 6 0.84 1.0016 -0.1.8) = ( x i – 6x i + 3 ) y i 4 2 where x i represents x 0 = 0.0 = 0.2 . Third Edition Copyright © Orchard Publications .0640 0.4 -0.0 -1. and 4 . Using the values of the coefficients of y i in (9.91 0. 1.000 3.8).3 -0.4).2 0.0000 0.4681 9 0.04 0.1 0.Chapter 9 Solution of Differential Equations by Numerical Methods Solution: For this example. A B C D E F G H 1 Differential Equation is y' = −xy 2 Numerical solution by Taylor method follows 3 2 3 4 2 3 4 2 xi xi xi -xi xi -1 -xi +3xi xi -6xi +3 xi 4 5 0.000004y i (4) (9.1 9−2 Numerical Analysis Using MATLAB® and Excel®. h = x 1 – x 0 = 0.99 0.0256 -0.0081 -0.96 0.873 2. Spreadsheet for Example 9.3 0.3 . 2.0656 10 0. and x 4 = 0.7616 8 0.136 2.0 0.0270 0.0001 -0.2 -0. y i + 1 = y i + 0.299 2.0000 0.7) = ( – x + 3x ) ( – xy ) + ( – 3x + 3 ) y = ( x – 6x + 3 ) y 3 2 4 2 We use the subscript i to express them as y' i = – x i y i y'' i = ( x i – 1 ) y i y''' i = ( – x i + 3x i ) y i yi (4) 3 2 (9. x 3 = 0.5) are: y' = – xy y'' = – xy' – y = – x ( – xy ) – y = ( x – 1 ) y y''' = ( x – 1 ) y' + 2xy = ( x – 1 ) ( – xy ) 2xy = ( – x + 3x ) y y (4) 2 2 2 3 (9.00 0. x 1 = 0.09 0.0010 0.1 .4 .1y' i + 0.4 Figure 9.005y'' i + 0.1 – 0. 3.6) for i = 0.592 2. we construct the spreadsheet of Figure 9.9401 7 0.1 and by substitution into (9.0 .16 0. The first through the fourth derivatives of (9.1 -0. x 2 = 0.1.00 0.01 0.0080 0.000167y''' i + 0.

6).8).7616y 2 4) y( 3 = ( x 3 – 6x 3 + 3 ) y 3 = 2.873y 3 y''' 4 = ( – x 4 + 3x 4 ) y 4 = 1.84 y 1 y''' 0 = ( – x 0 + 3x 0 ) y 0 = 0 y''' 1 = ( – x 1 + 3x 1 ) y 1 = 0. and H6:H10 of the spreadsheet of Figure 9.0656y 4 4 2 4 2 4 2 4 2 4 2 3 3 3 3 3 2 2 2 2 2 (9. and using the given initial condition y 0 = 1 .12) By substitution of (9.99 y 1 y''2 = ( x 2 – 1 ) y 2 = – 0.4y 1 y''0 = ( x 0 – 1 ) y 0 = – y 0 y''1 = ( x 1 – 1 ) y 1 = – 0.11) (9.592y 2 y''' 3 = ( – x 3 + 3x 3 ) y 3 = 0.4681y 3 4) y( 4 = ( x 4 – 6x 4 + 3 ) y 4 = 2.3y 1 y' 4 = – x 4 y 4 = – 0.9) through (9.12) into (9. Third Edition Copyright © Orchard Publications 9−3 .1.2y 1 y' 3 = – x 3 y 3 = – 0. are now substituted into (9.10) (9.Taylor Series Method The values in E6:E10. we obtain: Numerical Analysis Using MATLAB® and Excel®.1y 1 y' 2 = – x 2 y 2 = – 0.299y 1 y''' 2 = ( – x 2 + 3x 2 ) y 2 = 0.96 y 2 y''3 = ( x 3 – 1 ) y 3 = – 0.9) (9.9401y 1 4) y( 2 = ( x 2 – 6x 2 + 3 ) y 2 = 2.136y 4 4) y( 0 = ( x 0 – 6x 0 + 3 ) y 0 = 3y 0 4) y( 1 = ( x 1 – 6x 1 + 3 ) y 1 = 2. G6:G10. F6:F10.91 y 3 y'' 4 = ( x 4 – 1 ) y 4 = – 0. and we obtain the following relations: y' 0 = – x 0 y 0 = – 0y 0 = 0 y' 1 = – x 1 y 1 = – 0.

1y'0 + 0.00019 + 0. y 2 = y 1 + 0.000167y''' 0 + 0.97531 ( 0.000167y''' 2 + 0.18) 9−4 Numerical Analysis Using MATLAB® and Excel®.005 + 0.13) Similarly.00167 ( 0 ) + 0.88249 ----.005y''1 + 0.00005 + 0.0048 + 0.03 – 0.= – xy of this example can be solved analytically as follows: The differential equation dy dx dy ----.0001 + 0.1y'2 + 0.01 – 0.99501 ) = 0.02 – 0.000012 = 0.000004 ( 3 ) = 1 – 0.980194 ) = 0.005y''3 + 0.923107 ) = 0. 1 -(0) + C ln 1 = – -2 C = 0 –x ⁄ 2 1 2 -x ln y = – -2 or y = e (9.00001 ) y 1 = 0.0042 + 0.00001 ) y 3 = 0.16) (9.98511 ( 0. Third Edition Copyright © Orchard Publications .15) (4) (9.= – xdx y ∫ ----y dy = – x dx (4) (4) (9.955993 ) = 0.005y''4 + 0.1 ( 0 ) + 0.1y'3 + 0.00015 + 0.1y'1 + 0.95600 ( 0.1y'4 + 0.005 ( – 1 ) + 0.000004y 3 = ( 1 – 0.00455 + 0.923107 y 5 = y 4 + 0.04 – 0.000167y''' 4 + 0.99501 (9.000004y 2 = ( 1 – 0.9656 ( 0.00001 ) y 2 = 0.17) ∫ 1 2 -x + C ln y = – -2 and with the initial condition y = 1 when x = 0 .005y''0 + 0.955993 y 4 = y 3 + 0.000004y 1 = ( 1 – 0.Chapter 9 Solution of Differential Equations by Numerical Methods y 1 = y 0 + 0.000004y 0 (4) = 1 + 0.14) (4) (9.980194 y 3 = y 2 + 0.00001 ) y 4 = 0.000167y''' 3 + 0.005y''2 + 0.00495 + 0.000167y''' 1 + 0.000004y 4 = ( 1 – 0.

y' ) (9.( k + k2 ) y n + 1 = y n + -2 1 For Runge-Kutta Method of Order 2 (9. y n + h ) 1 . y n ) k 2 = hf ( x n + h .8825 and we observe that this value is in close agreement with the value of (9. y.18) yields y = e – 0.17).y''' h 2 + ----y h + … y' i + 1 = y' i + y''i h + ---2! i 3! i (9.125 = 0. Third Edition Copyright © Orchard Publications 9−5 . y ) at several points instead of the values of successive derivatives at a single point. For a Runge−Kutta method of order 2. The applicable formulas are as follows: Numerical Analysis Using MATLAB® and Excel®.20) 9.'x') z = exp(-1/2*x^2) The procedure used in this example. We can verify the analytical solution of Example 9.19) In this case.'y(0)=1'.1 with MATLAB’s dsolve(s) function using the following script: syms x y z z=dsolve('Dy=−x*y'. It differs from the Taylor series method in that we use values of the first derivative of f ( x. we can use order 3 or order 4. the following formulas are applicable.Runge−Kutta Method For x 5 = 0.2 Runge−Kutta Method The Runge−Kutta method is the most widely used method of solving differential equations with numerical methods. k 1 = hf ( x n . we need to apply the additional formula 1 1 (4) 3 . can be extended to apply to a second order differential equation y'' = f ( x.5 (9.21) When higher accuracy is desired.

and 4 Runge−Kutta methods with h = 0. we begin with x = 0 .2 Compute the approximate value of y at x = 0. Solution: a.( 0.2 from the solution y ( x ) of the differential equation y' = x + y 2 (9. k 1 = hf ( x n .24) given the initial condition y ( 0 ) = 1 .2 ( 0 + 1 ) = 0.264 .2 .2 [ 0 + 0. y n ) = k 1 l 1⎞ h . we use (9.25) 9−6 Numerical Analysis Using MATLAB® and Excel®.. Since we are given that y ( 0 ) = 1 . Use order 2. 3. and y = 1 . Third Edition Copyright © Orchard Publications .( m + 2m 2 + 2m 3 + m 4 ) y n + 1 = y n + -6 1 For Runge-Kutta Method of Order 4 (9.2 ) ] = 0.2 k 2 = hf ( x n + h . y + --l 2 = hf ⎛ x n + -⎝ 2 n 2⎠ l 3 = hf ( x n + h . y n ) = 0. y n + -----m 2 = hf ⎛ x n + h . y n + h ) = 0.2 + 0. y n + -----m 3 = hf ⎛ x n + -⎝ 2 2⎠ m 4 = hf ( x n + h .2 + ( 1 + 0.( l + 4l 2 + l 3 ) y n + 1 = y n + -6 1 For Runge-Kutta Method of Order 3 m 1 = hf ( x n . y n ) = l 1 = k 1 m 1⎞ -.= l2 ⎝ 2 2⎠ m 2⎞ h . y n + m 3 ) 1 .( k + k 2 ) = 1 + -y 1 = y 0 + -2 2 1 (9. Then.22) (9. y n + 2l 2 – l 1 ) 1 .23) Example 9.328 ) = 1.328 2 2 and 1 1 .. For order 2.Chapter 9 Solution of Differential Equations by Numerical Methods l 1 = hf ( x n .21)..

we use (9..( m + 2m 2 + 2m 3 + m 4 ) y 1 = y 0 + -6 1 1 . For order 4. Then. y n + --.262 + 2 × 0.276 + 0. but these will not be discussed in this text. Then.( 0. m 1 = hf ( x n .276 ) ] = 0..274 = 1 + -6 (9.( 0.2 [ 0 + 0.22).262 ⎝ 2 2⎠ m 2⎞ h 0. y n ) = k 1 = 0. the pair of 3rd−order formulas* are: * Third and fourth order formulas can also be used.2 m 1⎞ h .2 0 + -----m 3 = hf ⎛ x n + -⎝ 2 2⎠ 2 ⎝ 2 ⎠ 2 (9. y n ) = l 1 = k 1 = 0.= l 2 = 0.2 + 4 × 0. Third Edition Copyright © Orchard Publications 9−7 .2 + ( 1 + 0.2 ) ] = 0.2 ⎛ 0 + -l 2 = hf ⎛ x n + -⎠ ⎝ ⎠ ⎝ ⎝ 2 2 2 2⎠ (9.262 . l 1 = hf ( x n .391 ) = 1.30) For second order differential equations.23).391 2 and 1 1 .262⎞ 2 . y.262 – 0.366 ) = 1.26) l 3 = hf ( x n + h .28) = 0. They can be found in differential equations and advanced mathematics texts.= 0. For order 3.2 2 l1 ⎞ 1 1 h .⋅ 0.2 ⎛ 0.2 + 2 × 0.Runge−Kutta Method b. y n + -----m 2 = hf ⎛ x n + -.29) The Runge−Kutta method can also be used for second order differential equations of the form y'' = f ( x. we use (9.⋅ 0. y' ) (9.273 .2 [ ( 0 + 0.+ 1 + -----------.276 m 4 = hf ( x n + h .366 and 1 .27) c. y n + m 3 ) = 0.( l 1 + 4l 2 + l 3 ) = 1 + -y 1 = y 0 + -6 6 (9. y n + 2l 2 – l 1 ) = 0.2 ) + ( 1 + 2 × 0.2⎞ = 0.= 0. y n + -----. Numerical Analysis Using MATLAB® and Excel®.262 + 0.2⎞ + ⎛ 1 + --.

2 . y n + 2l 2 – l 1. Use h = 0.2 . We will use h = x 1 – x 0 = 0. compute the approximate values of y and y' at x = 0.2 .32) with the initial conditions y ( 0 ) = 1 . y n + ---. Third Edition Copyright © Orchard Publications . We rewrite the given equation as y'' = 2y = 0 ⋅ x + 2y 3 3 (9. y' n + 2l' 2 – l' 1 ) 1 . y' ( 0 ) = – 1 .2 .31) we obtain: 9−8 Numerical Analysis Using MATLAB® and Excel®. Solution: We are given the values of x 0 = 0 .( l' + 4l' 2 + l' 3 ) y' n + 1 = y' n + -6 1 For Runge-Kutta Method of Order 3 2nd Order Differential Equation (9.Chapter 9 Solution of Differential Equations by Numerical Methods l 1 = hy' n l' 1 = hf ( x n . y n . y' n + ---l' 2 = hf ⎛ x n + -⎝ 2 2 2⎠ l 3 = h ( y' n + 2l' 2 – l' 1 ) l' 3 = hf ( x n + h.3 Given the 2nd order non−linear differential equation y'' – 2y = 0 3 (9. y' 0 = – 1 and we are seeking the values of y 1 and y' 1 at x 1 = 0.( l + 4l 2 + l 3 ) y n + 1 = y n + -6 1 1 .33) and using (9. y 0 = 0 . y' n ) l' 1 ⎞ l 2 = h ⎛ y' n + ---⎝ 2⎠ l1 l' 1⎞ h -.31) Example 9.

( – 0.35) MATLAB has two functions for computing numerical solutions of Ordinary Differential Equations (ODE).2916 ) + 0.34) By substitution into the last two formulas of (9. we will use the ode23 function in our subsequent discussion. y 0 + ---. Example 9. we obtain: 1 1 .2 ( – 1 ) = – 0.458 ) = 0. uses fourth and fifth−order Runge−Kutta methods. Third Edition Copyright © Orchard Publications 9−9 . The syntax for ode23 is ode23(‘f’.88 ) ] = 0.31).= 0.2 ( 0 + 2 × 1 + 0 ) = 0. uses second and third−order Runge−Kutta methods. represents the initial condition or boundary point that is needed to determine a unique solution.2916 – 0.2726 3 3 3 3 3 (9.1634 l' 3 = hf ( x 0 + h.2 + 4 ( 0. y0. tspan.36) Numerical Analysis Using MATLAB® and Excel®.= 0. a set of x values and the corresponding set of y values that represent points of the function y = f ( x ) . The second.6935 .2 [ 2 ( 0.2 ) ] + 0 } = 0. defines the desired time span of the interval over which we want to evaluate the function y = f ( x ) . The first.2 ( 1.4 + 4 ( 0.2 ) ] = 0.16 l 2 = h ⎛ y' 0 + ---.( l' + 4l' 2 + l' 3 ) = – 1 + -y' 1 = y' 0 + -6 6 1 (9.2 [ 2 ( 1 – 0.Runge−Kutta Method l 1 = hy' 0 = 0. in single quotation marks.+0 ⎝ ⎠ ⎝ 2 2 2 2 ⎠ = 0.y0).8328 y 1 = y 0 + -6 6 1 1 .4 ) = – 0.2 ( – 1 + 2 × 0. The third argument.4 ⎝ ⎝ 2⎠ 2⎠ l1 l' 1 ⎞ h – 0.2 [ 2 ( 1 – 0. f. y' 0 + ---l' 2 = hf ⎛ x 0 + -.16 ) – 0. y 0.4 l' 1 ⎞ ------⎞ = – 0.4 Use the MATLAB ode23 function to find the analytical solution of the second order nonlinear equation y'' – 2y = 0 3 (9. y' 0 ) = 0.2916 l 3 = h ( y' 0 + 2l' 2 – l' 1 ) = 0.2⎞ 3 -. is the name of the user defined MATLAB function. The first argument. ode23. This function produces two outputs.32 + 0.tspan.2 l' 1 = hf ( x 0. ode45.( l 1 + 4l 2 + l 3 ) = 1 + -.( 0.2726 ) = – 0.2 { 0 + 2 [ 1 + 2 ( – 0.1 ) ] = 0. therefore. y 0 + 2l 2 – l 1. The second. y' 0 + 2l' 2 – l' 1 ) = 0.2 ⎛ – 1 + 0.1634 ) = 0.2 0 + 2 ⎛ 1 + --------.16 ) – ( – 0. Both have the same syntax.

Third Edition Copyright © Orchard Publications .y]=ode23('fex9_4'. Next.4').3. The script is shown below..'x') MATLAB displays the following message: Warning: Explicit solution could not be found. function d2y=fex9_4(x. yprime (lower curve)').Dy(0)=−1'.2*y(1)^3]. Solution: If we attempt to find the analytical solution with the following MATLAB script syms x y y=dsolve('D2y=2*y^3.2). tspan=[0 1]. ylabel('y (upper curve).Chapter 9 Solution of Differential Equations by Numerical Methods with the initial conditions y ( 0 ) = 1 and y' ( 0 ) = – 1 . y(:. at points y ( 0. xlabel('x'). '+r−'..2 ) and y' ( 0.2 ) .2.3. % Given initial conditions [x. These can be found in differential equations textbooks. % Output must be a column This file is saved as fex9_4. % Use 2nd and 3rd Order Runge−Kutta % Plot numeric values with the statements below plot(x. The numerical solution for this non−linear differential equation is obtained and plotted with the following script. d2y=[y(2). tspan. 9−10 Numerical Analysis Using MATLAB® and Excel®. We observe that the values at points y ( 0. This is because. in general.1). we write and execute the script below to obtain the plots for y and y ′ . Compare values with those of Example 9. by first writing a user defined m−file which we denote as fex9_4. Then.y(0)=1.y). plot the numerical solution using the function ode23 for the tspan interval 0 ≤ x ≤ 1 . % Interval over which we want to evaluate y=f(x) y0=[1. 'Ob−−') title('Numerical Solution for Differential Equation of Example 9..2 ) and y' ( 0. y(:. grid The plots for y and y' are shown in Figure 9. although few methods are available for special cases. compare favorably with those that we found in Example 9. non−linear differential equations cannot be solved analytically. x. y0).2 ) . This warning indicates that MATLAB could not find a closed−form solution for this non−linear differential equation.−1].

Then. the analytical solution of (9.Runge−Kutta Method Figure 9.4 Example 9.37) is .37) with the given initial conditions is found with MATLAB as follows: syms x y y=dsolve('x^2*D2y−x*Dy−3*y=x^2*log(x).2. Solution: The analytical solution of (9. compute and plot the numerical solution using the command ode23 along with points of the analytical solution. to verify the accuracy of the numerical solution. Plot for Example 9.37) with the initial conditions y ( 1 ) = – 1 and y' ( 1 ) = 0 .5 Use MATLAB to find the analytical solution of x y'' – x y' – 3y = x ln x 2 2 (9.7/9 1/x Numerical Analysis Using MATLAB® and Excel®. Dy(1)=0. y(1)=−1'. Third Edition Copyright © Orchard Publications 9−11 .1/3 log(x) .2/9) x and therefore. 'x') y = 1/9*(−3*x^3*log(x)−2*x^3−7)/x y=simple(y) y = (−1/3*log(x)−2/9)*x^2−7/9/x pretty(y) 2 (.

y(:. x.y). y'=1st der. Plot for Example 9. tspan=[1 4]. we create and save a user defined m−file.38) % Plot numeric and analytic values with the statements below plot(x.38) anal_yprime=((−2.5 9−12 Numerical Analysis Using MATLAB® and Excel®./3).^2−7.. ylabel('y (line with +).ln x – 2 -. % Interval over which we want to evaluate y=f(x) y0=[−1.5'). title('Numeric and Analytic Solutions of Differential Equation of Example 9. '−'. Figure 9. '+'. function d2y=fex9_5(x.. anal_y../9).*x./(9. y(2)/x+3*y(1)/x^2+log(x)]..3. logx=lnx % % we let y(1) = y and y(2)=y'. y(:. % This is the first derivative of (9. % This is the. then y(1)'=y(2) % % and y(2)'=y(2)/x^2+3*y(1)/x^2+log(x) % d2y=[y(2). x. x. % Use 2nd and 3rd Order Runge−Kutta anal_y=((−1. fex9_5.⎞ x 2 – ----y = ⎛ – -⎝ 3 9⎠ 9x (9. xlabel('x').2)...1). % analytic solution of the 2nd order differential equation of (9. % output must be a column The following MATLAB script computes and plots the numerical solution values for the interval 1 ≤ x ≤ 4 and compares these with the actual values obtained from the analytical solutions.*log(x)−2.38) Next. grid The numeric and analytical solutions are shown in Figure 9./(9.. Third Edition Copyright © Orchard Publications .*x). yprime (line with O)'). '−'). % Produces the derivatives of Example 9.^2). 'O'. y0)../3).5 % x^2*y''−x*y'−3*y=x^2*log(x) where y''=2nd der.Chapter 9 Solution of Differential Equations by Numerical Methods 1 7 .3.0]. anal_yprime. % Given initial conditions [x.y]=ode23('fex9_5'. tspan.*log(x)−7.*x+7./9).*x.

Obviously. 0. Solution: The spreadsheet of Figure 9.1.2.Δ f n + ----. find the value of y corresponding to x = 0.3 Adams’ Method In this method. Numerical Analysis Using MATLAB® and Excel®.6 correct to three decimal places using Adams’ method. y 3. and y5 using the third−order Runge−Kutta method as in Example 9. Third Edition Copyright © Orchard Publications 9−13 .39) where h = xn + 1 – xn f n = ( x n.Adams’ Method 9.40) with the initial condition y ( 0 ) = 1 . to form a table of differences. 0.4 and 0. Then.2. y n ) Δ fn = fn – fn – 1 Δ fn = Δ fn – Δ fn – 1 2 and so on. Example 9. it is necessary to have several (4 or more) approximate values of y ( x ) in addition to the given initial condition y ( 0 ) . y 4 . Adams’ method uses the formula 1 5 2 3 3 . 0.5 by the third − order Runge−Kutta method. the step from y n to y n + 1 is performed by a formula expressed in terms of differences of f ( x. y ) .6 Given the differential equation y' = 2y + x (9.4 shows the results of the computations of y 1.Δ fn + … y n + 1 = y n + h f n + -2 12 8 (9.Δ f n + -. These values can be found by other methods such as the Taylor series or Runge−Kutta methods. compute the approximate values of y for x = 0. y 2.3.

5619 h*f(xn+h.yn+0.6398 2.1000 0.5*h.0000 1.5*L(1))]= 0.1000 0.yn) h*(0.5148 2.8773)= 0.5*L(1)) h*[(0+0.2250 h*f(xn+h.1000 1.4000 2.yn+2*L(2)-L(1)) h*[(0+h)+2*(1+2*L(2)-L(1))]= 0.3)= Next value x(0) + 3*h y(3)= From previous computation L(1)= L(2)= L(3)= 39 40 41 42 43 44 45 46 47 48 h*f(xn.2859 h*f(xn+h.2+2*1.0 0.3000 1.2553 h*f(xn+0.yn+0.3313 y(4)= y(2) +(L(1) + 4*L(2) + L(3))/6 h= x(4)-x(3)= x(0.5*L(1))]= 0.8773)= 0.5*h)+2*(1+0.5146 y(2)= y(1) +(L(1) + 4*L(2) + L(3))/6 h= x(2)-x(1)= x(0.yn) h*(0.2000 1.yn) h*(0.4 0.2267 0.2267)= 0.5*h)+2*(1+0.4)= Next value x(0) + 4*h y(4)= From previous computation L(1)= L(2)= L(3)= 49 50 51 52 53 h*f(xn.5*L(1))]= 0.5*L(1)) h*[(0+0.2 0.2)= Next value x(0) + 2*h y(2)= From previous computation L(1)= L(2)= L(3)= h*f(xn.yn+0.1)= Next value x(0) + h y(1)= From previous computation L(1)= L(2)= L(3)= h*f(xn.3602 h*f(xn+h.yn) h*(0.1000 0.yn+2*L(2)-L(1)) h*[(0+h)+2*(1+2*L(2)-L(1))]= 0.8773 0.yn+2*L(2)-L(1)) h*[(0+h)+2*(1+2*L(2)-L(1))]= 0.5063 h*f(xn+0.4510 h*f(xn+h.5*L(1)) h*[(0+0.3313 0. Spreadsheet for Example 9.5*h.1 0.3229 h*f(xn+0.5146 0.yn+0.5146)= 0.5*L(1))]= 0.yn+2*L(2)-L(1)) h*[(0+h)+2*(1+2*L(2)-L(1))]= 0.5*h.3+2*1.5*h)+2*(1+0.yn+2*L(2)-L(1)) h*[(0+h)+2*(1+2*L(2)-L(1))]= 0.yn+0.2000 x5= h*f(xn+0.3286 1.5*h)+2*(1+0.8969 y(5)= y(2) +(L(1) + 4*L(2) + L(3))/6 Figure 9.yn) h*(0+2*1)= 0.5 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 L(1)= L(2)= L(3)= y(1)= y(0) +(L(1) + 4*L(2) + L(3))/6 h= x(2)-x(1)= x(0.1+2*1.2267 0.4055 h*f(xn+0.1000 0.3+2*1.4.5*h. Third Edition Copyright © Orchard Publications .5*h.5*h)+2*(1+0.6 9−14 Numerical Analysis Using MATLAB® and Excel®.5*L(1)) h*[(0+0.0000 1.5*L(1))]= 0.8773 y(3)= y(2) +(L(1) + 4*L(2) + L(3))/6 h= x(3)-x(2)= x(0.5*L(1)) h*[(0+0.Chapter 9 Solution of Differential Equations by Numerical Methods A 1 2 3 B C D E F Differential Equation y' = x + 2y Numerical solution by Runga-Kutte method follows h= x(0)= y(0)= x0= x1= x2= x3= x4= h*f(xn.1000 0.3 0.2600 x(1)-x(0)= (given) Initial condition (given) 0.4124 1.

do not differ considerably. y.2312 6.2638 + -8 12 2 (9.8969 fn=xn+2yn 2.2938 0.0626 1.42) and 1 .8773 2.Milne’s Method Next. we must reduce the interval h .41) As with the other methods.4 Milne’s Method Milne’s method also requires prior knowledge of several values of y .43) serves as a check for the value y n + 1 = f ( x n + 1.0272 Δ fn Δ 2 fn Δ 3 fn and by substitution into (7.2267 1. 9.2232 0.0406 0.( 0.3313 2.8969 + 0. xn 0.6758 3.0330 0.0406 ) = 3. Third Edition Copyright © Orchard Publications 9−15 .1 0. y n + 1 ) (9.1826 0.1496 0.1 6. we accept Y n + 1 as the best approximation.43) respectively.0080 5.5534 2.0000 0. Numerical Analysis Using MATLAB® and Excel®. we compute the following values to be used in Adams’ formula of (9.4 0. y' ) with initial conditions y ( x 0 ) = y 0 and y'( x 0 ) = y' 0.8254 4.h [ 2f n – f n – 1 + 2f n – 2 ] y n + 1 = y n – 3 + -3 (9.2292 0.5 yn 1.39).43) The corrector formula of (9. These are shown below. It uses the predictor−corrector pair 4 .2232 ) + -y 6 = 2.0000 1.599 .2 0. Adams’ method can also be applied to second order differential equations of the form y'' = f ( x.5146 1.1224 0.h[f Y n + 1 = y n – 1 + -+ 4f n + f n – 1 ] 3 n+1 (9.5534 0.0 0.39) 3 51 .0546 1.3 0.44) If y n + 1 and Y n + 1 in (9.2312 ) + ----( 0.( 1.42) and (9. If they differ significantly.

7 n 2 3 4 5 xn 0.46) and (9.2292 4.599 .3313 + -3 (9.Chapter 9 Solution of Differential Equations by Numerical Methods Example 9. Solution: This is the same differential equation as in Example 9.0626 ) = 3.2938 + 5. y.1 ( 7. y' ) with initial conditions y ( x 0 ) = y 0 and y'( x 0 ) = y'0 .7 Use Milne’s method to find the value of y corresponding to x = 0.× 0.6 + 2 × 3.1 ( f 6 + 4f 5 + f 4 ) Y 6 = y 4 + -3 1 . and from Example 9.6 where we found the following values: TABLE 9.2938 and using the predictor formula we find 4 .0.6 for the differential equation y' = 2y + x (9.0546 ) = 3. 9−16 Numerical Analysis Using MATLAB® and Excel®.× 0.7984 and 1 .599 = 2.3 0.2938 – 5.47) that the predictor−corrector pair is in very close agreement.6 . we must find the value of f 6 . Milne’s method can also be extended to second order differential equations of the form y'' = f ( x. f 6 = x 6 + 2y 6 = 0.1 Table for Example 9.8773 2.5146 1.0626 + 2 × 4.0626 6.599 = 1.4 0.5 y 6 = 3.1 ( 2 × 6. Third Edition Copyright © Orchard Publications .7984 + 4 × 6.5146 + -3 (9.47) We see from (9.3313 2.5 yn 1. this is found from f 6 = x 6 + 2y 6 where x 6 = 0.2 0.46) Before we use the corrector formula of (9.8969 fn=xn+2yn 3.45) with the initial condition y ( 0 ) = 1 .0546 5.43).599 = 7.1 ) [ 2f 5 – f 4 + 2f 3 ] y 6 = y 2 + -3 4 .( 0. Then.

y' ) using the relation 1 1 (4) 3 . • The Taylor series method can also be extended to apply to a second order differential equation y'' = f ( x.1 .y'''i h 2 + ----y h + … y' i + 1 = y' i + y''i h + ---2! 3! i • The Runge−Kutta method uses values of the first derivative of f ( x. it is the most widely used method of solving differential equations using numerical methods. We can use this series method to obtain approximate solutions of differential equations with the relation 111..( k + k2 ) y n + 1 = y n + -2 1 provided that h is sufficiently small such as h = 0. y n + --l 2 = hf ⎛ x n + -⎝ 2 2⎠ 1 . y n + -----m 3 = hf ⎛ x n + -⎝ 2 2⎠ m 1⎞ h .1 . y. y n + h ) 1 .( l + 4l 2 + l 3 ) y n + 1 = y n + -6 1 l 3 = hf ( x n + h . y n ) k 2 = hf ( x n + h .. y n + 2l 2 – l 1 ) • For a Runge−Kutta method of order 4 we use the relations m 1 = hf ( x n . • For a Runge−Kutta method of order 3 we use the relations l 1 = hf ( x n . y ) at several points. y n ) = l 1 = k 1 m 2⎞ h . • For a Runge−Kutta method of order 2 we use the relations k 1 = hf ( x n .Summary 9.5 Summary • The Taylor series method uses values of successive derivatives at a single point.= l2 m 2 = hf ⎛ x n + -⎝ 2 2⎠ m 4 = hf ( x n + h . y n + -----.( m + 2m 2 + 2m 3 + m 4 ) y n + 1 = y n + -6 1 • The Runge−Kutta method can also be used for second order differential equations of the form Numerical Analysis Using MATLAB® and Excel®.. y n + m 3 ) 1 . y n ) = k 1 l 1⎞ h . Third Edition Copyright © Orchard Publications 9−17 .( 4 ) 4 2 3 y'' h y''' h y h y i + 1 = y i + y' i h i + 1 + ---+ ---+ ---+… 2! i i + 1 3! i i + 1 4! i i + 1 provided that h is sufficiently small such as h = 0.

y n + 2l 2 – l 1. y. ode45. defines the desired time span of the interval over which we want to evaluate the function y = f ( x ) .Δ fn + … y n + 1 = y n + h f n + -2 n 12 8 where h = xn + 1 – xn f n = ( x n. The first argument. y0. 9−18 Numerical Analysis Using MATLAB® and Excel®. y' n + 2l' 2 – l' 1 ) 1 1 . y' n + ---l' 2 = hf ⎛ x n + -⎝ 2 2 2⎠ l' 3 = hf ( x n + h. represents the initial condition or boundary point that is needed to determine a unique solution.tspan. Both have the same syntax.( l 1 + 4l 2 + l 3 ) . it is necessary to have several (4 or more) approximate values of y ( x ) in addition to the given initial condition y ( 0 ) . The second. • The syntax for ode23 is ode23(‘f’. uses fourth and fifth−order Runge−Kutta methods. f. This function produces two outputs. tspan. y' ) • For second order differential equations. Third Edition Copyright © Orchard Publications . uses second and third−order Runge−Kutta methods. They can be found in differential equations texts. This method uses the formula 1 5 2 3 3 . The third argument. a set of x values and the corresponding set of y values that represent points of the function y = f ( x ) .Δ f + ----. y n ) Δ fn = fn – fn – 1 Δ fn = Δ fn – Δ fn – 1 2 and so on. The second. The first. y n . • Adams’ method provides the transition from y n to y n + 1 and the step is performed by a formula expressed in terms of differences of f ( x. y n + ---. y' n ) l1 l' 1⎞ h -.Δ f n + -.y0). in single quotation marks. • MATLAB has two functions for computing numerical solutions of Ordinary Differential Equations (ODE).Chapter 9 Solution of Differential Equations by Numerical Methods y'' = f ( x. These values can be found by other methods such as the Taylor series or Runge−Kutta methods. ode23. y ) . the pair of 3rd−order relations are: l 1 = hy' n l' 1 ⎞ l 2 = h ⎛ y' n + ---⎝ 2⎠ l 3 = h ( y' n + 2l' 2 – l' 1 ) yn + 1 l' 1 = hf ( x n .( l' + 4l' 2 + l' 3 ) = y n + -y' n + 1 = y' n + -6 6 1 Third and fourth order formulas can also be used but they were not be discussed in this text. is the name of the user defined MATLAB function. To use this method.

The procedure for this method was not discussed.Summary • Milne’s method also requires prior knowledge of several values of y . The corrector formula serves as a check for the value y n + 1 = f ( x n + 1.h[f Y n + 1 = y n – 1 + -+ 4f n + f n – 1 ] 3 n+1 where y n + 1 is the predictor formula and Y n + 1 is the corrector formula. Third Edition Copyright © Orchard Publications 9−19 . If they differ significantly. Numerical Analysis Using MATLAB® and Excel®.h [ 2f n – f n – 1 + 2f n – 2 ] y n + 1 = y n – 3 + -3 and 1 . It can be found in differential equations texts. we must reduce the interval h . y' ) with initial conditions y ( x 0 ) = y 0 and y' ( x 0 ) = y'0 . y. It uses the predictor−corrector pair 4 . y n + 1 ) If y n + 1 and Y n + 1 do not differ considerably. we accept Y n + 1 as the best approximation. Milne’s method can also be extended to second order differential equations of the form y'' = f ( x.

2 sin x 3 with the initial condition y ( 0 ) = 0. It is suggested that a spreadsheet is used to do all computations. 2. 5. Then.1 and x 0 + 0. compute and plot the numerical solution using the MATLAB function ode23 along with points of the analytical solution to verify the accuracy of the numerical solution for the interval 2 ≤ x ≤ 4 . and x 0 = 0. Use MATLAB to plot the numerical solution of the non−linear differential equation y' = – y + 0. Use MATLAB to find the analytical solution of y' = f ( x ) = 3x 2 with the initial condition y ( 2 ) = 0.6 Exercises 1. Use the MATLAB ode23 function to verify the analytical solution of Example 9. y' ( 0 ) = – 1 . It is suggested that a spreadsheet is used to do all computations. 4.1. 6.5 . using the third−order Runge−Kutta method. Given the differential equation y'' + y' = xy compute the approximate values of y and y' at x 0 + 0. Given the differential equation y' = x – y 2 with the initial condition y ( 0 ) = 1 and x 0 = 0.2 correct to four decimal places using the third−order Runge− Kutta method. 9−20 Numerical Analysis Using MATLAB® and Excel®.Chapter 9 Solution of Differential Equations by Numerical Methods 9. 3.0 find the values of y corresponding to the values of x 0 + 0.0 correct to four decimal places. Third Edition Copyright © Orchard Publications . Construct a spreadsheet for the numerical solutions of Example 9.1 and x 0 + 0.2 given that y ( 0 ) = 1 .2.707 using the command ode23 for the interval 0 ≤ x ≤ 10 .

tspan=[0 3]. we write and execute the MATLAB script below. '+r−'. y0). xlabel('x').1).7 Solutions to End−of−Chapter Exercises 1.2). tspan. We write and save the following function file: function dy = func_exer9_1(x. % Use 2nd and 3rd Order Runge−Kutta % Plot numeric values with the statements below plot(x. % Interval over which we want to evaluate y=f(x) y0=[1. % Given initial conditions [x. grid The plot below shows the function y = f ( x ) and its derivative dy ⁄ dx .. y(:. ylabel('y (upper curve).1').−1]. yprime (lower curve)').Solutions to End−of−Chapter Exercises 9. y(:... x.y]=ode23('func_exer9_1'. Numerical Analysis Using MATLAB® and Excel®. 'Ob−−') title('Numeric Solution of Differential Equation of Exercise 9. Third Edition Copyright © Orchard Publications 9−21 . Next.y) dy = −x*y.

2758 0. % Use 2nd and 3rd Order Runge−Kutta % Plot numeric values with the statements below plot(x.. ylabel('y'). grid 9−22 Numerical Analysis Using MATLAB® and Excel®..2731 0.yn) 12 L(1)= h*f(xn+0.5*h)+(1+0.5*h.5.2735 h*(0+1^2)= h*[(0+0.5*(D11+D12) h*(0+1^2)= h*[(0+0.y]=ode23('fexer9_3'.0000 h*(0+1^2)= h*(0+0.Chapter 9 Solution of Differential Equations by Numerical Methods 2.2620 0.yn+2*L(2)-L(1)) 14 L(3)= y(0) +(L(1) + 4*L(2) + L(3))/6 15 y(1)= 16 h*f(xn. tspan. it follows that y(0) = 7.5 [x.2+(1+0.y(2)=0.5*m(2)) 19 m(3)= h*f(xn+h.yn+0.yn+0.3906 1. xlabel('x').5'. A B 1 Differential Equation y' = x + y2 2 Numerical solution by Runga-Kutte method follows 3 h= (given) 4 x(0)= (given) 5 y(0)= Initial condition (given) 6 7 k(1)= h*f(xn.5*h)+(1+0.2)^2)= y(0)+0. Dy=3*x^2.yn+0.3').5*h.0000 1.2000 0.2000 0.5*h.. The MATLAB script for the numerical solution is as follows: tspan=[2 4].yn+h) 9 y(0)+0. Third Edition Copyright © Orchard Publications . % Interval over which we want to evaluate y=f(x) y0=7. y.yn) 17 m(1)= h*f(xn+0.3655 1.5*L(1)) 13 L(2)= h*f(xn+h. we write and save the following statements as function file fexer9_3 function Dy=fexer9_3(x.5(k(1)+k(2)) 10 y(1)= 11 h*f(xn.5.2620 0.5*L(1)^2)]= h*[(0+h)+(1+2*L(2)-L(1))^2]= 0.2000 0. '+r−') title('Numeric Solution of Differential Equation of Exercise 9.yn+m(3)) 20 m(4)= 21 y(1)= y(0) +(1/6)*(m(1) + 2*m(2)+2*m(3) + m(4)) C D 0. y0).2640 0.3280 1. The analytical solution is found with syms x y y=dsolve('Dy=3*x^2.2000 0.5*h)+(1+0.'x') and MATLAB displays y = x^3-15/2 Next.5*m(2))^2]= h*[(0+h)+(1+m(3))^2]= 3.5*m(1))^2]= h*[(0+0.yn) 8 k(2)= h*f(xn+h.5*m(1)) 18 m(2)= h*f(xn+0.y). % Initial condition: Since y=x^3−15/2 and y(2) = 0.

2sinx').. '−')... plot(x.num_x.707]. Dy=−y^3+0.Solutions to End−of−Chapter Exercises 4. ylabel('y=f(x)'). x. grid 5.x0). x0=[0.. We write and save the following statements as function file fexer4 function Dy=fexer4(x. The MATLAB script and the plot for the numerical solution are as follows: tspan=[0 10]. Third Edition Copyright © Orchard Publications 9−23 .tspan. The spreadsheet is shown on the following two pages. [x.y).num_x]=ode23('fexer4'.num_x.'+'. xlabel('x'). Numerical Analysis Using MATLAB® and Excel®.2*sin(x). title('Numeric solution of non−linear differential equation dy/dx=−x^3+0...

0 0.yn) 19 L(2)= h*f(xn+0.yn) 9 h*[(0+0.yn+0.1+h)^2-(0. Third Edition Copyright © Orchard Publications .2077 0.0305 37 38 h*(0.yn+0.2+2*1.yn) 39 h*[(0+0.5*L(1))]= L(2)= h*f(xn+0.0895 -0.yn+2*L(2)-L(1)) 31 32 y(3)= y(2) +(L(1) + 4*L(2) + L(3))/6 33 34 h= x(3)-x(2)= -0.0948 -0.5*h.8213 27 28 h*(0.yn+2*L(2)-L(1)) 41 42 y(4)= y(2) +(L(1) + 4*L(2) + L(3))/6 43 D E F x0= x1= x2= 0.5*h)^2-(1+0.yn+0.Chapter 9 Solution of Differential Equations by Numerical Methods A B C 1 Differential Equation is y' = x2 .2000 26 y(2)= From previous computation 0.2000 35 x(0.5*L(1)) 30 h*[(0+h)+2*(1+2*L(2)-L(1))]= L(3)= h*f(xn+h.yn+2*L(2)-L(1)) h*[(0.5*L(1))]= L(2)= h*f(xn+0.1)= Next value x(0) + h 0.9052+0.1000 25 x(0.8063 continued on next page 9−24 Numerical Analysis Using MATLAB® and Excel®.yn+0.3+2*1.5*L(1)) 10 h*[(0+h)^2-(1+2*L(2)-L(1))]= L(3)= h*f(xn+h.1981 0.0901 0.0000 7 8 h*(0-1)= L(1)= h*f(xn.2922 -0.yn+2*L(2)-L(1)) 11 12 y(1)= y(0) +(L(1) + 4*L(2) + L(3))/6 13 14 h= x(2)-x(1)= 0.2 -0.5*h)+2*(1+0.5*L(1)) 40 h*[(0+h)+2*(1+2*L(2)-L(1))]= L(3)= h*f(xn+h.8773)= L(1)= h*f(xn.1^2-0.5*h)^2-(0.2)= Next value x(0) + 2*h 0.1843 0.8213 0.5*h.5*L(1))]= 20 L(3)= h*f(xn+h.0838 -0.9052)= L(1)= h*f(xn.5*h)+2*(1+0.2138 -0.6000 36 y(3)= From previous computation 1.5*h.9052 17 18 h*(0.9052+2*L(2)-L(1))]= 21 22 y(2)= y(1) +(L(1) + 4*L(2) + L(3))/6 23 24 h= x(2)-x(1)= 0.y 2 Numerical solution by Runga-Kutte method follows 3 4 h= x(1)-x(0)= 0.2405 1.1 0.1000 16 y(1)= From previous computation 0.0787 0.5*h.1000 15 x(0.1000 5 x(0)= (given) 0.5146)= L(1)= h*f(xn.5*L(1))]= L(2)= h*f(xn+0.yn) 29 h*[(0+0.1000 -0.5*L(1)) h*[(0+0.0000 6 y(0)= Initial condition (given) 1.0305 -0.9052 -0.3)= Next value x(0) + 3*h -0.

0000 -0.yn) h*f(xn+0.0998 -0.0000 0.1000 0.8063 6.4)= y(3)= L(1)= L(2)= L(3)= y(4)= B x(4)-x(3)= Next value x(0) + 4*h From previous computation h*f(xn.0992 0.5*h.0901 0. y'(0) + 2*L'(2) .5*L(1)) h*f(xn+h.9050 -0.1)= Next value x(0) + h 21 y(1)= From previous computation 22 y'(1)= From previous computation 23 C 0.5*L(1))]= h*[(0+h)+2*(1+2*L(2)-L(1))]= D E F 0.1000 0.L'(1))= 14 L'(3)= h*f(x(0) + h.9003 continued on next page Numerical Analysis Using MATLAB® and Excel®. A B 1 Differential Equation is y''+y'=xy or y''=xy-y' 2 Numerical solution by Runga-Kutte method follows 3 4 h= x(1)-x(0)= 5 x(0)= Initial condition (given) 6 y(0)= Initial condition (given) 7 y'(0)= Initial condition (given) 8 9 L(1)= h*y'(0)= 10 L'(1)= h*f(x(0).5*L(1))= 13 L(3)= h*(y'(0) + 2*L'(2) .0000 0.0000 0.1000 -0.1000 0. y(0) + 0.L(1).0000 0.3+2*1.0000 1. y(0).yn+2*L(2)-L(1)) y(2) +(L(1) + 4*L(2) + L(3))/6 C 0.L'(1))= 15 16 y(1)= y(0) +(L(1) + 4*L(2) + L(3))/6 17 y'(1)= y'(0) + (L'(1) + 4*L'(2) + L'(3))/6= 18 19 h= x(1)-x(0)= 20 x(0.8773)= h*[(0+0. Third Edition Copyright © Orchard Publications 9−25 .yn+0. y'(0) + 0.0000 0.0950 0.5*L'(1))= 12 L'(2)= h*f(x(0) + 0. y(0) + 2*L(2) .1000 0.0000 -1.9050 -0.5*L(1). y'(0))= 11 L(2)= h*(y'(0) + 0.5*h.9003 0.Solutions to End−of−Chapter Exercises A 44 45 46 47 48 49 50 51 52 53 h= x(0.8063 h*(0.5*h)+2*(1+0.

0934 0. y(1) + 0. y(0) + 2*L(2) .Chapter 9 Solution of Differential Equations by Numerical Methods 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 A L(1)= L'(1)= L(2)= L'(2)= L(3)= L'(3)= y(2)= y'(2)= h= x(0.L(1). y'(0))= h*(y'(0) + 0.5*h.5*L(1).0754 0. y'(1) + 2*L'(2) .8023 0.7074 9−26 Numerical Analysis Using MATLAB® and Excel®. y'(1) + 0.0851 0.8199 -0.7444 -0.0803 0.L'(1))= h*f(x(1) + h. y(1).2000 0.5*h.L(1).L'(1))= y(0) +(L(1) + 4*L(2) + L(3))/6 y'(0) + (L'(1) + 4*L'(2) + L'(3))/6= C -0. y(0).5*L'(1))= h*f(x(0) + 0.1000 0.5*L(1))= h*(y'(0) + 2*L'(2) .0949 -0. Third Edition Copyright © Orchard Publications .5*L(1).5*L'(1))= h*f(x(1) + 0.0968 0.L'(1))= h*f(x(0) + h.0966 -0. y(0) + 0.8023 -0. y'(1))= h*(y'(1) + 0.0900 0.5*L'(1))= h*(y'(1) + 2*L'(2) .2)= y(2)= y'(2)= L(1)= L'(1)= L(2)= L'(2)= L(3)= L'(3)= y(3)= y'(3)= B h*y'(1)= h*f(x(1).0991 -0. y(1) + 2*L(2) .0980 -0.0709 0. y'(0) + 0.8199 -0. y'0 + 2*L'(2) .L'(1))= y(0) +(L(1) + 4*L(2) + L(3))/6 y'(0) + (L'(1) + 4*L'(2) + L'(3))/6= x(2)-x(1)= Next value x(0) + 2*h From previous computation From previous computation h*y'(0)= h*f(x(0).0802 0.

1 The Trapezoidal Rule Consider the function y = f ( x ) for the interval a ≤ x ≤ b . and so on. the integral from a to b is the sum of the integrals from a to x 1 .. Therefore.1. 10. f (x) P0 y0 0 a P1 P2 Pn Pn – 1 y1 x1 y2 . from x 1 to x 2 . x n – 1 = a + ( n – 1 )Δ x . The total area is ∫a b f ( x ) dx = ∫a x1 f ( x ) dx + ∫x x2 1 f ( x ) dx + … + ∫x b n f ( x ) dx = n–1 k=1 ∑ ∫x xk k–1 f ( x ) dx The integral over the first subinterval. Integration by the trapezoidal rule To evaluate the definite integral ∫a f ( x ) dx .... …. x 2 = a + 2 Δ x... the number of points between x 0 = a and x n = b is n x 1 = a + Δ x. can now be approximated by the area of the trapezoid Numerical Analysis Using MATLAB® and Excel®.Chapter 10 Integration by Numerical Methods T his chapter is an introduction to numerical methods for integrating functions which are very difficult or impossible to integrate using analytical means. we divide the interval a ≤ x ≤ b b into n subintervals –a each of length Δ x = b ---------... and Simpson’s rule that computes a function f ( x ) with a set of quadratic functions. Then.. shown in Figure 10. We will discuss the trapezoidal rule that computes a function f ( x ) with a set of linear functions.1.. yn – 1 xn – 1 yn x b x2 Figure 10. Third Edition Copyright © Orchard Publications 10−1 . and finally from x n – 1 to b ..

= 1 ---------.2) Compare with the exact value.33333 . Solution: The exact value of this integral is ∫1 2 x x dx = ---3 2 3 2 1 8 1 7 = -.= 2 Δx = b 4 4 n Then.1 Using the trapezoidal rule with n = 4 .Chapter 10 Integration by Numerical Methods 1 .( y + y 1 )Δ x plus the area of the trapezoid x 1 P 1 P 2 x 2 that is equal to aP0 P 1 x 1 that is equal to -2 0 1 -.= -.1) Example 10.( y + y 1 )Δ x + -. the trapezoidal approximation becomes 2 1 1 1 1 .( y + y 2 )Δ x . Third Edition Copyright © Orchard Publications . estimate the value of the definite integral ∫1 x d x 2 2 (10. Then. and so on.y + y1 + y2 + … + yn – 1 + 1 T = ⎛ -⎝2 0 2 n⎠ Trapezoidal Rule (10. and compute the percent error.3) For the trapezoidal rule approximation we have x0 = a = 1 xn = b = 2 n = 4 –1 –a -----------.( y + y 2 )Δ x + … + --(y T = -+ y n )Δ x 2 0 2 1 2 n–1 or 1 --y ⎞ Δx . y = f (x) = x 2 10−2 Numerical Analysis Using MATLAB® and Excel®.= 2.– -3 3 3 (10.

The Trapezoidal Rule x0 = a = 1 x1 = a + Δ x = 5 -4 -x2 = a + 2 Δ x = 6 4 -x3 = a + 3 Δ x = 7 4 x4 = b = 2 2 ----y 0 = f ( x 0 ) = 1 = 16 16 2 5 .dx ∫1 -x 2 1 = ln x 2 1 = ln 2 – ln 1 = 0. approximates the integral of a function y = f ( x ) using the trapezoidal rule. we let x 5 represent the row vector with n = 5 .⎞ = 25 y 1 = f ( x 1 ) = ⎛ -----⎝4 ⎠ 16 2 6 .+ -T = ⎛ -⎝ 2 16 16 16 16 2 16 ⎠ 4 16 4 32 (10.34375 – 2. Example 10.2 Use the MATLAB function trapz(x.6931 For the approximation using the trapezoidal rule. Numerical Analysis Using MATLAB® and Excel®.dx ∫1 -x 2 1 (10. 150 1 75 1 16 25 36 49 1 64 ⎞ 1 .= ----.4) From (10.⎞ = 36 y 2 = f ( x 2 ) = ⎛ -----⎝ 4⎠ 16 2 7 .4).45 % % Error = -------------------------------------------2.1).× ----.⎞ = 49 y 3 = f ( x 3 ) = ⎛ -----⎝ 4⎠ 16 8 -⎞ y 4 = f ( x 4 ) = ⎛ -⎝ 4⎠ 2 = 64 ----16 and by substitution into (10.+ ----. Third Edition Copyright © Orchard Publications 10−3 .6) and by comparison with the exact value.n) where y is the integral with respect to x.= -------.× -.y.6931 – 0.34375 . we find that the percent error is 2.= 2.× ----.× 100 = 0.× -.+ ----. and n (optional) performs integration along dimension n .33333 (10.33333 . compute the percent error when n = 5 and n = 10 Solution: The exact value is found from .+ ----.y) to approximate the value of the integral .0000 = 0.5) The MATLAB function trapz(x.3) and (10.

/x5. Use the MATLAB trapz(x. We use the following MATLAB script.6970 – 0.y5).^2)./x10.y) MATLAB displays the following result.7) where τ is a dummy variable of integration.56 % % Error = -------------------------------------0.y) function to find the area under this integral with n = 10 when the upper limit of integration is t = 2 .2.Chapter 10 Integration by Numerical Methods and x 10 the vector with n = 10 . and the areas under the curve as area5 and area10 respectively. y10=1. Solution: We use the same procedure as in the previous example. x5=linspace(1.3 The integral f( t) = ∫0 t e –τ 2 dτ (10.10). y=exp(−t.u –τ2 * The formal definition of the error function is erf ( u ) = -----∫ e dτ π 0 10−4 Numerical Analysis Using MATLAB® and Excel®.6970 area10 = 0. y5=1.6931 Example 10. area5=trapz(x5.y10) area5 = 0.12 % % Error = -------------------------------------0.× 100 = 0. The MATLAB script for this example is t=linspace(0.× 100 = 0.10). is called the error function* and it is used extensively in communications theory. area=trapz(t. Δ x = 1 ⁄ 5 and Δ x = 1 ⁄ 10 respectively.6939 – 0. Third Edition Copyright © Orchard Publications .6931 . that is.2.5). area10=trapz(x10.6931 .6931 and the percent error when Δ x = 1 ⁄ 10 is used is 0. 2 . The corresponding values of y are denoted as y 5 and y 10 .2.6939 The percent error when Δ x = 1 ⁄ 5 is used is 0. x10=linspace(1.

4 The i – v (current−voltage) relation of a non−linear electrical device is given by i ( t ) = 0.2*sin(3*t))−1).10). Third Edition Copyright © Orchard Publications 10−5 . By any means. but it is not easy. we choose n = 100 .The Trapezoidal Rule area = 0.2.*t).10) An analytical solution of the last integral is possible using integration by parts. For this example.1 ∫0 10 s sin 3t ( e 0.1 sin 3t ( e 0. v=sin(3.y) function.1 ( e 0.2 sin 3t – 1 ) dt (10. i=0. MATLAB returns the following message when integration is attempted with the symbolic expression of (10. t=linspace(0. When MATLAB cannot find a solution. We can try the MATLAB int(f. that is. it returns a warning.*i.0.8) where v ( t ) = sin 3t .a.*v)−1). Next. p=v.10. Numerical Analysis Using MATLAB® and Excel®. The energy W ( t 0. The instantaneous power is p ( t ) = v ( t ) i ( t ) = 0.2v ( t ) – 1) (10.1*sin(3*t)*(exp(0. MATLAB displays the following message.1. t 1 ) = ∫t t1 0 p ( t ) dt = 0. Warning: Explicit integral could not be found.8818 Example 10. The MATLAB script below will compute and plot the power. so that Δ x = 1 ⁄ 100 .100). s=int(0. we will find and sketch the power and energy by the trapezoidal rule using the MATLAB trapz(x. The energy is the integral of the instantaneous power. The instantaneous power p ( t ) b.b) function where f is a symbolic expression. t=sym('t').9) b.2 sin 3t – 1) (10. find a. Solution: a. For this example. and a and b are the lower and upper limits of integration respectively. t 1 ) dissipated in this device from t 0 = 0 to t 1 = 10 s.*(exp(0.10) When this script is executed. W ( t 0.

Chapter 10 Integration by Numerical Methods plot(t. '+').015 watts 0.01 0.p).p). 0.2.11) 10−6 Numerical Analysis Using MATLAB® and Excel®.2 shows that the power is uniform for all time. The MATLAB script below computes and plots the energy. and thus we expect the energy to be constant. Let the curve of Figure 10.2 Simpson’s Rule The trapezoidal and Simpson’s rules are special cases of the Newton−Cote rules which use higher degree functions for numerical integration.. The energy is shown in Figure 10. Plot for the power variation in Example 10..1013 joule. grid.energy.02 0. plot(t.. xlabel('seconds').1013 Thus. title('Energy vs Time'). energy=trapz(t. 10. xlabel('seconds'). title('Power vs Time'). ylabel('watts') The power varies in a uniform fashion as shown by the plot of Figure 10. grid. Third Edition Copyright © Orchard Publications .4 The plot of Figure 10.025 Power vs Time 0.3. ylabel('joules') energy = 0.4 be represented by the parabola y = αx + βx + γ 2 (10. the value of the energy is 0.005 0 0 1 2 3 4 5 seconds 6 7 8 9 10 Figure 10.2.

5 joules 0 -0.12) 1 .+ γx + β x + γ ) dx = -------3 2 2 3 2 αx 3 βx 2 h –h 3 αh βh h .4 y y0 y1 y2 x −h 0 h Figure 10.+ ------βh αh . Simpson’s rule of integration The area under this curve for the interval – h ≤ x ≤ h is Area h –h = ∫– h ( α x 3 h 2 .+ ------.Simpson’s Rule 1. Third Edition Copyright © Orchard Publications 10−7 .+ ------. and ( h.+ γh – ⎛– α -------. Then.– γ h⎞ = 2 ----------.4.11) we have: Numerical Analysis Using MATLAB® and Excel®.3.5 -1 0 1 2 3 4 5 seconds 6 7 8 9 10 Figure 10. y2 ) . by (10. y 1 ) . y 0 ) . ( 0.+ 2γh = -------⎝ 3 ⎠ 3 2 2 3 (10.5 Energy vs Time 1 0.h ( 2α h3 + 6 γ ) = -3 The curve passes through the three points ( – h. Plot for the energy of Example 10.

the area under segment AB is 10−8 Numerical Analysis Using MATLAB® and Excel®.18) Now.17) or 1 .5.h ( 2 α h 3 + 6 γ ) = -= -3 3 Area h –h (10. y 0 .18) to successive segments of any curve y = f ( x ) in the interval a ≤ x ≤ b as shown on the curve of Figure 10. y 1 and y 2 . This is done with the following procedure. γ and express (10.14) (10. Simpson’s rule of integration by successive segments From Figure 10.15) yields 2 α h = y 0 – 2y 1 + y 2 2 (10. A B C y0 y1 h a y2 y3 y4 yn – 1 yn b Figure 10.13) We can now evaluate the coefficients α.12) in terms of h . Thus.5. Third Edition Copyright © Orchard Publications .13) into (a) and (c) and rearranging we obtain α h – β h = y0 – y1 α h + β h = y2 – y1 2 2 (10.16) and by substitution into (10.h [ ( y 0 – 2y 1 + y 2 ) + 6y 1 ] . we can apply (10.12) we obtain Area h –h 1 1 .h ( y 0 + 4y 1 + y 2 ) = -3 (10. we observe that each segment of width 2h of the curve can be approximated by a parabola through its ends and its midpoint. β.14) with (10. By substitution of (b) of (10.15) Addition of (10.5.Chapter 10 Integration by Numerical Methods y0 = α h – β h + γ y1 = γ y2 = α h + β h + γ 2 2 (a) (b) (c) (10.

We can also find the analytical value with MATLAB’s int(f. to apply Simpson’s rule of numerical integration. Third Edition Copyright © Orchard Publications 10−9 . we obtain 1 .h ( y 0 + 4y 1 + 2y 2 + 4y 3 + 2y 4 + … + 2y n – 2 + 4y n – 1 + y n ) Area = -3 Simpson ′ s Rule of Numerical Integration (10.23) Solution: This is the same integral as that of Example 10. we construct the following table using the spreadsheet of Figure 10. and a and b are the lower and upper limits of integration respectively.5 Using Simpson’s rule with 4 subdivisions ( n = 4 ) . for convenience. the area under segment BC is Area BC 1 .dx ∫1 -x 2 1 (10. To use Simpson’s rule. When the areas under each segment are added.21) is found from b–a h = ---------n n = even (10.19) Likewise. Numerical Analysis Using MATLAB® and Excel®.6. the number n of subdivisions must be even.Simpson’s Rule Area AB 1 .h ( y 0 + 4y 1 + y 2 ) = -3 (10.6931 . compute the approximate value of .2 where we found that the analytical value of this definite integral is ln = 0. The value of h for (10. syms x Area=int(1/x.21) Since each segment has width 2h .a. This restriction does not apply to the trapezoidal rule of numerical integration.20) and so on.1.22) Example 10.2) Area = log(2) We recall that log(x) in MATLAB is the natural logarithm.b) function where f is a symbolic expression.h ( y 2 + 4y 3 + y 4 ) = -3 (10. For this example.

tol).50000 y2=1/x2= 14 0.31905 = E Products 1. q=quad(‘f’. and a and b are the lower and upper limits of –3 integration respectively.50000 8. The string ‘f’ is the name of a user defined function.57143 4 x4=b= 17 2.5 2 ∫ (1/x)dx evaluated from a = 1 to b = 2 with n = 4 3 Numerical integration by Simpson's method follows a= 1 4 Given b= 2 5 n= 4 6 0.00000 y0=1/x0= 10 1.20000 1. MATLAB displays a warning message but still provides an answer.b.33333 2.25000 y1=1/x1= 12 0. this means that these methods can handle irregularities such as singularities.a.6.23) By comparison of the numerical with the exact value. If tol is omitted.28571 0. 10−10 Numerical Analysis Using MATLAB® and Excel®.00000 3.69325 Figure 10.31905 0.80000 4 x2=a+2h= 13 1.75000 y3=1/x3= 16 0. h = (b-a)/n = Multiplier 8 x0=a= 9 1.66667 2 x3=a+3h= 15 1. Both of these functions use adaptive quadrature methods. that is.00000 1 x1=a+h= 11 1. Both functions have the same syntax.00000 y4=1/x4= 18 0.50000 1 19 Sum of Products = 20 Area = (h/3)*(Sum of Products) = (1/12)*8. the quad and quad8.Chapter 10 Integration by Numerical Methods A B C D 1 Example 10. MATLAB has two quadrature functions for performing numerical integration. Spreadsheet for numerical integration of (10. The quad8 function detects and handles irregularities more efficiently. The description of these can be seen by typing help quad or help quad8.2500 7 Then. while the quad8 function uses the so−called Newton−Cotes 8−panel rule. When such irregularities occur. Third Edition Copyright © Orchard Publications . The quad function uses an adaptive form of Simpson’s rule. we observe that the error is very small when Simpson’s method is applied. and integrate to a relative error tol which we must specify. it is understood to be the standard tolerance of 10 .

Obtain the value of this integral with the q=quad(‘f’. we need to create and save a function m−file.a.2) % Define symbolic variable x and integrate – 10 – 10 pretty(y) y = 1/2*erf(2)*pi^(1/2) 1/2 1/2 erf(2) pi erf is an acronym for the error function and we can obtain its definition with help erf b.2) We obtain the answer in standard tolerance form as y_std = 0. Obtain the value of this integral with the q=quad(‘f’. We will use format long to display the values with 15 digits.a.10^−10) y_tol = 0. y=int(exp(−x^2). we write and execute the following MATLAB script. We name it errorfcn1.b) function e. First.b.24) a.88208139076242 Numerical Analysis Using MATLAB® and Excel®.Simpson’s Rule Example 10.0. Obtain the value of this integral with the q=quad8(‘f’.a.88211275610253 c. the script and the answer are as follows: y_tol=quad('errorfcn1'. Third Edition Copyright © Orchard Publications 10−11 . Obtain the value of this integral with the q=quad8(‘f’. Use MATLAB’s symbolic int function to obtain the value of this integral b. With this file saved as errorfcn1.m as shown below. function y = errorfcn1(x) y = exp(−x.b) function c.tol) function where tol = 10 Solution: a.a. syms x.m. format long y_std=quad('errorfcn1'.^2).2.tol) function where tol = 10 d.0.6 Given the definite integral y= f(x ) = ∫0 2 e –x 2 dx (10.b. With the specified tolerance.0.

3 ) ) .2. b=input('Enter second point "b" (non−negative): '). we obtain y = ∫a b x 1⁄2 x dx = --------3⁄2 3⁄2 b a 2 .2) y_std8 = 0. y_std8=quad8('errorfcn1'. We will include the input function in the script. then.26) where a and b are non−negative values. % Insert line a=input('Enter first point "a" (non−negative): '). Third Edition Copyright © Orchard Publications . Solution: Evaluating the given integral.2. With the specified tolerance. we need not write a user defined m−file. the numerical values produced by the quad and quad8 functions for each set of data. The script is then saved as Example_10_7.b) where a and b are non−negative.25) at ( ( a.0. Use the fprintf function to display first the analytical values. % fprintf(' \n'). With the standard tolerance. ( 1.10^−10) y_tol8 = 0.8 ).Chapter 10 Integration by Numerical Methods d. % This script displays the approximations obtained with the quad and quad8 functions % with the analytical results for the integration of the square root of x over the % interval (a.( b3 ⁄ 2 – a3 ⁄ 2 ) = -3 (10. ∫a x dx (10.8) . y_tol8=quad8('errorfcn1'. Substitution of the values of the given values of a and b will be included in the MATLAB script below. The sqrt function in a built−in function and therefore.0. 10−12 Numerical Analysis Using MATLAB® and Excel®. 0.7 Using the quad and quad8 functions with standard tolerance. and (3. b ) = ( 0. evaluate the integral y = f(x) = b – 10 tolerance. both quad and quad8 produce the same result.4. 2.88208139076242 We observe that with the 10 Example 10.88208139076194 e.

*(b.k. kq8=quad8('sqrt'.b). quad8: 0. fprintf(' \n') % Insert two lines Now. Example_10_7 Enter first point "a" (non-negative): 0.a..kq8).a.4 Enter second point "b" (non-negative): 2. fprintf(' \n').. fprintf(' \n')... quad8: %f %f \n'.417396 Example_10_7 0.8 Analytical: 0.3 Analytical: 1.221080 Numerical quad.620843 Numerical quad. Third Edition Copyright © Orchard Publications 10−13 .221080 Enter first point "a" (non-negative): 3 Enter second point "b" (non-negative): 8 Analytical: 11.417399 Numerical quad.2 Enter second point "b" (non-negative): 0..5)). quad8: 11..221080 Example_10_7 1. that is.417399 Enter first point "a" (non-negative): 1.620825 11.^(1.620843 Numerical Analysis Using MATLAB® and Excel®. % Insert line fprintf(' Analytical: %f \n Numerical quad.kq. we execute this saved file by typing its name. quad8: 1.Simpson’s Rule k=2/3.5)−a.b).^(1. kq=quad('sqrt'.

it is understood to be the standard tolerance of 10 ..y + y 1 + y 2 + … + y n – 1 + --y Δx T = ⎛ -⎝2 0 2 n⎠ –a ---------. while the quad8 function uses the so−called Newton−Cotes 8− panel rule. Both of these functions use adaptive quadrature methods. • The MATLAB function trapz(x. q=quad(‘f’.n) where y is the integral with respect to x.y. The string ‘f’ is the name of a user defined function. • We can perform numerical integration with the MATLAB function int(f.h ( y 0 + 4y 1 + 2y 2 + 4y 3 + 2y 4 + … + 2y n – 2 + 4y n – 1 + y n ) Area = -3 where the number n of subdivisions must be even. and integrate to a relative error tol which we must specify. • MATLAB has two quadrature functions for performing numerical integration.b. • The trapezoidal and Simpson’s rules are special cases of the Newton−Cote rules which use higher degree functions for numerical integration.tol). The quad8 function detects and handles irregularities more efficiently.a. and a and b are the lower and upper limits of integration respectively. –3 10−14 Numerical Analysis Using MATLAB® and Excel®.a. Third Edition Copyright © Orchard Publications . • We can evaluate a definite integral using the expression ∫a f ( x ) dx b with Simpson’s rule of numerical integration 1 . The number n of by dividing interval a ≤ x ≤ b into n subintervals each of length Δ x = b n subdivisions can be even or odd. The quad function uses an adaptive form of Simpson’s rule. Both functions have the same syntax. and n (optional) performs integration along dimension n .b) function where f is a symbolic expression.Chapter 10 Integration by Numerical Methods 10. the quad and quad8.3 Summary • We can evaluate a definite integral ∫ f ( x ) dx with the trapezoidal approximation a b 1 1 ⎞ . and a and b are the lower and upper limits of integration respectively. approximates the integral of a function y = f ( x ) using the trapezoidal rule. that is. If tol is omitted.

Use Simpson’s rule to approximate the following definite integrals and compare your results with the analytical values. a. ∫ -------------2 0 1 x +1 n = 4 Numerical Analysis Using MATLAB® and Excel®. ∫ sin x dx 0 π n = 4 1 dx c.n) function.a. Verify your answers with the MATLAB quad(‘f’. ∫ x dx 3 0 2 n = 4 c. Verify your answers with the MATLAB trapz(x. ∫ x dx 2 0 2 n = 4 b.4 Exercises 1.y. a. ∫ x dx 0 2 n = 4 b.b) function. Third Edition Copyright © Orchard Publications 10−15 . ∫ ---2 1 2 x n = 4 2.Exercises 10. Use the trapezoidal approximation to compute the values the following definite integrals and compare your results with the analytical values. ∫ x dx 4 0 2 n = 4 1dx d.

y=x.2.= 4×1 -×2⎞ ×1 -×0+1 -.Chapter 10 Integration by Numerical Methods 10. area=trapz(x.y) area = 2 b.+ 1 + -.= 2 Δx = b 2 4 n y = f (x) = x x0 = a = 0 -x1 = a + Δ x = 1 2 x2 = a + 2 Δ x = 1 3 x 3 = a + 3 Δ x = -2 x4 = b = 2 y0 = f ( x0 ) = 0 y1 = f ( x1 ) = 1 -2 y2 = f ( x2 ) = 1 3 y 3 = f ( x 3 ) = -2 y4 = f ( x4 ) = 2 1 3 1 --=2 -.y + y1 + y2 + … + yn – 1 + 1 T = ⎛ -⎝2 0 2 n⎠ a. 1 --y ⎞ Δx .+ -T = ⎛ -⎝2 ⎠ 2 2 2 2 2 x=linspace(0. Third Edition Copyright © Orchard Publications .4). The exact value is 2 4 2 0 ∫0 x x dx = ---4 3 = 4 For the trapezoidal rule approximation we have 10−16 Numerical Analysis Using MATLAB® and Excel®. The exact value is ∫0 2 x x dx = ---2 2 2 0 = 2 For the trapezoidal rule approximation we have x0 = a = 0 xn = b = 2 n = 4 –0 –a -----------.5 Solution to End−of−Chapter Exercises 1.= 1 ---------.

= -n 4 2 y = f (x) = x x0 = a = 0 1 x 1 = a + Δ x = -2 x2 = a + 2 Δ x = 1 -x3 = a + 3 Δ x = 3 2 x4 = b = 2 3 y0 = f ( x0 ) = 0 1 y 1 = f ( x 1 ) = -8 y2 = f ( x2 ) = 1 y 3 = f ( x 3 ) = 27 ----8 y4 = f ( x4 ) = 8 1 1 1 27 1 .y) area = 4.⎞ × --. Third Edition Copyright © Orchard Publications 10−17 .= ----------.4).2.+ 1 + ----.× 8 ⎞ × --×0+1 -.= 4.+ -T = ⎛ -⎝ ⎠ ⎝2 ⎠ 2 2 2 8 8 2 x=linspace(0.^3. The exact value is 2 5 2 0 ∫0 4 xx dx = ---5 = 32 ----.= -n 4 2 y = f (x) = x 4 Numerical Analysis Using MATLAB® and Excel®.= ----------.25 . area=trapz(x. y=x. c.4 5 For the trapezoidal rule approximation we have x0 = a = 0 xn = b = 2 n = 4 b–a 2–0 1 Δ x = ---------.= ⎛5 + 7 .Solution to End−of−Chapter Exercises x0 = a = 0 xn = b = 2 n = 4 b–a 2–0 1 Δ x = ---------.= 6.4444 The deviations from the exact value are due to the small number of divisions n we chose.

Chapter 10 Integration by Numerical Methods x0 = a = 0 1 x 1 = a + Δ x = -2 x2 = a + 2 Δ x = 1 -x3 = a + 3 Δ x = 3 2 x4 = b = 2 y0 = f ( x0 ) = 0 1 y 1 = f ( x 1 ) = ----16 y2 = f ( x2 ) = 1 y 3 = f ( x 3 ) = 81 ----8 y 4 = f ( x 4 ) = 16 1 1 1 1 81 1 .= ----------.^4.+ --.⎞ × -. Third Edition Copyright © Orchard Publications .= ⎛ 3905 --⎞ ×1 -×1 . area=trapz(x.× 0 + --------.5090 -. The exact value is ∫1 2 1 ---.= -n 4 4 y = f (x) = 1 ⁄ x x0 = a = 1 -x1 = a + Δ x = 5 4 3 x 2 = a + 2 Δ x = -2 -x3 = a + 3 Δ x = 7 4 x4 = b = 2 2 y0 = f ( x0 ) = 1 y 1 = f ( x 1 ) = 16 ----25 4 y 2 = f ( x 2 ) = -9 y 3 = f ( x 3 ) = 16 ----49 1 y 4 = f ( x 4 ) = -4 1 1 -----------⎞ × 1 -.5720 d.+ 1 + ----. y=x.× 1 + 16 ----.dx = – 1 -2 x x 2 1 1 = -2 For the trapezoidal rule approximation we have x0 = a = 1 xn = b = 2 n = 4 b–a 2–1 1 Δ x = ---------.= 7.= ⎛ 9 + 41 .2.0625 .× 16 ⎞ × -.y) area = 7.+ -T = ⎛ -⎝ ⎠ ⎝2 ⎠ 2 2 8 16 16 2 x=linspace(0.= 0.+ 16 -----+4 T = ⎛ -⎝ ⎠ 4 ⎝2 ⎠ 1918 4 4 2 25 9 49 10−18 Numerical Analysis Using MATLAB® and Excel®.4).

00000 1. We write and execute the following MATLAB script: y_std=quad('exer_10_2_a'. y=1.m as shown below. The exact value is ∫0 A 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 2 x x dx = ---3 2 3 2 0 8 = -. B C D E Exercise 10.50000 0.0000 16.0000 2 2.5158 2.y) area = 0.6667 3 To use Simpson’s rule we construct the following table using a spreadsheet.0. h = (b-a)/n = x0=a= y0=x02= x1=a+h= y1=x12= x2=a+2h= y2=x22= x3=a+3h= y3=x32= x4=b= y4=x42= 0 2 4 0.31905 = 1 4.4).0000 1 0.h ( y 0 + 4y 1 + 2y 2 + 4y 3 + 2y 4 + … + 2y n – 2 + 4y n – 1 + y n ) Area = -3 a. We name it exer_10_2_a.6667 4 9.5000 Multiplier 0.a ∫ x 2 dx evaluated from a = 0 to b = 2 with n = 4 Numerical integration by Simpson's method follows Given a= b= n= Then.00000 0. function y = exer_10_2_a(x) y = x.00000 1.00000 Sum of Products = Area = (h/3)*(Sum of Products) = (1/12)*8.^2.0000 2.= 2.2.25000 2. area=trapz(x.^2.0000 Products We create and save a function m−file.6667 Numerical Analysis Using MATLAB® and Excel®./x.0000 4 1.00000 0.2. Third Edition Copyright © Orchard Publications 10−19 .2) y_std = 2.50000 2.25000 1.Solution to End−of−Chapter Exercises x=linspace(1.00000 4. 1 .

8284 1 0.0000 c.0000 7. function y = exer_10_2_b(x) y = sin(x). Third Edition Copyright © Orchard Publications .0.78540 0.0000 Products We create and save a function m−file.b ∫ sinxdx evaluated from a = 0 to b = π with n = 4 Numerical integration by Simpson's method follows Given a= b= n= Then.70711 3.31905 = 1 0.m as shown below. We write and execute the following MATLAB script: y_std=quad('exer_10_2_b'. We name it exer_10_2_b.2.0000 4 2.0046 4 2.00000 0.00000 Sum of Products = Area = (h/3)*(Sum of Products) = (1/12)*8.8284 2 2.35619 0.dx = tan – 1x 2 x +1 1 0 π = -. The exact value is π ∫0 sin x dx A 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 B = – cos x π 0 = –( – 1 – 1 ) = 2 To use Simpson’s rule we construct the following table using a spreadsheet. h = (b-a)/n = x0=a= y0=sinx0= x1=a+h= y1=sinx1= x2=a+2h= y2=sinx2= x3=a+3h= y3=sinx3= x4=b= y4=sinx4= 0 3.7854 Multiplier 0.pi) y_std = 2.14159 0.7854 4 To use Simpson’s rule we construct the following table using a spreadsheet.14159 4 0. 10−20 Numerical Analysis Using MATLAB® and Excel®.6569 2.Chapter 10 Integration by Numerical Methods b.70711 1.00000 2.00000 0. C D E Exercise 10.57080 1.= 0. The exact value is ∫0 1 1 -------------.

25000 0.80000 0.^2+1).0.5000 9.7647 1 1. h = (b-a)/n = x0=a= y0=1/(x02+1)= x1=a+h= y1=1/(x12+1)= x2=a+2h= y2=1/(x22+1)= x3=a+3h= y3=1/(x32+1)= x4=b= y4=1/(x42+1)= 0 1 4 0.0000 Products We create and save a function m−file.00000 0. We name it exer_10_2_c.64000 1.00000 0. We write and execute the following MATLAB script: y_std=quad('exer_10_2_c'. function y = exer_10_2_c(x) y = 1.50000 Sum of Products = Area = (h/3)*(Sum of Products) = (1/12)*8.m as shown below.2.00000 1.2500 Multiplier 0. Third Edition Copyright © Orchard Publications 10−21 .5600 2 1.1) y_std = 0.7854 4 2.31905 = 1 0.7854 Numerical Analysis Using MATLAB® and Excel®.50000 0./(x.6000 4 3.4247 0.75000 0.c ∫ (1/(x 2 +1))dx evaluated from a = 0 to b = 1 with n = 4 Numerical integration by Simpson's method follows Given a= b= n= Then.94118 0.Solution to End−of−Chapter Exercises A 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 B C D E Exercise 10.

Difference equations as used with discrete type systems. In terms of the interval h . we increase the argument of f by another unit. and the subscript k is normally omitted. The E operator increases the argument of a function by one interval h . are used in numerous applications such as engineering.1) where a k represents a constant coefficient and E is an operator similar to the D operator in ordinary differential equations.2 Definition. h = 1 . we obtain the second order operator E 2 . Third Edition Copyright © Orchard Publications 11−1 . Solutions. mathematics. that is. and a practical example in electric circuit theory is given at the end of this chapter. constant coefficient difference equation has the form ( ar E + ar – 1 E r r–1 + … + a1 E + a0 ) y = φ ( x ) (11. a recurrence relation is an equation which defines a sequence recursively: each term of the sequence is defined as a function of the preceding terms. The general form of a linear.Chapter 11 Difference Equations T his chapter is an introduction to difference equations based on finite differences. A difference equation is a specific type of recurrence relation.3). 11. and r is a positive integer that denotes the order of the difference equation. The Fibonacci numbers are defined.e. i. (11.2) The interval h is usually unity.2) is written as Ef ( x ) = f ( x + 1 ) = f x + 1 (11. Thus.1 Introduction In mathematics.3) If.. and other sciences. are discussed in Appendix A. the difference operator E is Ef ( x k ) = f ( x k + h ) = f ( x k + h ) (11. physics. and Applications The difference equations discussed in this chapter. 11. Numerical Analysis Using MATLAB® and Excel®. and this type is discussed in this chapter. in (11. The discussion is limited to linear difference equations with constant coefficients.

4) (11. then the complete solution is y = k1 y1 ( x ) + k2 y2 ( x ) + Y ( x ) (11. then. the right side of (11. any other solution of (11. is non−zero. we first find the solution of the homogeneous difference equation.1).7) If y 1 ( x ) and y 2 ( x ) are any two solutions of (11. cos kx .8) then.5) and in general. where k is a non−zero constant and n is a non−negative integer.Chapter 11 Difference Equations E f ( x ) = E [ Ef ( x ) ] = Ef ( x + 1 ) = f ( x + 2 ) = f x + 2 2 (11.3) is a linear combination of terms such as kx .10). we let r = 2 .1). analogous to the Wronskian determinant in ordinary differential equations. Also. we obtain the second order difference equation ( a2 E + a1 E + a0 ) y = φ ( x ) 2 (11.10) where ϕ ( x ) ≠ 0 . it is a non−homogeneous difference equation. y 2 ( x ) ] = y1 ( x ) Ey 1 ( x ) y2 ( x ) Ey 2 ( x ) ≠0 (11. if Y ( x ) is any solution of (11.6) and if the right side is zero.11) As with ordinary differential equations.7). the linear combination k 1 y 1 ( x ) + k 2 y 1 ( x ) is also a solution. and x . in (11. the equation is referred to as a homogeneous difference equation.7) can be expressed as y3 ( x ) = k1 y1 ( x ) + k2 y2 ( x ) (11. if C [ y 1 ( x ). Third Edition Copyright © Orchard Publications . 11−2 Numerical Analysis Using MATLAB® and Excel®. If. We find Y ( x ) by the Method of Undetermined Coefficients. E f ( x ) = f ( x + r ) = fx + r n r As with ordinary differential equations. if. we add the particular solution Y ( x ) to it to obtain the total solution. if the Casorati determinant. ϕ ( x ) = 0 . that is. in (11. and if ϕ ( x ) ≠ 0 . it reduces to ( a2 E + a1 E + a0 ) y = 0 2 (11.9) where k 1 and k 2 are constants. Moreover. For the non−homogeneous difference equation ( a2 E + a1 E + a0 ) y = φ ( x ) 2 (11.

1.Definition. zero. real and equal. As with algebraic quadratic equations. Solution: The characteristic equation of (11.13) can be real and unequal. the roots of (11. for the homogeneous difference equation we assume a solution of the form y = M x ax (11. Third Edition Copyright © Orchard Publications 11−3 . in analogy with the solution of the differential equation of the form y = ke .7). TABLE 11. we obtain a2 M x+2 + a1 M x+1 + a0 M x = 0 (11. Solutions.13) and this is the characteristic equation of a second order difference equation.10) are constants.14) with initial conditions y 0 = y ( 0 ) = 3 and y 1 = y ( 1 ) = 2 .1 Roots of the characteristic equation in difference equations Characteristic equation a 2 M + a 1 M + a 0 = 0 of ( a 2 E + a 1 E + a 0 ) y = 0 2 2 2 Roots M 1 and M 2 Real and Unequal M1 ≠ M2 Real and Equal M1 = M2 Discriminant a 1 – 4a 2 a 0 > 0 2 General Solution x x y = k1 M1 + k2 M2 k 1 and k 2 cons tan ts a 1 – 4a 2 a 0 = 0 2 x x y = k 1 M 1 + k 2 xM 2 k 1 and k 2 cons tan ts Complex Conjugates M1 = α + j β M2 = α –j β a 1 – 4a 2 a 0 < 0 2 x y = r ( C 1 cos θ x + C 2 sin θ x ) r = α +β 2 2 –1 β θ = tan --α Example 11.14) is Numerical Analysis Using MATLAB® and Excel®. and Applications We have assumed that the coefficients a i in (11. or negative. compute y 5 = y ( 5 ) .1 Find the solution of the difference equation ( E –6 E + 8 ) y = 0 2 (11. and recalling that Ef ( x ) = f ( x + 1 ) . or complex conjugates depending on whether the discriminant a 1 – 4a 2 a 0 is positive.12) By substitution into (11. These cases are summarized in Table 11. Then. then.

Chapter 11 Difference Equations M –6 M + 8 = 0 2 (11.21) ( –1 ) + ( 3 ) = 2 2 2 and its roots are M 1 = – 1 + j 3 and M 2 = – 1 – j 3 .18) yields k 1 = 5 and k 2 = – 2 .16) becomes y1 = 2 = k1 2 + k2 4 1 1 or 2k 1 + 4k 2 = 2 (11. we obtain the solution yx = y ( x ) = k1 2 + k2 4 x x (11. Therefore. Therefore. with reference to Table 11.16) becomes y0 = 3 = k1 2 + k2 4 0 0 or k1 + k2 = 3 (11.19) For x = 5 . From Table 11.1. we obtain y 5 = 5 ⋅ 2 – 2 ⋅ 4 = 5 × 32 – 2 × 1024 = – 1888 5 5 Example 11. (11. Thus.20) Solution: The characteristic equation of (11.2 Find the solution of the difference equation ( E + 2E + 4 ) y = 0 2 (11. the solution is –1 11−4 Numerical Analysis Using MATLAB® and Excel®. we let x = 0 .18) Simultaneous solution of (11. Then.20) is M + 2M + 4 = 0 2 (11.16) To make use of the first initial condition. we let x = 1 .1. Then. Third Edition Copyright © Orchard Publications .17) For the second initial condition.15) and its roots are M 1 = 2 and M 2 = 4 .17) and (11. (11. the solution is yx = 5 ⋅ 2 – 2 ⋅ 4 x x (11. r = and θ = tan ( 3 ⁄ ( – 1 ) ) = 2 π ⁄ 3 .

x⎞ y = 2 ⎛ C 1 cos ----⎝ 3 3 ⎠ (11.23) The characteristic equation of (11.11). φ ( x ) .22) The constants C 1 and C 2 can be evaluated from the initial conditions. we start with a linear combination of all the terms of the right side.24) Numerical Analysis Using MATLAB® and Excel®.3 Find the solution of the difference equation Solution: ( E –5 E + 6 ) y = x + 2 2 x (11. that is.2 shows the form of the particular solution for different terms of φ ( x ) . For the particular solution. For non−homogeneous difference equations of the form of (11.2 Form of the particular solution for a non−homogeneous difference equation Non−homogeneous difference equation ( a 2 E + a 1 E + a 0 ) y = φ ( x ) φ (x) α (constant) α x (k = positive integer) αk x k k 2 Form of Particular Solution Y ( x ) A (constant) Ak x + Ak – 1 x Ak x k–1 + … + A1 x + A0 α cos mx or α sin m x α x l cos mx or α x l sin m x k x k x A 1 cos mx + A 2 sin mx ( Akx + Ak – 1 x k k k–1 + … + A 1 x + A 0 ) l cos mx + … + B 1 x + B 0 ) l sin m x x x + ( Bk x + Bk – 1 x k–1 Example 11.23) is M –5 M + 6 = 0 2 (11. we combine the particular solution with the solution of the homogeneous equation shown in (11. Third Edition Copyright © Orchard Publications 11−5 . Solutions. this choice must be multiplied by x until there is no duplication of terms. and we apply the operator E . and Applications x 2π π . TABLE 11.x + C 2 sin 2 ----. Table 11. If any of the terms in the initial choice duplicates a term in the solution of the homogeneous equation.10).Definition.

**Chapter 11 Difference Equations
**

and its roots are M 1 = 2 and M 2 = 3 . From Table 11.1, the solution Y H of the homogeneous difference equation is

YH = k1 2 + k2 3

x x

(11.25)

x x x,

For the particular solution we refer Table 11.2. For the first term x of the right side of (11.23), we use the term A 1 x + A 0 , or Ax + B . For the second term 2 , we obtain A2 or C2 particular solution has the form

Y P = Ax + B + C2

x x

and thus, the

(11.26)

But the term C2 in (11.26), is also a term in (11.25). Therefore, to eliminate the duplication, we multiply the term C2 by x . Thus, the correct form of the particular solution is

Y P = Ax + B + Cx2

x x

(11.27)

To evaluate the constants

A,

**B , and C , we substitute (11.27) into (11.23). Then,
**

x+2

[A(x + 2) + B + C(x + 2) ⋅ 2

m+n

] – 5[A(x + 1) + B + C(x + 1) ⋅ 2 ] x x + 6 [ Ax + B + Cx2 ] = x + 2

m n

x+1

(11.28)

Using the law of exponents W

= W

**× W , simplifying, and equating like terms, we obtain
**

x

2Ax + ( – 3A + 2B ) – 2C2

= x+2

x

(11.29)

**Relation (11.29) will be true if
**

2A = 1 – 3A + 2B = 0 B = 0.75 – 2C = 1 C = – 0.5

or

A = 0.5

**By substitution into (11.28), we obtain the particular solution
**

Y P = 0.5x + 0.75 – 0.5 x2

x

(11.30)

**Therefore, the total solution is the sum of (11.25) and (11.30), that is,
**

y total = Y H + Y P = k 1 2 + k 2 3 + 0.5x + 0.75 – 0.5 x2

x x x

(11.31)

11−6

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Fibonacci Numbers 11.3 Fibonacci Numbers
**

The Fibonacci numbers are solutions of the difference equation

yx + 2 = yx + 1 + yx

(11.32)

that is, in a series of numbers, each number after the second, is the sum of the two preceding numbers. Example 11.4 Given that y 0 = 0 and y 1 = 1 , compute the first 12 Fibonacci numbers. Solution: For x = 0, 1, 2, 3 and so on, we obtain the Fibonacci numbers

1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, …

We will conclude this chapter with an application to electric circuit analysis. Example 11.5 For the electric network of Figure 11.1, derive an expression for the voltage V x at each point P x where x = 0, 1, 2, …, n , given that the voltage V 0 at point P 0 is known.

P0 R P1

+

P2 R V1 2R

+

P3 R 2R

+

Pn – 3

+

Pn – 2

+

Pn – 1 R

+

Pn R

+

+

V0

−

2R

−

−

V2

−

V3

2R

−

R V n – 3 2R

−

V n – 2 2R

−

Vn – 1

Vn

−

. Solution:

Figure 11.1. Electric network for Example 11.5

We need to derive a difference equation that relates the unknown voltage V x to the known voltage V 0 . We start by drawing part of the circuit as shown in Figure 11.2, and we denote the voltages and currents as indicated.

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

11−7

**Chapter 11 Difference Equations
**

Px

+

Px + 1

R 2R

+

Px + 2

R

+

2R

−

Vx

−

V x + 1 2R

−

Vx + 2

Figure 11.2. Part of the circuit of Figure 11.1

**By application of Kirchoff’s Current Law (KCL) at node P x + 1 of Figure 11.2, we obtain
**

Vx + 1 – Vx Vx + 1 Vx + 1 – Vx + 2 -------------------------+ ------------- + -------------------------------- = 0 R R 2R

(11.33)

**and after simplification,
**

2 ---(V – 2.5 V x + 1 + V x ) = 0 R x+2

**Of course, the term 2 ⁄ R cannot be zero. Therefore, we must have
**

V x + 2 – 2.5 V x + 1 + V x = 0

(11.34)

Relation (11.34) is valid for all points except P 1 and P n – 1 * as shown in Figure 11.1; therefore, we must find the current relations at these two points. Also, by application of Kirchoff’s current law (KCL) at node P 1 of Figure 11.1, we obtain

V1 – V0 V1 V1 – V2 ------------------ + ------ = 0 - + -----------------R R 2R

**and after simplification,
**

V 2 – 2.5V 1 + V 0 = 0

(11.35)

**Likewise, at node P n – 1 of Figure 11.1, we obtain
**

Vn – 1 – Vn – 2 Vn – 1 Vn – 1 – Vn - + -------------------------------------------------------- + ------------ = 0 2R R R

**Observing that V n = 0 , and simplifying, we obtain
**

2.5V n – 1 – V n – 2 = 0

(11.36)

* The voltages at nodes P 0 and P n are V 0 and Vn respectively.

11−8

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

Fibonacci Numbers

Relation (11.35) is a difference equation of the form

( E – 2.5E + 1 ) y = 0

2

**where y = V x . Its characteristic equation is
**

M – 2.5M + 1 = 0

2

(11.37)

**The roots of the characteristic equation of (11.37) are M 1 = 0.5 and M 2 = 2 . Thus, the solution is
**

y = V x = k 1 ( 0.5 ) + k 2 ( 2 )

x x

(11.38)

The constant coefficients k 1 and k 2 in (11.38), are found by substitution of this relation into (11.35) and (11.36). Thus, from (11.37) and (11.38), we obtain

k 1 ( 0.5 ) + k 2 ( 2 ) – 2.5 ( k 1 ( 0.5 ) + k 2 ( 2 ) ) + V 0 = 0

2 2 1 1

or

0.25k 1 + 4k 2 – 1.25k1 – 5k 2 + V 0 = 0

or

k1 + k2 = V0

(11.39)

**Likewise, from (11.38) and (11.36) we obtain
**

n – 1⎞ n–2 1 ⎞n – 1 1 ⎞n – 2 2.5 ⎛ k 1 ⎛ -+ k2 ( 2 ) – k 1 ⎛ -– k2 ( 2 ) = 0 ⎝ ⎝2⎠ ⎠ ⎝2⎠

or

k1 2.5k 1 n–1 n–2 ------------ + 2.5k 2 ( 2 ) – ----------– k2 ( 2 ) = 0 n–2 n–1 2 2

or

2 ( 2.5 ) k 1 2.5k 2 ( 2 ) 4k 1 k 2 ( 2 ) - – ------- = 0 -------------------- + ----------------------- – --------------n n 2 4 2 2 k1 n ----- + k2 ( 2 ) = 0 n 2

n n

or

(11.40)

**Simultaneous solution of (11.39) and (11.40) yields
**

2 - V0 k 1 = ---------------2n 2 –1

2n

–1 - V0 k 2 = ---------------2n 2 –1

(11.41)

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

11−9

**Chapter 11 Difference Equations
**

Finally, substitution of (11.41) into (11.38) yields a solution of the difference equation in terms of V 0 , that is,

x 2 1 ⎞x –1 - + ---------------- V 0 ⎛ -- V0 ( 2 ) y = V x = ---------------⎝ ⎠ 2n 2n 2 2 –1 2 –1

2n

or

⎛ 2 2n x⎞ V 0 y = V x = ⎜ ------- – 2 ⎟ ---------------⎝ 2x ⎠ 22 n – 1

(11.42)

**We observe that when x = 0 ,
**

2n V0 2 - = V0 y = V x = ⎛ ------- – 1⎞ ---------------⎝ 1 ⎠ 2n 2 –1

and when x = n ,

⎛ 2n V0 n⎞ V 0 - = ( 2 n – 2 n ) ---------------- = 0 ------- – 2 ⎟ ---------------y = Vx = ⎜ 2 2n 2n n ⎝2 ⎠2 – 1 2 –1

11−10

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

Summary 11.4 Summary

• The general form of a linear, constant coefficient difference equation has the form

( ar E + ar – 1 E

r r–1

+ … + a1 E + a0 ) y = φ ( x )

where a k represents a constant coefficient and E is an operator similar to the D operator in ordinary differential equations. As with ordinary differential equations, the right side is a linear combination of terms such as kx , cos kx , and x , where k is a non−zero constant and n is a non−negative integer. If ϕ ( x ) = 0 , the equation is referred to as a homogeneous difference equation, and if ϕ ( x ) ≠ 0 , it is a non−homogeneous difference equation. • The difference operator E is

Ef ( x k ) = f ( x k + h ) = f ( x k + h )

n

**The interval h is usually unity, i.e., h = 1 , and the subscript k is normally omitted. Thus, (11.3) is written as
**

Ef ( x ) = f ( x + 1 ) = f x + 1

and in general,

E f ( x ) = f ( x + r ) = fx + r

r

• If y 1 ( x ) and y 2 ( x ) are any two solutions of a homogeneous difference equation, the linear combination k 1 y 1 ( x ) + k 2 y 1 ( x ) , where k 1 and k 2 are constants, is also a solution. • If the Casorati determinant, analogous to the Wronskian determinant in ordinary differential equations, is non−zero, that is, if

C [ y 1 ( x ), y 2 ( x ) ] = y1 ( x ) Ey 1 ( x ) y2 ( x ) Ey 2 ( x ) ≠0

**then, any other solution of the homogeneous difference equation can be expressed as
**

y3 ( x ) = k1 y1 ( x ) + k2 y2 ( x )

**• For the non−homogeneous difference equation
**

( a2 E + a1 E + a0 ) y = φ ( x )

2

**where ϕ ( x ) ≠ 0 , if Y ( x ) is any solution of (11.11), then the complete solution is
**

y = k1 y1 ( x ) + k2 y2 ( x ) + Y ( x )

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

11−11

**Chapter 11 Difference Equations
**

As with ordinary differential equations, we first find the solution of the homogeneous difference equation; then, we add the particular solution Y ( x ) to it to obtain the total solution. We find Y ( x ) by the Method of Undetermined Coefficients. • In analogy with the solution of the differential equation of the form y = ke , for the homogeneous difference equation, we assume a solution of the form

y = M

x ax

**• Since Ef ( x ) = f ( x + 1 ) , the characteristic equation of a second order difference equation is
**

a2 M

x+2

+ a1 M

x+1

+ a0 M

x

= 0

and as with algebraic quadratic equations, the roots can be real and unequal, real and equal, or complex conjugates depending on whether the discriminant a 1 – 4a 2 a 0 is positive, zero, or negative. These cases are summarized in Table 11.1. • For non−homogeneous difference equations we combine the particular solution with the solution of the homogeneous equation. For the particular solution, we start with a linear combination of all the terms of the right side, that is, φ ( x ) , and we apply the operator E . If any of the terms in the initial choice duplicates a term in the solution of the homogeneous equation, this choice must be multiplied by x until there is no duplication of terms. The form of the particular solution for different terms of φ ( x ) is shown in Table 11.2. • The Fibonacci numbers are solutions of the difference equation

yx + 2 = yx + 1 + yx

2

that is, in a series of numbers, each number after the second, is the sum of the two preceding numbers.

11−12

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

Exercises 11.5 Exercises

Find the total solution of the following difference equations. 1. ( E + 7E + 12 ) y = 0 2. ( E + 2E + 2 ) y = 0 3. ( E – E – 6 ) y = x + 3 4. ( E + 1 ) y = sin x

2 2 2 2

x

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

11−13

**Chapter 11 Difference Equations 11.6 Solutions to End−of−Chapter Exercises
**

1. The characteristic equation is

( E + 7E + 12 ) y = 0 M + 7M + 12 = 0

2 2

**and its roots are M 1 = – 3 and M 2 = – 4 . Therefore, with reference to Table 11.1, we obtain the solution
**

yx = y ( x ) = k1 ( –3 ) + k2 ( –4 )

x x

(1)

The constants k 1 and k 2 can be evaluated from the initial conditions. Since they were not given, let us assume that y 0 = y ( 0 ) = 1 and y 1 = y ( 1 ) = 2 . To make use of the first initial condition, we let x = 0 . Then, (1) becomes

y0 = 1 = k1 ( –3 ) + k2 ( –4 )

0 0

or

k 1 + k 2 = 1 (2)

**For the second initial condition, we let x = 1 . Then, (1) becomes
**

y1 = 2 = k1 ( –3 ) + k2 ( –4 )

1 1

or

– 3 k 1 – 4k 2 = 2 (3)

**Simultaneous solution of (2) and (3) yields k 1 = 6 and k 2 = – 5 . Thus, the solution is
**

yx = y ( x ) = 6 × ( – 3 ) –5 × ( –4 )

x x

**2. The characteristic equation is
**

( E + 2E + 2 ) y = 0 M + 2M + 2 = 0

2 2

**and its roots are M 1 = – 1 + j and M 2 = – 1 – j . From Table 11.1, r =
**

θ = tan 1 ⁄ ( – 1 ) = – π ⁄ 4 . Therefore, the solution is y = 2 ( C 1 cos ( – π ⁄ 4 ) x + C 2 sin ( – π ⁄ 4 ) x )

x

–1

( –1 ) + ( 1 ) =

2

2

2 and

11−14

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

**Solutions to End−of−Chapter Exercises
**

The constants C 1 and C 2 can be evaluated from the initial conditions. For this exercise, they were not given. 3. The characteristic equation is

(E – E – 6) y = x + 3 M –M – 6 = 0

2 2

x

**and its roots are M 1 = – 2 and M 2 = 3 From Table 11.1, the solution Y H of the homogeneous difference equation is
**

Y H = k 1 ( – 2 ) + k 2 3 (1)

x x

For the particular solution we refer Table 11.2. For the first term x of the right side of the given equation we use the term A 1 x + A 0 , or Ax + B . For the second term 3 , we obtain A3 or C3

x, x x

**and thus, the particular solution has the form
**

Y P = Ax + B + C3

x x x

But the term C3 is also a term in the given equation. Therefore, to eliminate the duplication, we multiply the term C3 by x . Thus, the correct form of the particular solution is

Y P = Ax + B + Cx3

x

(2)

To evaluate the constants equation. Then,

A,

**B , and C , we substitute the last expression above into the given
**

x+2 x+1

[A(x + 2) + B + C(x + 2) ⋅ 3

m+n

] –[ A ( x + 1 ) + B + C ( x + 1 ) ⋅ 3 ] x x – 6 [ Ax + B + Cx3 ] = x + 3

m

Using the law of exponents W obtain

= W

× W , simplifying, and equating like terms, we

n

– 6 Ax + ( A – 6 B ) + 15C3

x

= x+3

x

**This relation will be true if
**

– 6A = 1 A –6 B = 0 B = 1 ⁄ 36 15C = 1 C = 1 ⁄ 15

or

A = –1 ⁄ 6

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

11−15

**Chapter 11 Difference Equations
**

By substitution into (2), we obtain the particular solution

Y P = ( – 1 ⁄ 6 ) x + 1 ⁄ 36 + ( 1 ⁄ 15 ) x2

x

(3)

**Therefore, the total solution is the sum of (1) and (3), that is,
**

y total = Y H + Y P = k 1 ( – 2 ) + k 2 3 + ( – 1 ⁄ 6 ) x + 1 ⁄ 36 + ( 1 ⁄ 15 ) x2

x x x

**4. The characteristic equation is
**

( E + 1 ) y = sin x (1) M +1 = 0

2 2

**and its roots are M 1 = j and M 2 = – j From Table 11.1, r = of the solution is
**

( 1 ) = 1 and θ = tan 1 ⁄ 0 = π ⁄ 2 . Therefore, the homogeneous part Y H = C 1 cos ( π ⁄ 2 ) x + C 2 sin ( π ⁄ 2 ) x (2)

2 –1

For the particular solution we refer Table 11.2 where we find that the solution has the form A 1 cos mx + A 2 sin mx , and for this exercise

Y P = A cos x + B sin x

Since the cosine and sine terms appear in the complimentary solution, we multiply the terms of the particular solution by x and we obtain

Y P = Ax cos x + Bx sin x (3)

To evaluate the constants we obtain Using the trig identities

A,

B , and C , we substitute the last expression above into (1) and

A ( x + 2 ) cos ( x + 2 ) + B ( x + 2 ) sin ( x + 2 ) + Ax cos x + Bx sin x = sin x

cos ( a + b ) = cos acos b – sin asin b sin ( a + b ) = sin acos b – sin b cos a

**expanding, rearranging, equating like terms, and combining the complimentary and particular solutions we obtain
**

π sin x + sin ( x – 2 ) - x + C 2 sin π -- x + ----------------------------------------y = C 1 cos -2 2 2 ( 1 + cos 2 )

11−16

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

Chapter 12

Partial Fraction Expansion

his chapter is an introduction to partial fraction expansion methods. In elementary algebra we learned how to combine fractions over a common denominator. Partial fraction expansion is the reverse process and splits a rational expression into a sum of fractions having simpler denominators.

T

**12.1 Partial Fraction Expansion
**

The partial fraction expansion method is used extensively in integration and in finding the inverses of the Laplace, Fourier, and Z transforms. This method allows us to decompose a rational polynomial into smaller rational polynomials with simpler denominators, from which we can easily recognize their integrals or inverse transformations. In the subsequent discussion we will discuss the partial fraction expansion method and we will illustrate with several examples. We will also use the MATLAB residue(r,p,k) function which returns the residues (coefficients) r of a partial fraction expansion, the poles p and the direct terms k. There are no direct terms if the highest power of the numerator is less than that of the denominator. Let

(s) ----------F(s) = N D(s)

(12.1)

**where N ( s ) and D ( s ) are polynomials and thus (12.1) can be expressed as
**

bm s + bm – 1 s + bm – 2 s + … + b1 s + b0 N(s) F ( s ) = ----------- = ------------------------------------------------------------------------------------------------------------------n n–1 n–2 D(s) an s + an – 1 s + an – 2 s + … + a1 s + a0

m m–1 m–2

(12.2)

The coefficients a k and b k for k = 0, 1, 2, …, n are real numbers and, for the present discussion, we have assumed that the highest power of N ( s ) is less than the highest power of D ( s ) , i.e., m < n . In this case, F ( s ) is a proper rational function. If m ≥ n , F ( s ) is an improper rational function. It is very convenient to make the coefficient a n of s in (12.2) unity; to do this, we rewrite it as

1 ---- ( bm sm + bm – 1 sm – 1 + bm – 2 sm – 2 + … + b1 s + b0 ) an N(s) F ( s ) = ----------- = -----------------------------------------------------------------------------------------------------------------------------a1 D(s) a0 n an – 1 n – 1 an – 2 n – 2 -s -s - s + ---+ ---------+ … + ---s + ---------an an an an

n

(12.3)

Numerical Analysis Using MATLAB® and Excel®, Third Edition Copyright © Orchard Publications

12−1

…. or repeated.+ … + ----------------F ( s ) = ----------------( s – p1 ) ( s – p2 ) ( s – p3 ) ( s – pn ) (12. p 2. 3s + 2 F 1 ( s ) = ------------------------2 s + 3s + 2 (12.= -------------------------------. r 2.8) 12−2 Numerical Analysis Using MATLAB® and Excel®.6) r k = lim ( s – p k ) F ( s ) = ( s – p k ) F ( s ) s → pk s = pk Example 12.1 Use partial fraction expansion to simplify F 1 ( s ) of (12.7) below. that is. we let s → p k . r 3.Chapter 12 Partial Fraction Expansion The roots of the numerator are called the zeros of F ( s ) . However.= --------------F 1 ( s ) = ------------------------2 (s + 1) (s + 2) (s + 1)(s + 2) s + 3s + 2 3s + 2 r 1 = lim ( s + 1 ) F ( s ) = --------------(s + 2) s → –1 = –1 s = –1 (12. p n of F ( s ) are distinct (different from each another).3) .4) as r2 r3 rn r1 . p 3.3) can be real and distinct.+ ----------------. To evaluate the residue r k . …. we multiply both sides of (12. or combinations of real and complex conjugates. then.+ ----------------. (12. r n are the residues of F ( s ) .4) where p k is distinct from all other poles.+ --------------. and are found by letting N ( s ) = 0 in (12. Third Edition Copyright © Orchard Publications . The zeros and poles of (12. we can factor the denominator of F ( s ) in the form N(s) F ( s ) = ------------------------------------------------------------------------------------------------( s – p1 ) ⋅ ( s – p2 ) ⋅ ( s – p3 ) ⋅ … ⋅ ( s – pn ) (12.5) by ( s – p k ) . or complex conjugates. Case I: Distinct Poles If all the poles p 1. Then. The roots of the denominator are called the poles of F ( s ) and are found by letting D(s) = 0 . We will consider the nature of the poles for each case.7) Solution: r2 r1 3s + 2 3s + 2 . the partial fraction expansion method allows us to express (12. in most engineering applications we are interested in the nature of the poles.5) where r 1.

2]. and it is always empty (has no value) whenever F ( s ) is a proper rational function. [r. The factor(s) function is a good alternative. Third Edition Copyright © Orchard Publications 12−3 . this function is used with symbolic expressions.10) Solution: First. these represent the residues. 3. we obtain 4 3s + 2 .+ --------------F 1 ( s ) = ------------------------2 (s + 1) (s + 2) s + 3s + 2 (12.8).2 Use partial fraction expansion to simplify F 2 ( s ) of (12. p. we will use the MATLAB function factor(s) to express the denominator polynomial of F 2 ( s ) in factored form. Example 12. the direct term k is empty. poles. Numerical Analysis Using MATLAB® and Excel®. Ds) r = 4 -1 p = -2 -1 k = [] where we have denoted Ns and Ds as two vectors that contain the numerator and denominator coefficients of F 1 ( s ) . and k vectors.9) We can us the MATLAB residue(r. Ds = [1. The first value of the vector r is associated with the first value of the vector p. by substitution into (12.= --------------– 1 . 3s 2 + 2s + 5 F 2 ( s ) = -----------------------------------------------s 3 + 12s 2 + 44s + 48 (12. the second value of r is associated with the second value of p. MATLAB displays the r. and so on. we observe that the highest power of the denominator is s 2 whereas the highest power of the numerator is s and therefore. p. k] = residue(Ns. we can use the roots(p) function. These expressions are * Of course.p. 2].* This function returns an expression that contains the prime factors of a polynomial. For this example. However. The vector k is referred to as the direct term.k) function to verify our answers with the following script: Ns = [3.10) below. and direct terms respectively.Partial Fraction Expansion 3s + 2 r 2 = lim ( s + 2 ) F ( s ) = --------------(s + 1) s → –2 = 4 s = –2 Therefore.

z . For the present. Symbolic expressions. we find the residues r 1 . factor(den) ans = (s+4)*(s+2)*(s+6) and thus. y . sin x 2 e –α t d. they are discussed in detail in MATLAB User’s Manual.+ --------------. a1 and k2 .dx ∫ -x 1 MATLAB contains the so-called Symbolic Math Toolbox. t etc. For example. y . we can use the syms function to define one or more symbolic variables with a single statement. The functions. and r 3 . Returning to Example 12. 9⁄8 – 37 ⁄ 4 89 ⁄ 8 3s + 2s + 5 . s . we must create a symbolic variable x . on the other hand.= --------------. syms x y z a1 k2 defines the symbolic variables x . s = sym (‘s’) creates the symbolic variable s . Alternately. they produce numerical results.= --------------. Some examples of symbolic expressions are given below.2 and using MATLAB we have: syms s. This is done with the sym function. For example.+ --------------F 2 ( s ) = -----------------------------------------------2 3 (s + 2) (s + 4) (s + 6) s + 12s + 44s + 48 2 12−4 Numerical Analysis Using MATLAB® and Excel®. which we have used before. This is a collection of tools (functions) which are used in solving symbolic expressions. These are 3s + 2s + 5 r 1 = -------------------------------( s + 4 )( s + 6) 2 s = –2 9 = -8 3s + 2s + 5 r 2 = -------------------------------(s + 2)(s + 6) 2 s = –4 37 =– ----4 3s + 2s + 5 r 3 = -------------------------------(s + 2)(s + 4) 2 s = –6 89 = ----8 Therefore. like roots(p). that is. are display numeric expressions. r 2 .3 2 y = -----( 3t – 4t + 5t + 8 ) 2 dt 2 u = .+ --------------. Before using symbolic expressions. Third Edition Copyright © Orchard Publications . our interest is in using the factor(s) to express the denominator of (12.10) as a product of simple factors. can manipulate mathematical expressions without using actual numbers.Chapter 12 Partial Fraction Expansion explained below.+ --------------F 2 ( s ) = -----------------------------------------------2 3 (s + 2)(s + 4)(s + 6) (s + 2) (s + 4) (s + 6) s + 12s + 44s + 48 Next. den=s^3+12*s^2+44*s+48.= ------------------------------------------------. 2 2 r1 r2 r3 3s + 2s + 5 3s + 2s + 5 .

0000-2. Using the MATLAB script syms s.0000+2. roots_p=roots(p) roots_p = -2.12) and the residues are * For some undocumented reason. s+3 F 3 ( s ) = -----------------------------------------3 s + 5s 2 + 12s + 8 (12.Partial Fraction Expansion Case II: Complex Poles Quite often. The partial fraction expansion method can also be used in this case. we express the denominator in factored form to identify the poles of F 3 ( s ) . Third Edition Copyright © Orchard Publications 12−5 .= -----------------------------------------------------------------------F 3 ( s ) = -----------------------------------------2 3 ( s + 1 ) ( s + 2 + j2 ) ( s + 2 – j2 ) s + 5s + 12s + 8 r1 r2 r3 s+3 . the poles of a proper rational function F ( s ) are complex. p=[1 4 8]. Example 12. Numerical Analysis Using MATLAB® and Excel®. Thus if p k is a complex pole. as illustrated by the following example. we will use the roots(p) function to find its roots by treating it as a polynomial.0000i Then. and since complex poles occur in complex conjugate pairs.= --------------= -----------------------------------------. the number of complex poles is even. factor(s^3 + 5*s^2 + 12*s + 8) we obtain ans = (s+1)*(s^2+4*s+8) Since the factor(s) function did not factor the quadratic term*. the factor(s) function does not seem to work with complex numbers. then its complex conjugate p k∗ is also a pole.11) Solution: As a first step.11) below.3 Use partial fraction expansion to simplify F 3 ( s ) of (12. s+3 s+3 .+ --------------------------+ -----------------------2 3 ( s + 1 ) ( s + 2 + j2 ) ( s + 2 – j 2 ) s + 5s + 12s + 8 (12.0000i -2.

14) Case III: Multiple (Repeated) Poles In this case.+ ---------------------------------(s + 2) ( s + 2 + j2 ) ( s + 2 –j 2 ) (12.13) to form a single term and now is written as 2⁄5 1 ( 2s + 1 ) . the partial fraction expansion of (12.+ j ----( – 8 + j4 ) ( – 8 – j4 ) 80 5 20 s+3 r 3 = -------------------------------------------( s + 1 ) ( s + 2 + j2 ) s = – 2 + j2 1 – j2 1 – j2 .13) We can express (12. r 12. has a multiplicity m .= – 1 = ----------------------.15) can be expressed as 12−6 Numerical Analysis Using MATLAB® and Excel®.– j -----=–1 ( – 8 – j4 ) ( – 8 + j4 ) 5 20 80 Of course. the last evaluation was not necessary since r 3 = r 2∗ or 3 * 3 1 ----1 ----. Third Edition Copyright © Orchard Publications .⋅ -----------------------------F 3 ( s ) = --------------( s + 2 ) 5 ( s 2 + 4s + 8 ) (12. say p 1 .---------------------.12). Then.= ----------------= --------------------------------– 8 – j4 ( – 1 + j2 ) ( j4 ) 3– 8 + j4 ) ( – 16 ) – j 12 ( 1 – j2 ) .+ j -⎞ = – --–j r 3 = ⎛ – -⎝ 5 20⎠ 5 20 and this is always true since complex roots occur in conjugate pairs.( ---------------------= ---------------------= ------------------------. Then. N(s) F ( s ) = ------------------------------------------------------------------------------------------m ( s – p 1 ) ( s – p 2 )… ( s – p n – 1 ) ( s – p n ) (12. … r 1m .15) and denoting the m residues corresponding to multiple pole p 1 as r 11.+ ----------------------------------.Chapter 12 Partial Fraction Expansion s+3 r 1 = ------------------------2 s + 4s + 8 2 = -5 1 – j2 1 – j2 = -----------------------------------. we obtain 2⁄5 – 1 ⁄ 5 + j3 ⁄ 20 – 1 ⁄ 5 – j3 ⁄ 20 F 3 ( s ) = --------------.= -----------------( – 1 – j2 ) ( – j4 ) – 8 + j4 s = –1 s+3 r 2 = ----------------------------------------( s + 1 ) ( s + 2 –j 2 ) s = – 2 – j2 3 ( 1 – j2 ) ( – 8 – j4 ) – 16 + j12 . F ( s ) has simple poles but one of the poles.13) in a different form if we want to eliminate the complex presentation.– -.= -----------------------. This is done by combining the last two terms on the right side of (12. by substitution into (12.

we first differentiate the relation of (12.16) by ( s – p 1 ) .18) twice with respect to s .[ ( s – p1 ) F ( s ) ] s → p 1 ds (12. we multiply both sides of (12. To find the second residue r 12 of the second repeated pole p 1 . then.19) yields the residue of the first repeated pole. that is.+ … + ----------------⎝(s – p ) (s – p ) ( s – p n )⎠ 2 3 (12.20) To find the third residue r 13 of the repeated pole p 1 . Then. r k = lim ( s – p k ) F ( s ) = ( s – p k ) F ( s ) s → pk s = pk m To find the first residue r 11 of the repeated pole.[ ( s – p1 ) F ( s ) ] r 13 = lim ------s → p 1 ds 2 2 (12.Partial Fraction Expansion r 11 r 12 r 13 r 1m F ( s ) = --------------------. we obtain s → p1 lim ( s – p 1 ) F ( s ) 2 m–1 s → p1 m = r 11 + lim [ ( s – p 1 ) r 12 + ( s – p 1 ) r 13 + … + ( s – p 1 ) r 1m ] (12. we differentiate (12.+ ----------------.17).19) and thus (12.+ ---------------------------+ ---------------------------+ … + ----------------m m–1 m–2 ( s – p1 ) ( s – p1 ) ( s – p1 ) ( s – p1 ) r3 rn r2 + ----------------.21) Numerical Analysis Using MATLAB® and Excel®.+ ----------------( s – p2 ) ( s – p3 ) ( s – pn ) (12. taking the limit as s → p 1 on both sides of (12.18) r3 rn ⎞ r2 m . Third Edition Copyright © Orchard Publications 12−7 . we let s → p 1 . ( s – p 1 ) F ( s ) = r 11 + ( s – p 1 ) r 12 + ( s – p 1 ) r 13 + … + ( s – p 1 ) m 2 m–1 r 1m r2 r3 rn ⎞ m + ( s – p 1 ) ⎛ ----------------.16) For the simple poles p 1. we let s → p 1 . then.+ … + ----------------+ lim ( s – p 1 ) ⎛ ----------------⎝ ( s – p2 ) ( s – p3 ) ( s – p n )⎠ s → p1 or r 11 = lim ( s – p 1 ) F ( s ) s → p1 m (12. that is. p 2. m d .+ ----------------. … p n we proceed as before.+ ----------------.18) with respect to s . that is.17) Next. m d r 12 = lim ---.

Third Edition Copyright © Orchard Publications .= --------------.26).+ -----------------.+ --------------2 2 ( s + 2 ) ( s + 1) (s + 2)(s + 1) (s + 1) 12−8 Numerical Analysis Using MATLAB® and Excel®.⎛ s ----------⎞ r 22 = ---ds ⎝ s + 2 ⎠ s = –1 (s + 2) – (s + 3) = -------------------------------------2 (s + 2) = –1 s = –1 Then.25) in partial fraction expansion form is r 21 r1 r 22 s+3 = --------------. (12.+ -----------------+ --------------F 4 ( s ) = ----------------------------------2 2 (s + 2) (s + 1) (s + 1) (s + 2)(s + 1) (12.Chapter 12 Partial Fraction Expansion This process is continued until all residues of the repeated poles have been found.4 Use partial fraction expansion to simplify F 4 ( s ) of (12.24) Example 12.25) below. In general. for repeated poles the residue r 1k can be derived from the relation ( s – p 1 ) F ( s ) = r 11 + r 12 ( s – p 1 ) + r 13 ( s – p 1 ) + … m 2 (12.------------[ ( s – p1 ) F ( s ) ] s → p 1( k – 1 )! ds k – 1 k–1 (12.22) whose ( m – 1 ) th derivative of both sides is d .25) Solution: We observe that there is a pole of multiplicity 2 at s = – 1 and thus.[ ( s – p1 )m F ( s ) ] ( k – 1 )! r 1k = lim ------------k – 1 s → p 1 ds k–1 (12. s+3 1 2 –1 F 4 ( s ) = ----------------------------------. by substitution into (12.23) or d m 1 r 1k = lim -----------------. s+3 F 4 ( s ) = ----------------------------------2 (s + 2)(s + 1) (12.26) The residues are s+3 r 1 = -----------------2 (s + 1) s + 3r 21 = --------------(s + 2) = 1 s = –2 = 2 s = –1 +3 d.

0000 k = [] Example 12. Like the factor(s) function. we will use the expand(s) function. % Multiplies polynomials d1 and d2 to express denominator D(s) as polynomial [r. and as such. it is true for any value of s. *We must remember that (2.0000 2. Numerical Analysis Using MATLAB® and Excel®. s+3 ----------------------------------2 (s + 1) (s + 2) 1 = --------------(s + 2) 2 + -----------------2 (s + 1) s=0 r 22 + --------------(s + 1) s=0 s=0 s=0 or 3 ⁄ 2 = 1 ⁄ 2 + 2 + r 22 from which r 22 = – 1 as before. Check with MATLAB: syms s expand((s + 1)^2) % Create symbolic variable s % Express it as a polynomial ans = s^2+2*s+1 Ns = [1 3].0000 p = -2. expand(s) is used with symbolic expressions.45) is an identity.p.0000 -1.0000 -1. the residue r 22 could be found by substitution of the already known values of r 1 and r 21 into (12.5 Use partial fraction expansion to simplify F 5 ( s ) of (12.Partial Fraction Expansion Instead of differentiation. Its description can be displayed with the help expand command. % Coefficients of (s + 2) term in D(s) Ds=conv(d1. To check our answers with MATLAB. Third Edition Copyright © Orchard Publications 12−9 . % Coefficients of the numerator N(s) d1 = [1 2 1]. % Coefficients of (s + 1)^2 = s^2 + 2*s + 1 term in D(s) d2 = [0 1 2]. and letting s = 0 *.26).0000 -1.k]=residue(Ns.d2).27) below. that is.Ds) r = 1.

---. the residue r 21 if found by evaluating F 5 ( s ) at s = – 2 . Third Edition Copyright © Orchard Publications .⎛ -------------------------⎟ r 12 = ---⎜ 2 ds ⎝ ( s + 2 ) ⎠ 2 s = –1 2 (12.30) s = –1 ( s + 2 ) ( 2s + 3 ) – 2 ( s + 2 ) ( s + 3 s + 1 ) = --------------------------------------------------------------------------------------------4 (s + 2) s+4 = -----------------.------. Therefore. Then.= 3 3 (s + 2) The residue r 13 is found by taking the second derivative of F 5 ( s ) and evaluating it at s = – 1 . in partial fraction expansion form r 21 r 11 r 12 r 13 r 22 + -----------------+ --------------.( --------------------------------------------------------------= -6 2 (s + 2) –s–5 = -----------------4 (s + 2) = –4 s = –1 (12. Then. and the residue r 22 is found by first taking the first derivative of F 5 ( s ) and evaluating it at s = – 2 . 2 ⎞ s + 3s + 1 d.+ -----------------+ --------------F 5 ( s ) = -----------------3 2 ( s + 1 ) ( s + 2 )2 ( s + 2 ) (s + 1) (s + 1) (12.⎜ -------------------------⎟ r 13 = ---2! ds 2 ⎝ ( s + 2 ) 2 ⎠ 2 1 d d ⎛ s + 3 s + 1⎞ .31) 1 ⎛ s + 2 – 3s – 12 ⎞ . Thus.28) We find the residue r 11 by evaluating F 5 ( s ) at as s = – 1 + 3s + 1 r 11 = s ------------------------2 (s + 2) 2 = –1 s = –1 (12.---------------------------------= -4 ⎠ 2⎝ (s + 2) s = –1 Similarly.29) The residue r 12 is found by first taking the first derivative of F 5 ( s ) . and evaluating it at s = – 1 .⎛ -----------------1 . 2 2 1 d ⎛ s + 3 s + 1⎞ . 12−10 Numerical Analysis Using MATLAB® and Excel®.Chapter 12 Partial Fraction Expansion s2 + 3 s + 1 F 5 ( s ) = ------------------------------------( s + 1 )3 ( s + 2 )2 (12.⎜ -------------------------⎟ = -2 ds ds ⎝ ( s + 2 ) 2 ⎠ 3 2 s = –1 s = –1 s+4 ⎞ d.27) Solution: We observe that there is a pole of multiplicity 3 at s = – 1 .---= -2 ds ⎝ ( s + 2 ) 3 ⎠ s = –1 s + 2) – 3(s + 2) (s + 4) 1 .---. and a pole of multiplicity 2 at s = – 2 .

0000 3. Ds=[1 7 19 25 16 4]. syms s.+ -----------------.+ -----------------. i.0000 Numerical Analysis Using MATLAB® and Excel®..⎜ -------------------------⎟ r 22 = ---ds ⎝ ( s + 1 ) 3 ⎠ 3 s = –2 2 2 s + 1 ) ( 2s + 3 ) – 3 ( s + 1 ) ( s + 3 s + 1 ) = ( -------------------------------------------------------------------------------------------------6 (s + 1) ( s + 1 ) ( 2s + 3 ) – 3 ( s + 3 s + 1 ) r 22 = ---------------------------------------------------------------------------4 (s + 1) 2 s = –2 2 s = –2 – s – 4s = -------------------4 (s + 1) = 4 s = –2 By substitution of these residues into (12. The MATLAB script for this example is as follows. polynomial coefficients only.28).q) function is used with numeric expressions.0000 -1.0000 1.+ --------------F 5 ( s ) = -----------------. % We must first define the variable s in symbolic form % The function "collect" below multiplies (s+1)^3 by (s+2)^2 Ds=collect(((s+1)^3)*((s+2)^2)) Ds = s^5+7*s^4+19*s^3+25*s^2+16*s+4 % We now use this result to express the denominator D(s) as a % polynomial so we can use its coefficients with the "residue" function % Ns=[1 3 1].+ --------------3 2 2 ( s + 1 ) ( s + 2) (s + 1) (s + 1) (s + 2) (12. Third Edition Copyright © Orchard Publications 12−11 .Ds) r = 4.k]=residue(Ns.p. Its description can be displayed with the help collect command.e. [r. we obtain F 5 ( s ) in partial fraction expansion as –1 3 –4 1 4 .32) We will now verify the values of these residues with MATLAB.0000 -4. We must remember that the conv(p. we introduce the collect(s) function that we can use to multiply two or more symbolic expressions to obtain the result in a polynomial form.Partial Fraction Expansion s + 3s + 1 r 21 = ------------------------3 (s + 1) 2 = 1 s = –2 2 d ⎛ s + 3 s + 1⎞ . Before we do this.

Ds = [1 1]. i. m < n . and before we apply the partial fraction expansion. Ds) r = 1 p = -1 k = 1 1 Numerical Analysis Using MATLAB® and Excel®.34) below in partial expansion form. that is.0000 k = [] Case for m ≥ n Our discussion thus far. p. s + 2s + 2 F 6 ( s ) = ------------------------s+1 2 (12. k] = residue(Ns.0000 -1. F ( s ) is an improper rational function. Example 12. m > n and thus we need to express F 6 ( s ) in the form of (12.33) so that m < n . the highest power m of the numerator is less than the highest power n of the denominator. Third Edition Copyright © Orchard Publications 12−12 ..34).6 Express F 6 ( s ) of (12.34) Solution: In (12.33).Chapter 12 Partial Fraction Expansion p = -2.0000 -1. If m ≥ n . we must divide the numeraror N ( s ) by the denominator D ( s ) to obtain an expression of the form F ( s ) = k0 + k1 s + k2 s + … + km – n s 2 m–n N(s) + ----------D(s) (12. 1 s + 2s + 2 . By long division. was based on the condition that F ( s ) is a proper rational function.0000 -1.= ----------+s+1 F 6 ( s ) = ------------------------s+1 s+1 2 (12. [r.e.35) Check with MATLAB: Ns = [1 2 2].0000 -2.

we let s – a be a linear factor of D ( s ) and we suppose that ( s – a ) is the highest power of s – a that divides D ( s ) . we can express F ( s ) as rm r1 r2 ( s ) = -------------------. we assign the sum of n partial fractions as shown below.+ … -----------------F(s) = N 2 D(s) s – a (s – a) ( s – a )m n m (12. Then. 12. we perform the following steps: 1.+ ----------------. We arrange the terms of both sides in decreasing powers of s . r2 s + k2 r1 s + k1 rn s + kn -------------------------. 5. We also assume that the denominator D ( s ) can be expressed as a product of real linear and quadratic factors. let s 2 + α s + β be a quadratic factor of D ( s ) and suppose that ( s 2 + α s + β ) is the highest power of this factor that divides F ( s ) . We multiply each term of the right side by the appropriate factor to make the denominators of (12. 6.38) below as a sum of partial fractions using the equating the numerators procedure. we first perform a long division and then work with the quotient and the remainder as before.+ … + --------------------------------2 2 n 2 s + α s + β ( s2 + α s + β ) (s + αs + β) 2. We repeat Step 1 for each of the distinct linear and quadratic factors of D ( s ) . To this factor. We set the given F ( s ) equal to the sum of these partial fractions.36) Next.+ --------------------------------.54). Now. We equate the coefficients of corresponding powers of s. 4. Third Edition Copyright © Orchard Publications 12−13 . Numerical Analysis Using MATLAB® and Excel®. If not.37) both sides equal. We solve the resulting equations for the residues.7 Express F 7 ( s ) of (12.Alternate Method of Partial Fraction Expansion The direct terms k = [ 1 1 ] are the coefficients of the s term and the constant in (2. If these assumptions prevail.2 Alternate Method of Partial Fraction Expansion The partial fraction expansion method can also be performed by the equating the numerators procedure thereby making the denominators of both sides the same. 7. and then equating the numerators. 3. We assume that the degree on the numerator N ( s ) is less than the degree of the denominator. Example 12.

6. we obtain 0 = r 1 + r 22 0 = – 2r 1 + A – r 22 + r 21 – 2 = r 1 – 2A + r 22 4 = A – r 22 + r 21 (12. by equating like powers of s . Third Edition Copyright © Orchard Publications . therefore.42. we obtain 4 = 2r 1 or r 1 = 2 (12.38) Solution: By Steps 1 through 3 above.39) By Step 4.42) Subtracting the second equation from the fourth in (12.41) Relation (12.– --------------– 2s + 4 2s + 1 + ----------------2 = -----------------F 7 ( s ) = ------------------------------------2 2 2 2 (s – 1) (s + 1)(s – 1) (s + 1) (s – 1) 12−14 Numerical Analysis Using MATLAB® and Excel®. and 7. r 21 r1 s + A r 22 – 2s + 4 F 7 ( s ) = ------------------------------------= -----------------+ ----------------.46) Therefore F 7 ( s ) in partial fraction expansion form becomes 1 .43) and by substitution into the first equation of (12. – 2s + 4 = ( r 1 s + A ) ( s – 1 ) 2 + r 21 ( s + 1 ) + r 22 ( s – 1 ) ( s + 1 ) 2 2 (12.+ --------------2 2 2 2 (s – 1) (s + 1)(s – 1) (s + 1) (s – 1) (12. we obtain 0 = 2 + r 22 or r 22 = – 2 (12.44) Next. we obtain: 4 = 1 + 2 + r 21 or r 21 = 1 (12.40) and by Steps 5.44) into the third equation of (12.Chapter 12 Partial Fraction Expansion – 2s + 4 F 7 ( s ) = ------------------------------------2 2 (s + 1)(s – 1) (12. the coefficients of each power of s on the left and right sides are equal.42).43) and (12. substitution of (12. Accordingly.41) is an identity in s .45) and using the fourth equation of (12.42). yields – 2 = 2 – 2A – 2 or A = 1 (12. – 2s + 4 = ( r 1 + r 22 ) s 3 + ( – 2r 1 + A – r 22 + r 21 ) s 2 + ( r 1 – 2A + r 22 ) s + ( A – r 22 + r 21 ) (12.42).

that is. we obtain 0 = r1 + r2 (12.+ ------------------------F 7 ( s ) = ----------------------------------------------. Third Edition Copyright © Orchard Publications 12−15 . where we found that the denominator can be expressed in factored form of a linear and a quadratic factor.k) function. r2 s + r3 r1 s+3 . we only need two equations in r 2 and r 3 . s+3 F 8 ( s ) = -----------------------------------------2 3 s + 5s + 12s + 8 (12.Alternate Method of Partial Fraction Expansion Example 12. we equate the constant terms of (12. We will conclude the partial fraction expansion topic with a few more examples. Then. we use the equating the numerators procedure and we obtain ( s + 3 ) = r 1 ( s + 4s + 8 ) + ( r 2 s + r 3 ) ( s + 1 ) 2 (12.49) As in Example 12. with the coefficients of s 2 on the right side of (12. using the residue(r.48) and in partial fraction expansion form.8 Use the equating the numerators procedure to obtain the partial fraction expansion of F 8 ( s ) in (12. Equating the coefficient of s 2 on the left side. we obtain 2 ⁄ 5 1 ( 2s + 1 ) . which is zero.51). (12.50) s = –1 To compute r 2 and r 3 .p.51) Since r 1 is already known. To find the third residue r 3 . we obtain r 3 = – 1 ⁄ 5 . 3 = 8r 1 + r 3 . and with r 1 = 2 ⁄ 5 .47) Solution: This is the same rational function as that of Example 12.3. The remaining steps are the same as in Example 12. that is.52) With r 1 = 2 ⁄ 5 . Numerical Analysis Using MATLAB® and Excel®.3.= ---------2 2 s + 1 s + 4s + 8 ( s + 1 ) ( s + 4s + 8 ) (12.52) yields r 2 = – 2 ⁄ 5 .47) below.51). s+3 F 7 ( s ) = ----------------------------------------------2 ( s + 1 ) ( s + 4s + 8 ) (12.3. we first find the residue of the linear factor as s+3 r 1 = ------------------------2 s + 4s + 8 2 = -5 (12.49). by substitution into (12.-----------------------------F 7 ( s ) = --------------( s + 2 ) 5 ( s 2 + 4s + 8 ) as before.– -.

F 9 ( s ) in partial fraction expansion form is written as r2 r1 14 –6 .56) below. F 9 ( s ) can be written as r2 r1 . it can be written as 12−16 Numerical Analysis Using MATLAB® and Excel®. but recommended den=[1 3 2]. [r. r 2 .k) function to compute the poles and residues of F 10 ( s ) in (12. Third Edition Copyright © Orchard Publications .p.54) The MATLAB script for this example is as follows: num=[0 8 2].p.Chapter 12 Partial Fraction Expansion Example 12.+ ---------.k) function to compute the poles and residues of the function 8s + 2 F 9 ( s ) = ------------------------2 s + 3s + 2 (12.56) Solution: Let p 1 . s+3 F 10 ( s ) = ----------------------------------------------2 ( s + 1 ) ( s + 4s + 8 ) (12. % The semicolon suppress the display of the row vector typed % and zero is typed to make the numerator have same number % of elements as the denominator.55) Example 12. not necessary.53) Solution: Let p 1 and p 2 be the poles (the denominator roots) and r 1 and r 2 be the residues.9 Use the residue(r.= ---------F 9 ( s ) = ------------s + p1 s + p2 s+2 s+1 (12. and p 3 be the poles (the denominator roots) and r 1 . Then.den) r = 14 -6 p = -2 -1 k = [] Therefore. Then.p.k]=residue(num.10 Use the residue(r.+ ------------F 9 ( s ) = ------------s + p1 s + p2 (12.+ ------------.and r 3 be the residues of F 10 ( s ) . p 2 .

[r. we need to express the denominator as a polynomial.2 – 0.+ -----------------------------.15j ----------------------------. we can reduce a rational polynomial to simple terms of a polynomial. den).4000 p = -2.15j + ---------0. x + 2x + 1 f 1 ( x ) = ----------------------------0.0000 k = [] 5 12 8 0. c. We will use the function conv(a.+ ------------.5x – 1 3 2 (12.den) function. where the last term is a rational polynomial whose order of the numerator is less than that of the denominator as illustrated by the following example. We recall that we can write two or more statements on one line if we separate them by commas or semicolons. a=[1 1].11 Use the deconv(num.k]=residue(num. b=[1 4 8]. num=[1.2 + 0.0000i Therefore.b) to multiply the two factors of the denominator of (12.2000-0.3].p.+ ------------.= – F 10 ( s ) = ------------s+1 s + 2 + 2j s + 2 – 2j s + p1 s + p2 s + p3 (12.56).2000+ 0.den) c = 1 r = -0.1500i 0.k]=residue(num.58) By repeated use of the deconv(num.0000+ -2. c=conv(a.0000-1.+ ------------.59) Numerical Analysis Using MATLAB® and Excel®.b).4– 0. whereas semicolons will suppress the display. Then. Example 12.den) function to express the following rational polynomial as a polynomial with four terms.Alternate Method of Partial Fraction Expansion r1 r2 r3 F 10 ( s ) = ------------.+ ------------s + p1 s + p2 s + p3 (12. We also recall that commas will display the results.p.0000i 2. Before we use this statement.57) The poles and the residues can be found with the statement [r.1500i 2. den=c. Third Edition Copyright © Orchard Publications 12−17 . F 10 ( s ) in partial fraction expansion form is r2 r3 r1 0.

60) It is important to remember that the function roots(p) is used with polynomials only. This is discussed in Chapter 1. 12−18 Numerical Analysis Using MATLAB® and Excel®. f 1 ( x ) can now be written as (12.r]=deconv(num. and x0 is a required initial value.5x + 6. We can approximate this value by first plotting f 2 ( x ) to find out where it crosses the x −axis.5x ) 3 2 5 2 (12.+ --------------------------------------------------. den=[0 0 0.11x + 2.5x – 1 8 16 Therefore. Page 1−27. Third Edition Copyright © Orchard Publications .35 1 f 2 ( x ) = ------------------------------------------------------------.3x + 4.Chapter 12 Partial Fraction Expansion Solution: num=[1 2 0 1].61) we must use the function fzero(‘function’. [q.+ ----------------------------6 4 2 6 3 4.5 −1]. If we want to find the zeros of any function.den) q = 2 r = 0 0 0 17 2 17 f 1 ( x ) = 2x + 8x + 16 + ------------------0.75 ( 12x + 2x + 13x + 25 ) ( 23x + 16x + 7.x0). such as the function f 2 ( x ) defined as 3x + 7x + 9 0. where function is a pre−defined string.

3 Summary • The function + bm – 2 s + … + b1 s + b0 bm s + bm – 1 s N(s) F ( s ) = ----------. 1. • If all the poles p 1.s + ---s + ---------+ ---------+ … + ---an an an an to make a n unity. n are real numbers. ….= ------------------------------------------------------------------------------------------------------------------n n – 1 n – 2 D(s) + an – 2 s + … + a1 s + a0 an s + an – 1 s m m–1 m–2 is a proper rational function where a n is a non−zero integer other than unity.. In most engineering applications we are interested in the nature of the poles. p 2. p n of F ( s ) are distinct we can factor the denominator of F ( s ) in the form Numerical Analysis Using MATLAB® and Excel®. or combinations of real and complex conjugates. p 3.= ------------------------------------------------------------------------------------------------------------------n n – 1 n – 2 D(s) + an – 2 s + … + a1 s + a0 an s + an – 1 s m m–1 m–2 where the coefficients a k and b k for k = 0. is a proper rational function if the highest power of the numerator N ( s ) is less than the highest power of of the denominator D ( s ) . or complex conjugates. If F ( s ) is an improper rational function we divide the numeraror N ( s ) by the denominator D ( s ) to obtain an expression of the form F ( s ) = k0 + k1 s + k2 s + … + km – n s 2 m–n (s) +N ----------D(s) so that m < n . m < n .e. i. …. we rewrite this function as 1m m–1 m–2 ---( bm s + bm – 1 s + bm – 2 s + … + b1 s + b0 ) a N(s) n F ( s ) = ----------. • Partial fraction expansion applies only to proper rational functions. F ( s ) is an improper rational function. and the roots of the denominator are called the poles of F ( s ) and are found by letting D(s) = 0 . • If the function bm s + bm – 1 s + bm – 2 s + … + b1 s + b0 N(s) F ( s ) = ----------. and are found by letting N ( s ) = 0 . Third Edition Copyright © Orchard Publications 12−19 . • The roots of the numerator are called the zeros of F ( s ) . or repeated. • The zeros and poles can be real and distinct. 2.Summary 12. If m ≥ n .= -----------------------------------------------------------------------------------------------------------------------------a1 D(s) a0 n an – 1 n – 1 an – 2 n – 2 -s -s .

+ ----------------. Since complex poles occur in conjugate pairs. the function is expressed as N(s) F ( s ) = ------------------------------------------------------------------------------------------m ( s – p 1 ) ( s – p 2 )… ( s – p n – 1 ) ( s – p n ) and denoting the m residues corresponding to multiple pole p 1 as r 11. that is. the partial fraction expansion can be expressed as r 12 r 13 r 1m r 11 + ---------------------------+ … + ----------------. r 2. r 3.p. if p k is a complex pole. • If a rational function F ( s ) has simple poles but one of the poles. r k = lim ( s – p k ) F ( s ) = ( s – p k ) F ( s ) s → pk s = pk • We can use the MATLAB residue(r. and a direct term. r 12. Then. …. • The partial fraction expansion can also be used if the poles are complex. For proper rational functions there is no direct term. r n are the residues of F ( s ) . To evaluate the residue r k . say p 1 . then. The first residue of a repeated pole is found from r 11 = lim ( s – p 1 ) F ( s ) s → p1 m The second repeated pole is found from 12−20 Numerical Analysis Using MATLAB® and Excel®. we let s → p k . we multiply both sides of (12. say p 1 .+ ---------------------------F ( s ) = --------------------m m–1 m–2 ( s – p1 ) ( s – p1 ) ( s – p1 ) ( s – p1 ) r3 rn r2 + ----------------.k) function to verify our answers.+ … + ----------------F ( s ) = ----------------( s – p1 ) ( s – p 2 ) ( s – p3 ) ( s – pn ) where r 1. for the simple poles we use the same procedure as for distinct poles. the partial fraction expansion method allows us to write the above expression as r2 r3 rn r1 . This function returns the residues. their associated poles.+ ----------------( s – p2 ) ( s – p3 ) ( s – pn ) • If a rational function F ( s ) has simple poles but one of the poles.+ ----------------.Chapter 12 Partial Fraction Expansion N(s) F ( s ) = ------------------------------------------------------------------------------------------------( s – p1 ) ⋅ ( s – p 2 ) ⋅ ( s – p 3 ) ⋅ … ⋅ ( s – p n ) where p k is distinct from all other poles. Third Edition Copyright © Orchard Publications . has a multiplicity m . … r 1m .5) by ( s – p k ) .+ ----------------. has a multiplicity m . then its complex conjugate p k∗ is also a pole.

Numerical Analysis Using MATLAB® and Excel®.Summary d. • With the alternate method of partial fraction expansion we use the equating the numerators procedure thereby making the denominators of both sides the same.[ ( s – p ) m F ( s ) ] r 12 = lim ---1 s → p 1 ds the third from m d .[ ( s – p1 ) F ( s ) ] r 13 = lim ------s → p 1 ds 2 2 and this process is continued until all residues of the repeated poles have been found. Third Edition Copyright © Orchard Publications 12−21 . and then equating the numerators. We assume that the denominator D ( s ) can be expressed as a product of real linear and quadratic factors.

----------------------------------2 1 8. ----------------------------2 s(s + s + 1) (s + 1)(s + 1 ) 12−22 Numerical Analysis Using MATLAB® and Excel®. ---------2 1 –s 1 2. ------------------------2 s + 2s + 1 s(s + 1) 1 6.4 Exercises Perform partial fraction expansion for the following. 11. Third Edition Copyright © Orchard Publications . Use MATLAB to simplify and to verify your results. -------------------2 1 7.Chapter 12 Partial Fraction Expansion 12. ------------------------2 s + 4s – 5 s 3. ---------------------2 s –2 s – 3 2 s 5. ---------------------2 s –2 s – 3 5s – 3 4.

Ds) r = 0.+ ---------2 2 (s + 1)(s – 1) s+1 s–1 s –1 1 –s –1 r 1 = ---------s+1 = –1 ⁄ 2 s=1 –1 r 2 = ---------s–1 = 1⁄2 s = –1 Then.= ---------2 s–1 s+5 (s – 1)(s + 5) s + 4s – 5 1r 1 = ---------s+5 = 1⁄6 s=1 1r 2 = ---------s–1 = –1 ⁄ 6 s = –5 Then. k] = residue(Ns. 0.Solutions to End−of−Chapter Exercises 12. Ds = [1. −5].+ ---------2 s +1 s–1 s –1 Ns = [0.= ------------------.5000 p = -1 1 k = [] 2. Third Edition Copyright © Orchard Publications 12−23 . r1 r2 – 1 . k] = residue(Ns. Ds = [1. –1 1⁄2 1⁄2 -----------. Ds) r = -1/6 1/6 p = -5 1 k = [] Numerical Analysis Using MATLAB® and Excel®. 0. [r. p. −1].= -------------------------------–1 1 .= – ---------. [r. Ns = [0.= -----------.– ---------------------------------2 s–1 s+5 s + 4s – 5 format rat. −1]. 4.= -------------------------------. p.5000 -0. 0.+ ---------------------------------.5 Solutions to End−of−Chapter Exercises 1. 1 ⁄ 61⁄6 1 = ---------. 1]. r2 r1 1 1 .

= ---------. −3]. Ds) r = 3/4 1/4 p = 3 -1 k = [] 4. r1 r2 s s . Ns = [0.+ ------------------------------. −2. 32 . k] = residue(Ns. −2. Ds = [1. r2 r1 5s – 3 5s – 3 .+ ---------5s – 3 . [r.= ------------------------------2 s+1 s–3 s –2 s – 3 Ns = [0. p. −3]. Ds) r = 3 2 p = 3 -1 k = [] 12−24 Numerical Analysis Using MATLAB® and Excel®. 1.= ---------. 0].+ ---------2 s+1 s–3 s –2 s – 3 format rat. Ds = [1.= -------------------------------.+ ------------------------------2 ( s + 1 )( s – 3) s+1 s–3 s –2 s – 3 s r 1 = ---------s–3 = 1⁄4 s = –1 s r 2 = ---------s+1 = 3⁄4 s=3 Then.Chapter 12 Partial Fraction Expansion 3. k] = residue(Ns. [r. 5.= -------------------------------. s 1⁄4 3⁄4 ---------------------.= ---------2 s+1 s–3 ( s + 1 )( s – 3) s –2 s – 3 5s – 3 r 1 = ------------s–3 = 2 s = –1 5s – 3 r 2 = ------------s+1 = 3 s=3 Then. p. Third Edition Copyright © Orchard Publications . −3].

Solutions to End−of−Chapter Exercises 5.den) q = 1 r = 0 and thus 2 r1 r2 s – 2 s – 1 .den) function to express the following rational polynomial as a polynomial with four terms. 0]. 0. [r. and before we apply the partial fraction expansion. r 21 r1 r 22 1 .r]=deconv(num. p. Ds) 2 r = -2 1 p = -1 -1 k = 1 6. num=[1 0 0 ].= --2 2 s ( s + 1) s(s + 1) (s + 1) Numerical Analysis Using MATLAB® and Excel®. [q.= 1 + ------------------------. 1]. This is an improper rational function.= 1 + -----------------. Ds = [1. k] = residue(Ns.= -----------------2 2 (s + 1) (s + 1) (s + 1) r2 = –2 –2 s – 1 = r1 + r2 ( s + 1 ) r1 = 1 r1 + r2 = –1 Then.= 1 + ------------------------+ --------------2 2 2 2 (s + 1) s + 2s + 1 (s + 1) s + 2s + 1 (s + 1) -2 -1 r1 r2 –2 s – 1 + -------------------------------. we must divide the numeraror N ( s ) by the denominator D ( s ) to obtain an expression of the form F ( s ) = k0 + k1 s + k2 s + … + km – n s 2 m–n (s) +N ----------D(s) We could perform long division but we will use the MATLAB deconv(num.+ --------------2 2 2 (s + 1) s + 2s + 1 s + 2s + 1 (s + 1) Ns = [1. Third Edition Copyright © Orchard Publications 12−25 . s –2 –2 s – 1 1 ------------------------.+ -----------------+ ---------------------------------.= 1 – -----------------2s + 1 = 1 – -----------------------------------------. 2. den=[1 2 1].

1].+ ----------------2 2 2 s + 1 s2 + 1 (s + 1)(s + 1) (s + 1 )(s + 1) (s + 1)(s + 1 ) 2 Equating numerators and like terms we obtain 1 = r1 s + r1 + r2 s + r2 s + r3 s + r3 r1 + r2 = 0 syms r1 r2 r3 eq1=r1+r2−0 eq2=r2+r3−0 eq3=r1+r3−1 S=solve(eq1. 1. p. 0. 1 1 –1 –1 -------------------. r1 r1 ( s + 1 ) r2 s + r3 ( r2 s + r3 ) ( s + 1 ) 1 ----------------------------------= ---------= ----------------------------------+ -------------------------------------. Third Edition Copyright © Orchard Publications .= -.⎛ 1 --⎞ r 22 = ---ds ⎝ s ⎠ s = –1 1= – --2 s = –1 s = –1 Then. expand(s*(s+1)^2) ans = s^3+2*s^2+s Ns = [0. [r. k] = residue(Ns.+ -----------------. 0]. 2. Ds = [1.Chapter 12 Partial Fraction Expansion 1r 1 = ---------s+1 = 1 s=0 -r 21 = 1 s = –1 s = –1 d. 0. Ds) r = -1 -1 1 p = -1 -1 0 k = [] 7.+ --------------2 2 s ( s + 1) s(s + 1) (s + 1) syms s. eq3) r2 + r3 = 0 r1 + r3 = 1 2 2 eq1 = r1+r2 eq2 = 12−26 Numerical Analysis Using MATLAB® and Excel®. eq2.

eq3.r2 ans = -1/2 S. p. .= ---------. 1]. 1].1i k = [] Numerical Analysis Using MATLAB® and Excel®. 1.Solutions to End−of−Chapter Exercises r2+r3 eq3 = r1+r3-1 S = r1: [1x1 sym] r2: [1x1 sym] r3: [1x1 sym] S. 1. k] = residue(Ns. Thus. Then. 1 1 ⁄ 2 ( –1 ⁄ 2 ) s + 1 ⁄ 2 ----------------------------------. 0.. r 2 = – 1 ⁄ 2 .r1 ans = 1/2 S. and r 3 = 1 ⁄ 2 .r3 ans = 1/2 The statement S=solve(eq1.eqN) returns the solutions in the structure S whose named fields hold hold the solution for each variable. Ds = [1. r 1 = 1 ⁄ 2 . Third Edition Copyright © Orchard Publications 12−27 . eq2.1/4i -1/4 + 1/4i p = -1 -1/6004799503160662 + 1i -1/6004799503160662 . 0.. expand((s+1)*(s^2+1)) ans = s^3+s+s^2+1 Ns = [0. [r. Ds) r = 1/2 -1/4 .+ -----------------------------------2 2 s +1 (s + 1 )(s + 1 ) s +1 syms s.

0].= --= ------------------------------. Ds) r = -1/2 -1/2 1 p = -1/2 + 1170/1351i -1/2 . p. 0. r 1 = 1 . [r. The MATLAB help residue command displays the following: Warning: Numerically. Then. 1.Chapter 12 Partial Fraction Expansion These values are inconsistent with those we’ve found. can make arbitrarily large changes in the resulting poles and residues.+ ----------------------------2 2 2 2 s s +s+1 s(s + s + 1) s(s + s + 1) s(s + s + 1) 2 Equating numerators and like terms we obtain 1 = r1 s + r1 s + r1 + r2 s + r3 s r1 + r2 = 0 r2 + r3 = 0 r1 = 1 2 2 By inspection. 1]. If the denominator polynomial. then small changes in the data.1170/1351i 0 k = [] As in Exercise 7. 1. is near a polynomial with multiple roots. + 390/1351i 390/1351i 12−28 Numerical Analysis Using MATLAB® and Excel®. A(s). 8. k] = residue(Ns. these values are inconsistent with those we’ve found.+ -------------------------------------------------. ( r2 s + r3 ) s r2 s + r3 r1 r1 ( s + s + 1 ) 1 . r 2 = – 1 . and r 3 = 1 . Third Edition Copyright © Orchard Publications . Problem formulations making use of state-space or zero-pole representations are preferable.+ -------------------------------------------------------= 1 2 2 s s +s+1 (s + 1 )(s + 1 ) syms s. Ds = [1. including roundoff errors. –s+3 1 -. the partial fraction expansion of a ratio of polynomials represents an ill-posed problem. expand(s*(s^2+s+1)) ans = s^3+s^2+s Ns = [0. 0.

Chapter 13 The Gamma and Beta Functions and Distributions T his chapter is an introduction to the gamma and beta functions and their distributions used with many applications in science and engineering. ∫ f ( x ) dx where the limits of integration a or b or both are infinite b.2) We will evaluate the integral of (13. Third Edition Copyright © Orchard Publications 13−1 .3). and in the computation of certain integrals.1) as * Improper integrals are two types and these are: a. ∫ f ( x ) dx where f ( x ) becomes infinite at a value x between the lower and upper limits of integration inclusive.4) (13. 13. a a b b Numerical Analysis Using MATLAB® and Excel®. They are also used in probability.5) we obtain –x xdu = – e dx and v = ---n n Then. denoted as Γ ( n ) .1) and this improper* integral converges (approaches a limit) for all n > 0 . We will derive the basic properties of the gamma function and its relation to the well known factorial function n! = n ( n – 1 ) ( n – 2 )… 3 ⋅ 2 ⋅ 1 (13.1 The Gamma Function The gamma function. is also known as generalized factorial function. with (13. we write (13. It is defined as Γ(n) = ∫0 x ∞ n – 1 –x e dx (13.3) (13.1) by performing integration by parts using the relation ∫ u dv Letting u = e –x = uv – v du and dv = x n–1 ∫ (13.

. we consider the limit lim ---------. for x = 0 . To work around this problem. then.= … x m–1 n x → ∞ ne x x → ∞ dm x → ∞ d x ne ne m m–1 dx dx n ( n – 1 ) ( n – 2 )… ( n – m + 1 ) x = lim -----------------------------------------------------------------------------------x x→∞ ne n – 1 ) ( n – 2 )… ( n – m + 1 )=0 = lim ( ------------------------------------------------------------------m–n x x→∞ x e n–m x m n d m–1 m–1 nx n–1 (13. we observe that (n + 1) -------------------Γ(n) = Γ n (13. (13. where m ≥ n .1) we have Γ(n) = ∫0 ∞ x n – 1 –x 1 e dx = -n ∫0 x e n –x dx (13. This can be proved with L’ Hôpital’s rule* by differentiating both numerator and denominator m times.9). the ratio of two functions. and we wish to find this limit.6) reduces to 1 Γ ( n ) = -n ∫0 x e ∞ n –x dx ∞ (13.9) By comparing the two integrals of (13.= lim ------lim -----------. It also vanishes at the upper limit as x → ∞ .6) vanishes at the lower limit.11) f(x) .= -..6) With the condition that n > 0 .= lim ----------------------------------.10) or nΓ(n) = Γ(n + 1) (13. Third Edition Copyright © Orchard Publications .Chapter 13 The Gamma and Beta Functions and Distributions x e Γ ( n ) = -----------n n –x ∞ 1 + -n x=0 ∫0 x e ∞ n –x dx (13. d n –x n x→∞ m x e dx dx x . results in the indeterminate form * Quite often. the first term on the right side of (13. such as ---------- g( x) f(a) 0 f(x ) ---------.= lim -----------------. say a . and if the limit ----dx dx x→a ddf(x) .. that is.8) and with (13.= lim ⎛ ----lim ---------f ( x ) ⁄ ----g ( x )⎞ ⎠ dx g(x) x → a ⎝ dx 13−2 Numerical Analysis Using MATLAB® and Excel®. if it exists.7) Therefore. tal’s rule states that if f ( a ) = g ( a ) = 0 . for some value of x .Then. L’Hôpig( a) 0 x → a g(x) ddf ( x ) ⁄ ----g ( x ) as x approaches a exists.

we obtain values which are not consistent with the definition of the Γ ( n ) function. when n = – 0. but not negative integer values. Γ(1) = 1 (13. For instance. the Γ ( n ) function is defined for all positive integers and positive fractional values.1) yields Γ(1) = ∞ –x –x ∞ 0 ∫0 e dx = – e = 1 (13.5 .14) and in general Γ ( 4 ) = 3 ⋅ Γ ( 3 ) = 3 ⋅ 2 = 3! Γ ( n + 1 ) = n! for n = 1.1. – 3 . For n = 1 . From (13. (13.1 shows the plot of the function Γ ( n ) versus n . – 3 and so on in (13. title('The Gamma Function'). we have derived the important relation. Third Edition Copyright © Orchard Publications 13−3 . whereas the factorial n! is defined only for zero (recall that 0! = 1 ) and positive integer values.11). plot(n. – 2. We can use MATLAB’s gamma(n) function to plot Γ ( n ) versus n . g=gamma(n). grid.15) is a very useful relation.10) for n < 0 . – 2. ylabel('Gamma(n)') Figure 13.g). and so on. – 1.12) Thus. Stated in other words. This is done with the script below which produces the plot shown in Figure 13.15) The formula of (13. 3.The Gamma Function It is convenient to use (13.11) for n > 0 .5 ) .5 ) in terms of Γ ( 0. that is. xlabel('n').10).11). and (13. … (13. we see that Γ ( n ) becomes infinite as n → 0 .13) From the recurring relation of (13. we can find Γ ( – 0. but if we substitute the numbers 0. the gamma function exists (is continuous) everywhere except at 0 and negative integer numbers. as defined in that relation. 2.05: 4. we obtain Γ(2) = 1 ⋅ Γ(1) = 1 Γ ( 3 ) = 2 ⋅ Γ ( 2 ) = 2 ⋅ 1 = 2! (13. n=−4: 0. Numerical Analysis Using MATLAB® and Excel®. it establishes the relationship between the Γ ( n ) function and the factorial n! . axis([−4 4 −6 6]). We must remember that. – 1. and for all negative fractional.

5 + 1 ) Γ ( 0.6 ) ( 1. we can use MATLAB to find any valid values of n .6 ) = ( 2.10) Γ(n + 1) Γ ( n ) = -------------------n a.5 13−4 Numerical Analysis Using MATLAB® and Excel®. From (13.717 Then. Example 13. and from math tables Therefore.6 ) = 2.11) Then. Third Edition Copyright © Orchard Publications .6 ) b.6 ) ( 0.5 0.5 ) = -----------------------.5 ) Γ(n + 1) = nΓ(n) Γ ( 3. b.= --------------0. but we can use (13.8953 ) = 3. Γ ( 0.Chapter 13 The Gamma and Beta Functions and Distributions 6 The Gamma Function 4 2 Gamma(n) 0 -2 -4 -6 -4 -3 -2 -1 0 n 1 2 3 4 Figure 13.6 ) = 0. Γ ( – 0. Γ ( 1.5 ) c. can be found in math tables. Plot of the gamma function Numerical values of Γ ( n ) for 1 ≤ n ≤ 2 .6 ) = ( 2.6 Γ ( 2. From (13. Of course. Γ ( 3.1.11) to compute values outside this range.6 )Γ ( 1.6 ) ( 1.5 ) Γ ( 0.8953 Γ ( 3.10) or (13.6 ) Γ ( 1.1 Compute: Solution: a.

2.6 0.5 and using the result of (b).5 gammaln(x) 1.3129 0.544 We can verify these answers with MATLAB as follows: a=gamma(3. c.5 ) = ( 2 ) ( 0.5 ) = ( – 2 ) ( 1.5 ) = – 2 Γ ( 0. b=gamma(0.5 ) = – 2 Γ ( 0.5) a = 3.772 ) = – 3.8862 Γ ( 0. But because it first computes the natural log.7170 1. we can use the EXP(GAMMALN(n)) function to evaluate Γ ( n ) at some positive value of n .5 + 1 ) = Γ ------------------------------------------Γ ( – 0. x 3.7725 c = -3.5449 Excel does not have a function which evaluates Γ ( n ) directly. It does. c=gamma(−0. Third Edition Copyright © Orchard Publications 13−5 . have the GAMMALN(x) function.6). Therefore.7725 #NUM! Figure 13. it does not produce an answer if n is negative as shown in Figure 13. From (13. ( 0.The Gamma Function and from math tables Therefore.5724 #NUM! exp(gammaln(x))= gamma(x) 3.5 ) ( – 0. Using Excel to find Γ ( n ) Numerical Analysis Using MATLAB® and Excel®.8862 ) = 1. Γ ( – 0.772 Then.5).5 ) = Γ – 0.5 -0.5 – 0. however.2.5 ) = 0.10) Γ(n + 1) Γ ( n ) = -------------------n Γ ( 1.7170 b = 1.

19) into (13. Proof: From (13.1).22) (13.18) yields Γ ( n ) = ( n – 1 ) ( n – 2 )Γ ( n – 2 ) (13.23) Example 13.Chapter 13 The Gamma and Beta Functions and Distributions Example 13. Then. we obtain Γ ( n – 1 ) = ( n – 2 )Γ ( n – 2 ) (13.25) 13−6 Numerical Analysis Using MATLAB® and Excel®. we have or Γ ( n ) = ( n – 1 ) ( n – 2 ) ( n – 3 )… 1 Γ ( n ) = ( n – 1 )! (13.18) Next.11).16) is true. Γ(n) = ∫0 x ∞ n – 1 –x e dx ∞ (13. we obtain Γ ( n ) = ( n – 1 ) ( n – 2 ) ( n – 3 )… 1 Γ ( 1 ) (13.2 Prove that when n is a positive integer.18).21) and since Γ ( 1 ) = 1 . Third Edition Copyright © Orchard Publications .17) (13. the relation Γ ( n ) = ( n – 1 )! (13. 1⎞ .3 Use the definition of the Γ ( n ) function to compute the exact value of Γ ( 1 ⁄ 2 ) Solution: From (13.20) By n repeated substitutions.= Γ ⎛ -⎝2⎠ ∫0 ∞ x 0.19) Substitution of (13.5 – 1 – x e dx = ∫0 x – 0.24) Then. replacing n with n – 1 on the left side of (13.5 – x e dx (13. Γ(n + 1) = nΓ(n) Γ ( n ) = ( n – 1 )Γ ( n – 1 ) (13.

30) and by recalling that: 1. Third Edition Copyright © Orchard Publications 13−7 . Numerical Analysis Using MATLAB® and Excel®. we let 1⎞ .27) by (13. from differential calculus 2 2 d 2 d u2 .= 2 Γ ⎛ -⎝2⎠ ∫0 ∫0 2 ∞ e –x 2 dx (13.The Gamma Function Letting x = y2 we obtain dx ----.e = e u ----du du (13.= 2 Γ ⎛ -⎝2⎠ 1⎞ .29) Now. we obtain 1⎞ .26) Next. that is.5 ) – y e 2 2ydy = 2 ∫0 ∞ y ye –1 –y 2 dy = 2 ∫0 e ∞ –y 2 dy (13.u = 2ue u ----.= Γ ⎛ -⎝2⎠ ∫0 ∞ y 2 ( – 0. ∫ρ ρ2 1 ρe –ρ 2 1 –ρ 2 -e dρ = – -2 (13.28) yields 1⎞ Γ ⎛ -⎝2⎠ 2 = 4 ∫0 ∞ e –x 2 dx ∫0 ∞ e –y dy = 4 ∫0 ∫0 e ∞ ∞ –( x + y ) 2 2 dx dy (13.33) We observe that as x → ∞ and y → ∞ .31) 2.32) Then.27) (13.29) to polar coordinates by making the substitution ρ 2 = x +y 2 2 (13. we define Γ ( 1 ⁄ 2 ) as a function of both x and y .28) ∞ e –y 2 dy Multiplication of (13. the total area of a region is found by either one of the double integrals A = ∫ ∫ dx dy = ∫ ∫ r dr dθ (13.25).= 2y dy or dx = 2ydy By substitution of the last three relations into (13. we convert (13.

= ------------. for n = – 1. 4 Γ ( – 1.29) yields 1⎞ Γ ⎛ -⎝2⎠ 2 = –2 ∫0 π⁄2 ⎛ e–ρ ⎝ 2 ⎞ dθ = – 2 ρ = 0⎠ ∞ ∫0 π⁄2 ( 0 – 1 ) dθ = 2 ∫0 π⁄2 dθ = 2 θ π⁄2 0 = π and thus.30).34) Substitution of (13.Chapter 13 The Gamma and Beta Functions and Distributions ρ → ∞ and θ → π ⁄ 2 (13.36) -.and Γ ( 0.= – ----.5 ) b .= -3 – 1.= –2 π – 0.5 .= ---------– 2.5 Other interesting relations involving the Γ ( n ) function are: π Γ ( n )Γ ( 1 – n ) = ------------sin n π for 0 < n < 1 2 2n – 1 (13.π Γ ( – 1.4 Compute: Solution: Using the relations we obtain: a.= ------------------.5 – 1.5 ) = -----------------------------. Γ ( – 2. Third Edition Copyright © Orchard Publications .5 ) = Γ ( n ) = --------------------n π b. Γ ( – 0.5 ) –2 π .5 – 1.37) for any n ≠ negative integer 13−8 Numerical Analysis Using MATLAB® and Excel®.5 ) = -----------------------------.5 ) π Γ ( – 0.5 c.5 ) Γ (n + 1) .5 – 2.5 . we have obtained the exact value 1 -⎞ = Γ ⎛ -⎝2⎠ π (13.5 ) c.5 .= ------------------.5 a. for n = – 0.π 3 8 Γ ( – 2.π Γ ( – 2.5 15 – 2. Γ ( 0.34) into (13.5 ) = ---------------.= --------.5 + 1 ) Γ ( – 0.5 – 0.33) and (13. (13.5 ) .5 + 1 ) Γ ( – 1. 4 -.⎞ = Γ ( n )Γ ⎛ n + 1 ⎝ 2 ⎠ πΓ ( 2n ) (13.35) Example 13. Γ ( – 1. for n = – 2.

6 Compute the product 1⎞ ⎛2⎞ .The Gamma Function Γ ( n + 1 ) = n! = n –n ⎧ 1 .5 Use (13.⎞ = ---------⎝ 3⎠ ⎝ ⎠ 3 sin π -3 or Numerical Analysis Using MATLAB® and Excel®. If n is a positive integer. Third Edition Copyright © Orchard Publications 13−9 .⎞ = Γ ⎛ -Γ ⎛ -⎝ ⎠ ⎝ ⎠ ⎠ ⎝2⎠ ⎝ 2 2 π 2 sin -2 or 1 -⎞ Γ ⎛ -⎝2⎠ 2 = π Therefore.– ------------------------571 .⎞ Γ ⎛1 – 1 -.36) to prove that 1⎞ .+ … ⎫ 2 π nn e ⎨ 1 + -------⎬ 12n 288n 2 51840n 3 2488320n 4 ⎩ ⎭ (13.⎞Γ ⎛ 1 .38) is referred to as Stirling’s asymptotic series for the Γ ( n ) function.36).Γ -Γ ⎛ -⎝3⎠ ⎝3⎠ Solution: Using (13.– ------------------1 .= Γ ⎛ -⎝2⎠ π Proof: 1 1 π-. the factorial n! can be approximated as n! ≈ 2 π nn e n –n (13.38) Relation (13.⎞ = ---------. 1 -⎞ = Γ ⎛ -⎝2⎠ π Example 13.39) Example 13. we obtain 1⎞ ⎛ π .+ ------------139 .Γ 1–1 Γ ⎛ --.

we type and execute the expression sqrt(2*pi*50)*50^50*exp(−50) ans = 3.39) to compute 50! Solution: 50! ≈ 2 π × 50 × 50 50 ×e – 50 We use MATLAB as a calculator.+ -.0414e+064 We can check this answer with the Excel FACT(n) function.04141E+64 13−10 Numerical Analysis Using MATLAB® and Excel®.Γ -Γ ⎛ -. we use the relation Γ ( n + 1 ) = n! and the MATLAB gamma(n) function.⎞ Γ(2) = 2 Γ ⎛ -⎝2⎠ 3⎞ πΓ ⎛ 2 ⋅ -⎝ 2⎠ πΓ ( 3 ) 3 2! π πΓ(3) π .= ----------. Third Edition Copyright © Orchard Publications . To find the exact value.Γ -. Then. that is. gamma(50+1) ans = 3.7 Use (13.= -----.⎞ = -----------------.0363e+064 This is an approximation.= -----Γ ⎛ -⎝2⎠ 4Γ(2) 4⋅1 2 Example 13. =FACT(50) and Excel displays 3.37) to find 3 -⎞ Γ ⎛ -⎝2⎠ Solution: 2 3–1 3 ⎞ ⎛3 1 ⎞ .Chapter 13 The Gamma and Beta Functions and Distributions 1⎞ ⎛ 2⎞ 2π 2 3π π .= ------------⎝ 3⎠ ⎝ 3⎠ 3 3⁄2 3 Example 13. that is.= ------------Γ ⎛ -.8 Use (13.= ⎝ 2 ⎠ ⎝2 2 ⎠ 3 2 .

Some examples follow.9 Using the definition of the Γ ( n ) function.= ---6 ⎝2⎠ 2 2 ∫0 ∞ y e dy 5 –y Γ ( 6 ) 5! 120 15 = ----------. ∫0 ∞ x e dx b. Solution: By definition. evaluate the integrals a. then. and by substitution.⎞ e – y ----. Example 13.10 A negatively charged particle is α meters apart from the positively charged side of an electric field. considered to be the center of attraction 0 . dx = dy ⁄ 2 . Numerical Analysis Using MATLAB® and Excel®. It is initially at rest. Solution: Let the center of attraction 0 be the point zero on the x −axis. ∞ n – 1 –x e dx = Γ ( n ) ∫0 b.= ----64 64 64 8 Example 13. and then moves towards the positively charged side with a force inversely proportional to its distance from it.3. ∞ ∞ x e dx = Γ ( 5 ) = 4! = 24 4 –x Let 2x = y .The Gamma Function The Γ ( n ) function is very useful in integrating some improper integrals.= -------.= ----. Assuming that the particle moves towards the center of the positively charged side. derive an expression for the time required the negatively charged particle to reach 0 in terms of the distance α and its mass m . ∫0 x e 5 – 2x dx = ∫0 ∞ 5 dy 1 ⎛y -. as indicated in Figure 13. a. Third Edition Copyright © Orchard Publications 13−11 . 4 –x ∫0 ∞ x e 5 – 2x dx ∫0 x Then.

( dx ) mvdv = – -x (13.----. we obtain mv --------.45) where C represents the constants of integration of both sides. dx ----.10 α x By Newton’s law.42) Substitution of (13.42) into (13. and it is evaluated from the initial condition that v = 0 when x = α .Chapter 13 The Gamma and Beta Functions and Distributions movement of particle 0 Figure 13.44) Integrating both sides of (13.43) or (13.44).40) yields dv . 13−12 Numerical Analysis Using MATLAB® and Excel®.= ----.= ----2 dx dx dt dt dt (13.= – k ln x + C 2 2 (13. Then. and moves towards the origin at x = 0 .= v dt 2 dv dv dx dx dv .= v ----------. dx . Sketch for Example 13.3. the particle is assumed to be located on the x −axis at point x = α . Third Edition Copyright © Orchard Publications . Let the velocity of the particle be v . At t = 0 .40) where m = mass of particle x = distance (varies with time) k = positive constant of proportionality and the minus (−) sign indicates that the distance x decreases as time t increases.= –k -m ------2 x dt 2 (13.= –k -mv ----dx x k . Then.41) and (13.

49) Solving (13. Third Edition Copyright © Orchard Publications 13−13 . and at τ = t . we choose the negative sign.46) and by substitution into (13.ln -----. x = 0 . that is.51) To simplify (13. we let .= – ----m x dt (13.51). then e y = ln ⎛ -⎝x⎠ α y α = -x –y (13. we express (13.ln -v = ----. are being replaced with 0 and ∞ respectively. Then. since x = α e .53) x→α the lower and upper limits of integration in (13.51). mv α --------.The Gamma Function C = k ln α (13.⎞ .-----------------------dt = – ----2k ln ( α ⁄ x ) (13. noting that the integration on the right side is with respect to the distance x where at t = 0 . Therefore.= ± ----m x dt 2 (13.= k ln α – k ln x = k ln -2 x 2 (13.45). 2k dx α .47) Solving for v and taking the square root of both sides we obtain 2k α dx .51) as Numerical Analysis Using MATLAB® and Excel®.50) We are interested in the time required for the particle to reach the origin 0 .52) or Also. and dx = – α e dy α⎞ α⎞ .48) Since x decreases as t increases. it is found from the relation below.= 0 and lim ln ⎛ -. x = α . We denote this time as T .=∞ lim ln ⎛ -⎝ x⎠ x→0 ⎝ x⎠ –y (13. T = ∫ m dτ = – ----2k 0 t ∫α -----------------------ln ( α ⁄ x ) 0 dx (13.49) for dt we obtain dx m.

we convert (13.59) to polar coordinates by letting x = ρ cos θ and y = ρ sin θ Then.57) For m > 0 and n > 0 .58) it written as Γ ( m )Γ ( n ) = ∫0 ∞ x 2 (m – 1) 2xe –x 2 dx ∫0 e ∞ y 2 2 (n – 1) 2ye –y 2 dy = 4 ∫0 ∞ x 2m – 2 xe –x 2 dx ∫0 y ∞ 2n – 2 ye –y 2 dy = 4 ∫0 ∫ 0 ∞ ∞ x 2m – 1 2n – 1 – ( x + y ) y 2 (13. Γ(m) = ∫0 x m – 1 –x e dx (13.58) where u and v are dummy variables of integration.54) Example 13. Third Edition Copyright © Orchard Publications .56) Also.= α ----2k y –y ∫0 y ∞ –1 ⁄ 2 –y e dy Finally.57) yields Γ ( m )Γ ( n ) = ∫0 ∞ u m – 1 –u e du ∫0 v ∞ n – 1 –v e dv (13. letting u = x 2 and v = y 2 . relation (13.55) From the definition of the Γ ( n ) function.59) dx dy Next. using the definition of the Γ ( n ) function. Γ(n) = ∫0 x ∞ ∞ n – 1 –x e dx (13. 13−14 Numerical Analysis Using MATLAB® and Excel®.11 Evaluate the integrals ∫0 Solution: π⁄2 cos θ dθ and n ∫0 π⁄2 sin θ dθ n (13. multiplication of (13.56) by (13.= α π ----⎝ 2 ⎠ 2k 2k 2k (13. we obtain du = 2xdx and dv = 2ydy . Next. Then. with these substitutions. we obtain 1⎞ m m πm .Chapter 13 The Gamma and Beta Functions and Distributions m T = – ----2k ∫0 ∞ – α e dy m -------------------.----.= α ------T = α Γ ⎛ -.

= -------------------------sin θ dθ = --------------------------------------------2 m 1 m+1⎞ ⎛ ⎞ ⎛ -+1 Γ --2 Γ -. m = ( 1 2n – 1 with 0 .61) we obtain ∫0 π⁄2 cos 2m – 1 θ ⋅ sin 2n – 1 θ dθ = ------------------------- Γ ( m )Γ ( n ) 2Γ(m + n) 2 (13.62) as 2 π⁄2 ∫0 n+1⎞ n+1⎞ ⎛ 1 ⎞ .-----sin θ dθ = 2 n ⎛ ⎞ -+1 Γ -⎝2 ⎠ ∫0 π⁄2 cos θ dθ = n ∫0 π⁄2 n > –1 (13. that is.+ ------------⎝ 2 ⎠ ⎝2 2 ⎠ (13.64) are equal since m and n can be interchanged.= ------------------------. we obtain the integral of the sin θ function as n ∫0 π⁄2 m+1⎞ +1 1 Γ ⎛ -------------------------⎞ .65) Numerical Analysis Using MATLAB® and Excel®.62).62) n + 1) ---------------.The Gamma Function Γ ( m )Γ ( n ) = 4 = 2 ∫0 ∫0 ∫0 π⁄2 π⁄2 ∞ ( ρ cos θ ) 2m – 1 2m – 1 ( ρ sin θ ) θ dθ 2n – 1 – ρ 2 e ρ dρ dθ 2 cos θ ⋅ sin 2n – 1 ∫0 ρ ∞ 2m + 2n – 2 – ρ (13. Therefore.60) 2 ρ dρ e To simplify (13.-----cos θ dθ = -----------------------------------------2 n n+1 1⎞ ⎛ ⎞ ⎛ Γ --+1 2 Γ ----------.63) If. and and this expression can be simplified by replacing 2m – 1 with n .+ -⎝2 ⎠ ⎝ 2 2⎠ (13. we replace 2m – 1 with 0 and 2n – 1 with m . n + 1⎞ Γ ⎛ ----------⎝ 2 ⎠ π n ---------------------.Γ -Γ ⎛ ----------Γ ⎛ ----------⎝ ⎝ ⎠ ⎝ ⎠ 2 ⎠ π 2 2 n .61) 2m – 1 2n – 1 θ dθ ⋅ Γ ( m + n ) Rearranging (13.60)..63) and (13. then. Then.60) is written as Γ ( m )Γ ( n ) = 2 = 2 ∫0 ∫0 π⁄2 cos cos 2m – 1 θ ⋅ sin θ ⋅ sin 2n – 1 θ dθ ∫0 w ∞ m + n – 1 –w e dw π⁄2 (13. n = -.-----. dw = 2 ρ d ρ and thus relation (13. Third Edition Copyright © Orchard Publications 13−15 . we obtain the special case of (13. we let ρ2 = w .⎞Γ ⎛ m Γ ⎛ -⎝ ⎝ ⎠ ⎝ ⎠ 2 ⎠ π 2 2 m . that is. in (13.64) We observe that (13..

0 1.Chapter 13 The Gamma and Beta Functions and Distributions The relations of (13.4 1.8 2.0 0.66) A detailed discussion of this probability distribution is beyond the scope of this book. it will suffice to say that it is used in reliability and queuing theory. Figure 13.0823 0.1084 0.4 n 3.0494 0.05 0. 13−16 Numerical Analysis Using MATLAB® and Excel®. n. The pdf for the gamma distribution.0 Γ(n) 2.0920 0.0023 0.66) * Several probability distributions are presented in Mathematics for Business.0 2.0000 0.0082 0.66) beta = the parameter β in (13.20 0. n.0 β^n 8. Science.beta.4 shows the probability density function (pdf) of the gamma distribution for n = 3 and β = 2 .cumulative) where: x = value at which the distribution is to be evaluated alpha = the parameter n in (13.0608 0. When n is a positive integer.0268 0. β > 0 n β Γ(n) (13. 13.6 1.0 f(x) 0.4 0. Third Edition Copyright © Orchard Publications .15 f(x) 0. and Technology.1007 0. it is referred to as Erlang distribution.10 0.x > 0.0 β 2.65) are known as Wallis’s formulas.0167 0.2 1.2 The Gamma Distribution One of the most common probability distributions* is the gamma distribution which is defined as n – 1 –x ⁄ β e x f ( x.8 1.00 0 2 4 6 x 8 10 12 Figure 13. x 0. ISBN 0−9709511− 0−8.2 2.2 0.6 0.0379 Probability Density Function of the gamma distribution for n = 3 and β = 2 0. β ) = ------------------------.alpha. We can evaluate the gamma distribution with the Excel GAMMADIST function whose syntax is GAMMADIST(x.4.0719 0.

Third Edition Copyright © Orchard Publications 13−17 . n = 3 . denoted as B ( m. Then.2.0925 We observe that these values are consistent with the plot of Figure 13. and β = 2 Solution: Since we are interested in the probability density function (pdf) values. y → 0 . dx = – dy .3. a.3 The Beta Function The beta function. x = 7 . that is.12 Use Excel’s GAMMADIST function to evaluate f ( x ) .1353 =GAMMADIST(7. Example 13. it returns the probability density function* (pdf). n ) .3. Example 13. Numerical Analysis Using MATLAB® and Excel®.2. n ) = B ( n. 13. the pdf of the gamma distribution if: a. we specify the FALSE condition.The Beta Function cumulative = a TRUE / FALSE logical value.13 Prove that B ( m. if TRUE. and β = 2 b. Therefore. is defined as B ( m.67) where m > 0 and n > 0 .FALSE) returns 0. n = 3 . b. m ) (13. x = 4 . =GAMMADIST(4. n ) = ∫0 x 1 m–1 (1 – x) n–1 dx (13. then. and if FALSE.FALSE) returns 0.4.68) Proof: Let x = 1 – y . GAMMADIST returns the cumulative distribution function (cdf). * Several probability density functions are also presented on the text mentioned on the footnote of the previous page. y → 1 and as x → 1 . We observe that as x → 0 .

70). n ) = -----------------------Γ(m + n) (13. dx = 2 sin θ cos θ d θ . n ) = 2 ∫0 π⁄2 cos 2m – 1 θ ⋅ sin 2n – 1 θ dθ (13. Third Edition Copyright © Orchard Publications . θ → 0 and as x → 1 . Example 13.62) and (13. n ) function is also useful in evaluating certain integrals as illustrated by the following examples. m ) and thus (13.70) ∫0 π⁄2 2m – 1 θ ) ( cos 2m – 1 θ ) dθ Example 13. then. 13−18 Numerical Analysis Using MATLAB® and Excel®. We observe that as x → 0 .14 Prove that B ( m.Chapter 13 The Gamma and Beta Functions and Distributions B ( m. n ) = = ∫0 ∫0 1 x m–1 (1 – x) y n–1 dx = – dy = ∫1 ( 1 – y ) n–1 0 m–1 [1 – (1 – y)] n–1 dy 1 (1 – y) m–1 n–1 ∫0 y 1 (1 – y) m–1 dy = B ( n. The B ( m. θ → π ⁄ 2 . n ) = ∫0 x ∫0 1 m–1 (1 – x) 2 n–1 dx 2 n–1 = 2 = 2 π⁄2 ( sin θ ) ( sin m–1 ( cos θ ) sin θ cos θ dθ (13.68) is proved.71) Proof: The proof is evident from (13. Then.15 Prove that Γ ( m )Γ ( n ) B ( m. B ( m.69) Proof: We let x = sin 2θ .

n) function. However. p g Γ(m) exp(gammaln(m)) 24.16 Evaluate the integral ∫0 x ( 1 – x ) dx 4 3 1 (13.4) z = 1/280 Excel does not have a function that computes the B ( m.74) Numerical Analysis Using MATLAB® and Excel®.= -------Γ(9) 8! 40320 40320 280 (13.n)= Γ(m) x Γ(n) / Γ(m+n) 1/280 m= 5 Figure 13.= -------------. format rat. % display answer in rational format z=beta(5.The Beta Function Example 13. 4 ) Using (13.5.00 n= 4 Γ(n) exp(gammaln(n)) 6. Example 13.71) for its computation as shown in Figure 13. n ) = ∫0 x 1 m–1 (1 – x) n–1 dx and thus for this example.= -------------.dx 0 2–x 2 2 (13.73) We can also use MATLAB’s beta(m.= --------.17 Evaluate the integral ∫ x --------------. Computation of the beta function with Excel. Third Edition Copyright © Orchard Publications 13−19 . For this example. 4 ) = ----------------------. ∫0 x ( 1 – x ) d x 4 3 1 = B ( 5.71) we obtain Γ ( 5 )Γ ( 4 ) 4! 3! 24 × 6 144 1 B ( 5.00 Γ(m+n) exp(gammaln(m+n)) 40320. n ) function directly. we can use (13.00 Beta(m.5.72) Solution: By definition B ( m.

n > 0 (13. etc.dv = 4 2 1–v 2 ∫0 v ( 1 – v ) 2 1 –1 ⁄ 2 dv 1⎞ Γ ( 3 )Γ ( 1 ⁄ 2 ) . Then. Using (13.75) Γ(1⁄2) = π Γ ( 7 ⁄ 2 ) = ( 7 ⁄ 2 – 1 )Γ ( 7 ⁄ 2 – 1 ) = ( 5 ⁄ 2 ) Γ ( 5 ⁄ 2 ) = ( 5 ⁄ 2 ) ( 5 ⁄ 2 – 1 )Γ ( 5 ⁄ 2 – 1 ) = 5 ⁄ 2 ⋅ 3 ⁄ 2 ⋅ ( 3 ⁄ 2 – 1 )Γ ( 3 ⁄ 2 – 1 ) = ( 15 ⁄ 8 )Γ ( 1 ⁄ 2 ) = 15 π ⁄ 8 (13. is shown in Figure 13.4 The Beta Distribution The beta distribution is defined as x ( 1 – x ) . As with the gamma probability distribution.dx = 4 -----------------------------------------. then x = 4v .77) 13.78) A plot of the beta probability density function (pdf) for m = 3 and n = 2 . -⎝ 2⎠ Γ(7 ⁄ 2) (13. driving habits. m. (13.Chapter 13 The Gamma and Beta Functions and Distributions Solution: Let x = 2v .76) Then. Third Edition Copyright © Orchard Publications . and dx = 2dv . and as x → 2 .71) we can express the beta distribution as Γ(m + n) . it will suffice to say that it is used in computing variations in percentages of samples such as the percentage of the time in a day people spent at work. n ) = -------------------------------------B ( m.79) 13−20 Numerical Analysis Using MATLAB® and Excel®. m. v → 1 . We observe that as x → 0 . eating times and places. v → 0 .75) and (13.= 4 2 -----------------------------= 4 2 ⋅ B ⎛ 3.74) becomes 2 2 ∫0 where Γ ( 3 ) = 2! 1 4v 8 -----------------.76) we obtain ∫ x 2 2 ⋅ 2! ⋅ π --------------. a detailed discussion of the beta probability distribution is beyond the scope of this book.2 dv = -----2 2 – 2v 2 ∫0 1 v --------------.x < 0 < 1.= 64 15 15 π ⁄ 8 0 2–x 2 2 (13. n > 0 f ( x. from (13.⋅ xm – 1 ( 1 – x )n – 1 f ( x.74). m. n ) = -----------------------Γ ( m )Γ ( n ) x < 0 < 1.6. n ) m–1 n–1 (13. m. (13.

16 0.2580 0.alpha.9800 0.0406 0.0047 0.6600 f(x. However.6.8 0.0400 0.5253 0.08 0.9156 Probability Density Function 0. we see that when x = 1 .6.12 0.79) beta = the parameter n in (13.9400 0. f ( x.0784 0.9200 0.7000 0.06 0.0004 0.0.8356 0.1080 0.0324 0.0100 0.6800 0. m.3. the cumulative distribution (the area under the curve) at this point is 100 % or unity since this is the upper limit of the x −range.n) 0.14 0.1) which returns 1.8200 0.The Beta Distribution x 0.0 0.0576 0.0256 0.0 x(m-1) 0.6 0.4530 0.m.22 0.6774 0.1521 0.8400 0.18 0.10 0.9600 0.28 0.1156 Figure 13.0484 0.34 m 3.0000 0.02 0.32 0.0 Γ(m+n) 24.8 f(x.0707 0.A.0900 0.7560 0.0 0.0 Γ(n) 1.6003 0.7600 0.0144 2.0036 of the Beta Distribution 0.00 0.0 n 2.0676 0.24 0. The pdf of the beta distribution We can evaluate the beta cumulative distribution function (cdf) with Excels’s BETADIST function whose syntax is BETADIST(x.0000 0.m.20 0. This value can be verified by =BETADIST(1.8000 0.0 0.8800 0.0184 0.79) A = the lower bound to the interval of x B = the upper bound to the interval of x From the plot of Figure 13.n) 1.B) where: x = value between A and B at which the distribution is to be evaluated alpha = the parameter m in (13.3840 0.0 Γ(m) 2.3188 0.7400 1.7200 0.4 0.26 0.8600 0.1024 0.2 0.0016 (1-x)(n-1) 1.0000 Numerical Analysis Using MATLAB® and Excel®.04 0.0196 0.beta.2 0. Third Edition Copyright © Orchard Publications 13−21 .7800 0.4 x 0. n ) which represents the probability density function.9000 0.2.30 0.6 1.0000 0.0064 for m = 3 and n = 20.2023 0. is zero.0 0.

+ … ⎫ 2 π nn e ⎨ 1 + -------⎬ 12n 288n 2 51840n 3 2488320n 4 ⎩ ⎭ is referred to as Stirling’s asymptotic series for the Γ ( n ) function.– ------------------------571 . the 13−22 Numerical Analysis Using MATLAB® and Excel®.Chapter 13 The Gamma and Beta Functions and Distributions 13. denoted as Γ ( n ) . … • We can use MATLAB’s gamma(n) function to obtain values of Γ ( n ) . 2. • The Γ ( n ) function and the factorial n! are related as Γ ( n + 1 ) = n! for n = 1. • To evaluate Γ ( n ) when n is a positive integer. 3. Third Edition Copyright © Orchard Publications . • The Γ ( n ) function is defined for all positive integers and positive fractional values. and for all negative fractional.– ------------------1 . Γ ( n ) = ( n – 1 )! Γ(1⁄2 ) = π π Γ ( n )Γ ( 1 – n ) = ------------sin n π for 0 < n < 1 2 2n – 1 Γ(n) π = πΓ ( 2n ) • The relation Γ ( n + 1 ) = n! = for any n ≠ negative integer n –n ⎧ 1 . is also known as generalized factorial function. It is defined as Γ(n) = ∫0 x ∞ n – 1 –x e dx • It is convenient to use the relation (n + 1) Γ(n) = Γ -------------------n for n < 0 and the relation nΓ(n) = Γ(n + 1) for n > 0 . but not negative integer values. • We can use the EXP(GAMMALN(n)) function to evaluate Γ ( n ) at some positive value of n . If n is a positive integer.+ ------------139 . we can use the relation • Other useful relations are shown below.5 Summary • The gamma function.

• The gamma distribution which is defined as n – 1 –x ⁄ β x e x > 0. • The relations n+1 -⎞ Γ ⎛ ----------⎝ 2 ⎠ π n . n ) function is also useful in evaluating certain integrals. n ) = -----------------------Γ(m + n) • The beta B ( m. • The beta distribution is defined as (1 – x) x . β ) = ------------------------n β Γ(n) • The beta function.Summary factorial n! can be approximated as n! ≈ 2 π nn e n –n • The Γ ( n ) function is very useful in integrating some improper integrals. β > 0 f ( x. Third Edition Copyright © Orchard Publications 13−23 .n) function to evaluate the beta B ( m. • We can use MATLAB’s beta(m. n ) m–1 n–1 Numerical Analysis Using MATLAB® and Excel®. n ) and gamma function Γ ( n ) are related by Γ ( m )Γ ( n ) . n ) function. B ( m.-----sin θ dθ = ---------------------2 n ⎛ ⎞ Γ --+1 ⎝2 ⎠ ∫0 π⁄2 cos θ dθ = n ∫0 π⁄2 n > –1 are known as Wallis’s formulas. n ) where m > 0 and n > 0 is defined as B ( m.x < 0 < 1. n. n > 0 f ( x. n. n ) = ∫0 x 1 m–1 (1 – x) n–1 dx • The beta function B ( m. B ( m. m. m. n ) = -------------------------------------B ( m.

Chapter 13 The Gamma and Beta Functions and Distributions 13.25 .5 and n = – 1. e. ∫0 π⁄2 tan θ dθ dx ∫0 --------------------2 3x – x 3 13−24 Numerical Analysis Using MATLAB® and Excel®. ∫0 ∞ e –x 3 dx 3 ∫0 ∞ xe –x dx ∫0 -----------------4 1–x 1 dx d. Third Edition Copyright © Orchard Publications . Given that m = 2. n ) Verify your answer with MATLAB and Excel 3. b. Evaluate the following integrals a.6 Exercises 1. compute B ( m. Given that m = 10 and n = 8 . compute Γ(m + n) -----------------------Γ ( m )Γ ( n ) Verify your answer with MATLAB and Excel 2. c.

1739 We cannot check the answer with Excel because it cannot compute negative values.6256 5.25.1419 × 10 = -----------------------------------------------------------------------------17 ⋅ 16 ⋅ 15 ⋅ 14 ⋅ 13 ⋅ 12 ⋅ 11 ⋅ 10 980179200 Check with MATLAB: beta(10.5 )Γ ( – 1. By repeated use of the relations n Γ ( n ) = Γ ( n + 1 ) for n > 1 and Γ ( n ) = Γ n we obtain Γ(m + n) Γ ( 2.25 ) Γ ( 2.= --------------------------------------------------------------------------------Γ ( –5 ⁄ 4 + 1 ) 3 ⁄ 2 ⋅ Γ ( 3 ⁄ 2 ) ⋅ ( –4 ⁄ 5 ) ⋅ Γ ( –1 ⁄ 4 ) Γ ( 3 ⁄ 2 + 1 ) ⋅ -----------------------------(1) ( –5 ⁄ 4 ) 5 Γ(1 ⁄ 4) 1 ⁄ 4 ⋅ Γ(1 ⁄ 4) .25 ) ) Γ ( 1. where we find that Γ ( 1 ⁄ 4 ) = 3. n ) = -----------------------. we obtain their approximate values from tables.--------------------------Γ ( m + n ).5 + ( – 1.for n < 1 . (n + 1) -------------------. Third Edition Copyright © Orchard Publications 13−25 .25 ) Γ(5 ⁄ 4) -----------------------. Then.= ----48 π ⋅ Γ ( 3 ⁄ 4 ) Γ ( 3 ⁄ 4 ) 3 ⁄ 2 ⋅ 1 ⁄ 2 ⋅ Γ ( 1 ⁄ 2 ) ⋅ ( – 4 ⁄ 5 ) ⋅ -----------------–1 ⁄ 4 There are no exact values for Γ ( 1 ⁄ 4 ) and Γ ( 3 ⁄ 4 ) .5.25 ) Γ ( 5 ⁄ 2 ) ⋅ Γ ( –5 ⁄ 4 ) Γ(1 ⁄ 4 + 1) 1 ⁄ 4 ⋅ Γ( 1 ⁄ 4 ) = --------------------------------------------------------------. by substitution into (1) we obtain: 3.5 )Γ ( – 1. 2.= ---------------------------------------.= ----⋅ = 0.= -----------------------------. gamma(m+n)/(gamma(m)*gamma(n)) ans = 0.= -------------------------.2254 Γ ( m )Γ ( n ) Check with MATLAB: m=2. therefore.= -------------------------------------------Γ ( m )Γ ( n ) Γ ( 2.6256 and Γ ( 3 ⁄ 4 ) = 1.2254 .= 5. Γ ( m )Γ ( n ) Γ ( 10 ) ⋅ Γ ( 8 ) ( 9! ) × ( 7! ) B ( m. n=−1.1739 -----------------------48 π ⋅ 1.1419e-006 Numerical Analysis Using MATLAB® and Excel®.= -------------------------------------.⋅ -----------------------------= --------------------------------------------------------------------------------------------.Solutions to End−of−Chapter Exercises 13.7 Solutions to End−of−Chapter Exercises 1.= ------------------------17! Γ(m + n) Γ ( 18 ) 9⋅8⋅7⋅6⋅5⋅4⋅3⋅2⋅7⋅6⋅5⋅4⋅3⋅2 = -----------------------------------------------------------------------------------------------------------------------------------------17 ⋅ 16 ⋅ 15 ⋅ 14 ⋅ 13 ⋅ 12 ⋅ 11 ⋅ 10 ⋅ 9 ⋅ 8 ⋅ 7 ⋅ 6 ⋅ 5 ⋅ 4 ⋅ 3 ⋅ 2 –6 7⋅6⋅5⋅4⋅3⋅2 5040 .8) ans = 5.

∫0 π⁄2 tan θ dθ = ∫0 π⁄2 1⁄2 sin θ ⎛ ----------. Third Edition Copyright © Orchard Publications .= -4 4 1–x 1 = -4 ∫0 1 . Γ ( 3 ⁄ 4 ) = -----------------Γ(1 ⁄ 4) and by substitution into (1) ∫0 -----------------4 1–x 1 dx 1 Γ(1 ⁄ 4) ⋅ π 1 Γ(1 ⁄ 4) ⋅ Γ(1 ⁄ 2) .⋅ y ( – 2 ⁄ 3 ) dy .62). then x = y 3 1⁄3 1 .Chapter 13 The Gamma and Beta Functions and Distributions 3. 1 1 dx ----------------.⎞ dθ = ⎝ cos θ ⎠ ∫0 π⁄2 ( sin θ ) 1⁄2 ( cos θ ) –1 ⁄ 2 ( dθ ) From (13. ∞ e –x 3 ∫0 e ∫0 ∞ ∞ –y 1 1⁄3–1 1 ⎛1⎞ -⋅y .⋅ ----------------------------------------dy = -4 Γ(3 ⁄ 4) (1) or π π Γ ( 3 ⁄ 4 ) ⋅ Γ ( 1 – 3 ⁄ 4 ) = ------------------------.⋅ -------------------------------.⋅ { Γ ( 1 ⁄ 4 ) } 2 = --------------------------= ------------4 2π 4 2π d. Let x = y . 4 ∞ xe –x 3 dx = x 2 – 1 –x e dx = Γ ( 2 ) = 1! = 1 ( –3 ⁄ 4 ) 1 We let x = y or x = y 1⁄4 1 . dx = -3 dx = ∫0 b. a.= ------------sin ( 3 π ⁄ 4 ) 2⁄2 Γ(3 ⁄ 4) ⋅ Γ(1 ⁄ 4) = 2π 2 π .( –3 ⁄ 4 ) 1 --------------⋅y dy = -4 1–y 1⁄2–1 ∫0 ( 1 – y ) ⋅y ( –3 ⁄ 4 ) dy ∫0 ( 1 – y ) 1 ⋅y 1⁄4–1 1 Γ(1 ⁄ 4) ⋅ Γ(1 ⁄ 2) .= -.⋅ ----------------------------------------= -4 2π ⁄ Γ(1 ⁄ 4) 4 Γ(3 ⁄ 4) 2 π {Γ(1 ⁄ 4)} . dx = ( 1 ⁄ 4 ) y dy and thus ( –1 ⁄ 2 ) ∫0 Also. ∫0 13−26 π⁄2 cos 2m – 1 θ ⋅ sin 2n – 1 θ dθ = ------------------------- Γ ( m )Γ ( n ) 2Γ(m + n) Numerical Analysis Using MATLAB® and Excel®.Γ -⋅ -dy = -3 3 ⎝3⎠ 3 ∫0 c. so . Then.

= B ⎛ -⎝2 2⎠ Γ(1 ⁄ 2 + 1 ⁄ 2) Γ(1) 2 3x – x Numerical Analysis Using MATLAB® and Excel®. n – 1 = – 1 ⁄ 2 .Then. and the integral of (1) becomes 3 ------⋅3 3 –1 ⁄ 2 .= --------------------------.= ----------------------x(3 – x) –1 ⁄ 2 ∫0 x –1 ⁄ 2 –1 ⁄ 2 ⋅ (3 – x) –1 ⁄ 2 dx (1) = ( 3y ) = ( 3 ⁄ 3 )y Let x = 3y . x when x = 3 . When x = 0 . Third Edition Copyright © Orchard Publications 13−27 .= ---------2π tan θ dθ = ------------------------------------. 1 ----------------------------------------. ∫0 e. y = 1 .Solutions to End−of−Chapter Exercises Letting 2m – 1 = 1 ⁄ 2 and 2n – 1 = – 1 ⁄ 2 we obtain m = 3 ⁄ 4 and n = 1 ⁄ 4 . π⁄2 Γ ( 3 ⁄ 4 )Γ ( 1 ⁄ 4 ) 2 π . m = 1 ⁄ 2 . y = 0 and ∫0 1 y –1 ⁄ 2 ⋅ ( 3 – 3y ) –1 ⁄ 2 dy = = 3 1 ∫0 y 1 –1 ⁄ 2 –1 ⁄ 2 1 -(1 – y) ⋅ -----dy 3 –1 ⁄ 2 (2) ∫0 y –1 ⁄ 2 (1 – y) dy Recalling that B ( m.= π --------------------.= -------------2Γ(3 ⁄ 4 + 1 ⁄ 4 ) 2Γ(1) 2 3 3 ∫0 3 dx = --------------------2 3x – x ∫0 dx . m = 1 ⁄ 2 and thus ∫0 3 2 1 (1 ⁄ 2) ⋅ Γ(1 ⁄ 2) {Γ(1 ⁄ 2)} dx --⎞ = Γ -. n ) = ∫0 x 1 m–1 (1 – x) n–1 Γ(m) ⋅ Γ(n) dx = ----------------------------Γ(m + n) it follows that m – 1 = – 1 ⁄ 2 . then dx = 3dy .

1 Orthogonal Functions Orthogonal functions are those which are perpendicular to each other. orthogonal vectors. Figure 14.2 shows the angle between two curves C 1 and C 2 .2. Cholesky. QR.Chapter 14 Orthogonal Functions and Matrix Factorizations T his chapter is an introduction to orthogonal functions.1. Orthogonal curves Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications 14−1 .1. We begin with orthogonal lines and functions. This is shown in Figure 14. 14. Orthogonal lines Orthogonality applies also to curves. and we conclude with the factorization methods LU. From analytic geometry and elementary calculus we know that two lines are orthogonal if the product of their slopes is equal to minus one. orthogonal trajectories. Mutually orthogonal systems of curves and vectors are of particular importance in physical problems. y slope = m 1 slope = m 2 m1 ⋅ m2 = –1 x Figure 14. Figure 14. and Singular Value Decomposition.

= – -dx x (14.3) and (14. Third Edition Copyright © Orchard Publications .4) we see that these two curves are orthogonal since their slopes are negative reciprocals of each other. Example 14. The cases where x = 0 or y = 0 cannot occur because we defined a ≠ 0 and b ≠ 0 .2) the slope is 2xdx – 2ydy = 0 or dy -----.1).= x y dx (14. the angle between the curves C 1 and C 2 is the angle β between their tangent lines L 1 and L 2 . y ) on any curve of (14. The two families of these curves are shown in Figure 14. in Figure 14. 14.3) On any curve of (14. Other orthogonal functions are the cos x and sin x functions as we’ve learned in Chapter 6. L 1 and L 2 are orthogonal if m 2 = – 1 ⁄ m 1 . the slope is xdy + ydx = 0 or dy y ----. then.3.1). Thus.2) are orthogonal trajectories of the curves of (14.4) From (14.2. 14−2 Numerical Analysis Using MATLAB® and Excel®.1 Prove that every curve of the family xy = a a≠0 (14.1) is orthogonal to every curve of the family x –y = b 2 2 b≠0 (14. If m 1 and m 2 are the slopes of these two lines.2 Orthogonal Trajectories Two families of curves with the property that each member of either family cuts every member of the other family at right angles are said to be orthogonal trajectories of each other.2) Proof: At a point P ( x. the curves of (14.Chapter 14 Orthogonal Functions and Matrix Factorizations By definition.

= k ( cons tan t ) 2 2 2 ∫ ∫ x + 2y = C ( cons tan t ) 2 2 (14.6) From (14. the slope of the orthogonal family we are seeking must be xdy ----.5) dy ----.5) is c≠0 (14.x = 2y ----. Orthogonal trajectories Example 14.= 2cx dx (14.5).9) Numerical Analysis Using MATLAB® and Excel®.Orthogonal Trajectories Figure 14.= 2 -------2 dx x x 2 (14.6) as y dy .8) or 2ydy + xdx = 0 2 y dy + x dx = 0 2 y x . Third Edition Copyright © Orchard Publications 14−3 .= – ----2y dx (14.7) Therefore. c = y ⁄ x and thus we rewrite (14.+ ---2 ---.2 Find the orthogonal trajectories of the family of parabolas y = cx 2 Solution: The slope of (14.3.

Chapter 14 Orthogonal Functions and Matrix Factorizations Relation (14. Their inner (dot) product is defined as X ⋅ Y = x 1 y 1 + x 2 y 2 + x 3 y 3 + … + x n y n (a scalar ) (14. Orthogonal trajectories for families of parabolas and ellipses.9) represents a family of ellipses and the trajectories are shown in Figure14. Example 14.3 Given that X = [ 1 1 1 ] and Y = [ 2 1 2 ] find the dot product X ⋅ Y Solution: X ⋅ Y = (1) ⋅ (2) + (1) ⋅ (1) + (1) ⋅ (2) = 5 Definition:Two vectors X 1 and X 2 are said to be orthogonal if their dot product is zero. Third Edition Copyright © Orchard Publications .4. Figure 14. 14.10) Example 14.4. 14−4 Numerical Analysis Using MATLAB® and Excel®.3 Orthogonal Vectors Let X = [ x 1 x 2 x 3 … x n ] and Y = [ y 1 y 2 y 3 … y n ] be two vectors of the same length.4 Test the vectors X 1 = [ 1 1 1 ] and X 2 = [ 1 – 2 1] for orthogonality.

11) where A is the transpose of A and I is the identity matrix. 1 2 2 2 4 4 .-. Solution: First. If the column (or row) vectors of a square matrix A are mutually orthogonal unit vectors. we compute the magnitude X . the basis is called orthonormal basis. For this example. Third Edition Copyright © Orchard Publications 14−5 . Thus.-. If these vectors are also unit vectors.= -. X = 2 +4 +4 = 6 2 2 2 To compute the unit vector U X we divide each element of X by the magnitude X . This process is called normalization. the vectors X 1 and X 2 are orthogonal to each other. With any vector X ≠ 0 we may associate a unique unit vector U which is obtained by dividing each component of X by each magnitude X defined as X = x1 + x2 + … + xn 2 2 2 where x i represents an element of the vector X .Orthogonal Vectors Solution: X1 ⋅ X2 = ( 1 ) ⋅ ( 1 ) + ( 1 ) ⋅ ( –2 ) + ( 1 ) ⋅ ( 1 ) = 0 Therefore.-. the matrix A is orthogonal and A⋅A = I T (14. Example 14.5 Given that X = [2 4 4] compute the unit vector U X . T Numerical Analysis Using MATLAB® and Excel®.-U X = -2 3 3 6 6 6 A basis that consists of mutually orthogonal vectors is referred to as an orthogonal basis.

Third Edition Copyright © Orchard Publications . 14−6 Numerical Analysis Using MATLAB® and Excel®.11). To find a 2 × 2 square matrix Z such that we begin with z1 z2 –z1 z2 ⋅ Z⋅Z = I z1 –z1 z2 z2 2 2 T = 1 0 0 1 (14.6 Given that A = 1⁄2 1⁄4 1⁄2 1⁄2 find an orthonormal set of eigenvectors* and verify that the result satisfies (14. with these eigenvalues we can obtain an infinite number of eigenvectors. Solution: First.12) or z1 + z2 2 –z1 2 2 –z1 + z2 2 z1 + 2 z2 + 2 z2 = 1 0 0 1 (14.13) Equating like terms we obtain z1 + z2 = 1 2 2 2 2 –z1 + z2 = 0 2 2 2 From the second equation we obtain z 1 = z 2 and by substitution into the first we obtain 2z 1 = 1 or * It is strongly suggested that the reader reviews the definitions of eigenvalues and eigenvectors in Chapter 5 at this time. we find the eigenvalues of the matrix A from the relation det ( A – λ I ) = 0 where for this example ⎛ ⎞ det ⎜ 1 ⁄ 2 1 ⁄ 4 – λ 1 0 ⎟ = 0 ⎝ 1⁄2 1⁄2 0 1 ⎠ det 1 ⁄ 2 – λ 1⁄2 2 1⁄4 1⁄2–λ = 0 λ – λ + 3 ⁄ 16 = 0 from which λ 1 = 1 ⁄ 4 and λ 2 = 3 ⁄ 4 and as we’ve learned in Chapter 5.Chapter 14 Orthogonal Functions and Matrix Factorizations Example 14.

Third Edition Copyright © Orchard Publications 14−7 .The Gram-Schmidt Orthogonalization Procedure z1 = z2 = 1 ⁄ ( ± 2 ) This result indicates that we can choose either 1 ⁄ 2 or 1 ⁄ ( – 2 ) for the values of z 1 and z 2 .⋅ Y1 Y m = X m – -------------------------------Ym – 1 ⋅ Ym – 1 Y1 ⋅ Y1 (14.12) is Z = 1⁄ 2 1⁄ 2 1 ⁄ (– 2) 1 ⁄ 2 and as a check. 1 ⁄ 2 ⋅ 1 ⁄ 2 1 ⁄ (– 2) = 1 0 0 1 1 ⁄ (– 2) 1 ⁄ 2 1⁄ 2 1⁄ 2 1⁄ 2 The computations for finding orthonormal sets of eigenvectors for larger size ( 3 × 3 or higher) matrices using the above procedure becomes quite involved. … Y m using the following relations.⋅ Y1 Y 2 = X 2 – ----------------Y1 ⋅ Y1 Y1 ⋅ X3 Y2 ⋅ X3 .14) Also. 14. A simpler procedure is the GramSchmidt orthogonalization procedure which is discussed on the next section.14) below are the inner (dot) products and if X = [ x 1 x 2 x 3 … x n ] and Y = [ y 1 y 2 y 3 … y n ] are two vectors of the same length their dot product is defined as X ⋅ Y = x 1 y 1 + x 2 y 2 + x 3 y 3 + … + x n y n (a scalar ) .⋅ Y1 Y 3 = X 3 – ----------------Y2 ⋅ Y2 Y1 ⋅ Y1 … Ym – 1 ⋅ Xm Y1 ⋅ Xm . We must remember that the products in (14.⋅ Y m – 1 – … – -----------------.4 The Gram-Schmidt Orthogonalization Procedure Let X 1. the unit vectors Numerical Analysis Using MATLAB® and Excel®. X 2.14) the dot products on the numerator and denominator must be found first and the result must be from the dot product of it and Y 1 Y1 = X1 Y1 ⋅ X2 .We choose the value 1 ⁄ 2 and then the first (left most) matrix in (14. … X m be some column vectors. Y 2. Thus in the second equation in (14.⋅ Y 2 – ----------------. We can find an orthogonal basis Y 1.

A=[1/sqrt(3) 1/sqrt(6) −1/sqrt(2). Example 14.[ 1 1 1 ]T .15) are mutually orthogonal and form an orthonormal basis.= [ –1 ⁄ 2 0 1 ⁄ 2 ] U 3 = -------Y3 and denoting the matrix whose elements are the unit vectors as A .⋅ Y 2 – ----------------.⋅ Y1 = [ 1 – 2 1 ]T . X 2 = [ 1 – 2 1 ] . I=A*A' 14−8 Numerical Analysis Using MATLAB® and Excel®. 1/sqrt(3) −2/sqrt(6) 0. In our subsequent discussion the column vectors will be denoted as row vectors transposed.Chapter 14 Orthogonal Functions and Matrix Factorizations Yi U i = ------Yi i = 1. m (14.15) Y1 T U 1 = -------.= [1 ⁄ 3 1 ⁄ 3 1 ⁄ 3] Y1 Y2 T . 2. Solution: From (14.7 Given that X 1 = [ 1 1 1 ] .⋅ Y 2 – -. we have: 1⁄ 3 A = 1⁄ 3 1⁄ 3 T 1⁄ 6 –2 ⁄ 6 1⁄ 6 –1 ⁄ 2 0 1⁄ 2 We can verify that A ⋅ A = I with the MATLAB script below.⋅ Y 1 = [ 1 2 3 ] T – -Y 3 = X 3 – ----------------3 Y2 ⋅ Y2 Y1 ⋅ Y1 6 = [ 1 2 3 ] – [ 2 2 2 ] = [ –1 0 1 ] T T T and from (14. find an orthonormal basis. 1/sqrt(3) 1/sqrt(6) 1/sqrt(2)].⋅ Y1 = [ 1 – 2 1 ]T – 0 Y 2 = X 2 – ----------------3 Y1 ⋅ Y1 Y1 ⋅ X3 Y2 ⋅ X3 6 0 . … . Third Edition Copyright © Orchard Publications . and X 3 = [ 1 2 3 ] .= [ 1 ⁄ 6 –2 ⁄ 6 1 ⁄ 6 ] U 2 = -------Y2 Y3 T .14) Y1 = X1 = [ 1 1 1 ] T T T T Y1 ⋅ X2 -.

0000 0.0000 0 1. 1 2 3]. L 11 0 0 x1 x3 b1 b3 L 21 L 22 0 L 31 L 32 L 33 ⋅ x2 = b2 The unknowns are found from Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications 14−9 .The LU Factorization I = 1.4027 We observe that the vectors of the C matrix produced by MATLAB are different from those we derived with the Gram-Schmidt orthogonalization procedure.0000 0. Consider the following 3 × 3 lower triangular case.0000 14.0000 0 1.0000 -0. B=[1 1 1.0000 0.0000 0 0.0000 -0.4027 0.0000 1.0000 1.0000 -0.0000 0 0.9153 0.0000 -0. C=orth(B) C = -0.0000 0.0000 -0. As shown below. computers use the so-called matrix factorization methods to decompose a matrix A into a product of other smaller matrices. In Chapter 4 we saw how the method of Gaussian elimination proceeds by systematically removing the unknowns from a system of linear equations. 1 −2 1.0000 1. I=C*C' T I = 1. that is. we may find different values depending on the process being used.0000 -0.0000 We can also use the MATLAB function orth(A) to produce an orthonormal basis as shown below. The reason for this difference is that the orthogonalization process is not unique.5 The LU Factorization In matrix computations. the vectors produced by MATLAB also satisfy the condition C ⋅ C = I .9153 0. The LU factorization method decomposes a matrix A into a lower triangular matrix L and an upper triangular matrix U so that A = L ⋅ U .

we begin by multiplying the first equation by – 2 and subtracting it from the second equation. the unknowns are written in reverse order. Likewise. to solve U 11 U 12 U 13 0 0 U 33 x1 x3 b1 0 U 22 U 23 ⋅ x 2 = b 2 b3 (14. With these two reductions we obtain 2v 1 – v 2 + 3v 3 = 5 – 5v 2 + 4v 3 = 18 2.16) provided that L 11 ⋅ L 22 ⋅ L 33 ≠ 0 . we multiply the first equation by 3 ⁄ 2 and we subtract it from the third equation.5v 2 – 5.16) is referred to as forward substitution.20) 14−10 Numerical Analysis Using MATLAB® and Excel®. x 3 = b 3 ⁄ U 33 x 2 = ( b 2 – U 23 x 3 ) ⁄ U 22 x 1 = ( b 1 – U 12 x 2 – U 13 x 3 ) ⁄ U 11 (14.5 v 3 = – 3. 2v 1 – v 2 + 3v 3 = 5 – 4v 1 – 3v 2 – 2v 3 = 8 3v 1 + v 2 – v 3 = 4 (14. Thus.18) provided that U 11 ⋅ U 22 ⋅ U 33 ≠ 0 .5 (14. Third Edition Copyright © Orchard Publications . This removes v 1 from the second equation.17) we start from the bottom to the top as shown below.19) To find the three unknowns. The substitution order in (14. For the upper triangular case.18) is referred to as backward substitution.Chapter 14 Orthogonal Functions and Matrix Factorizations x 1 = b 1 ⁄ L 11 x 2 = ( b 2 – L 21 x 1 ) ⁄ L 22 x 3 = ( b 3 – L 31 x 1 – L 31 x 2 ) ⁄ L 33 (14. The substitution order in (14. Example 14.8 Let us review the example given in Chapter 4 which consists of the following equations.

20) by – 1 ⁄ 2 and we subtract it from the third equation of (14.21) We see that the eliminations have transformed the given square system into an equivalent upper triangular system that gives the same solution which is obtained as follows: v 3 = – 11 ⁄ 7 v 2 = ( 18 – 4v 3 ) ⁄ ( – 5 ) = – 34 ⁄ 7 v 1 = ( 5 + v 2 – 3v 3 ) ⁄ 2 = 17 ⁄ 7 The elements of the upper triangular matrix U are the coefficients of the unknowns in (14.20) and we obtain the system of equations below. Thus. let us use the relations of (14. 0 0 −3. To find the elements of matrix L we use MATLAB to multiply matrix A by the inverse of matrix U . Third Edition Copyright © Orchard Publications 14−11 . Thus.22) where the first matrix on the left side is the lower triangular matrix L and the second is the upper triangular matrix U . and v 3 in (14. L=A*inv(U) L = Numerical Analysis Using MATLAB® and Excel®.20). A=[2 −1 3. U=[2 −1 3.16) and (14. 3 1 −1].21). 2v 1 – v 2 + 3v 3 = 5 – 5v 2 + 4v 3 = 18 – 3.5 (14. −4 −3 −2.5 v 3 = 5.5 Now. U 11 U 12 U 13 0 U 22 U 23 0 0 U 33 2 –1 3 = 0 –5 4 0 0 – 3. 0 −5 4.The LU Factorization Next we multiply the second equation of (14. The elements of matrix U are the coefficients of v 1 .18) to find the lower and upper triangular matrices of our example where 2 –1 3 A = –4 –3 –2 3 1 –1 We want to find L ij and U ij such that L 11 0 0 ⋅ U 11 U 12 U 13 2 –1 3 0 U 22 U 23 = A = – 4 – 3 – 2 3 1 –1 0 0 U 33 L 21 L 22 0 L 31 L 32 L 33 (14.5]. v 2 .

5 Check with MATLAB: L=[1 0 0.9 Use the MATLAB function [L. Example 14. 0 −5 4. and L 32 = – 1 ⁄ 2 . the matrix A has been decomposed to a lower triangular matrix L and an upper matrix U as shown below.Chapter 14 Orthogonal Functions and Matrix Factorizations 1 -2 3/2 0 1 -1/2 0 0 1 Therefore. For a square matrix where none of the diagonal elements are zero. 0 0 −3. Was this necessary? The answer is no. −2 1 0. 2 –1 3 1 0 = –4 –3 –2 –2 1 3 1 –1 3 ⁄ 2 –1 ⁄ 2 0 2 –1 3 ⋅ 0 0 –5 4 1 0 0 – 3. 3/2 −1/2 1]. the lower triangular matrix has the form L = A⋅U 1 0 0 L = L 21 1 0 L 31 L 32 1 –1 –1 and in our example we found that the values of the subdiagonal elements are L 21 = – 2 . L 31 = 3 ⁄ 5 . A=L*U A = 2 -4 3 -1 -3 1 3 -2 -1 In the example above.U]=lu(A) to decompose the matrix 2 –3 1 A = –1 5 –2 3 –8 4 14−12 Numerical Analysis Using MATLAB® and Excel®. U=[2 −1 3. Third Edition Copyright © Orchard Publications . we found the elements of the lower triangular matrix L by first computing the inverse of the upper triangular matrix U and performing the matrix multiplication but not L = U ⋅ A . These values are the multipliers that we’ve used in the elimination process in succession.5].

that is.e. Then. help lu LU LU factorization.U] = lu(X) stores an upper triangular matrix in U and a "psychologically lower triangular matrix" (i. U=[3 −8 4. A1=L1*U A1 = 3 -1 2 -8 5 -3 4 -2 1 We observe that matrix A is now permuted.The LU Factorization into a lower and an upper triangular. 2/3 1 1]. 0 7/3 −2/3. Solution: format rat. 3 −8 4]. Let us now use MATLAB to see if L' ⋅ U = A . the lower triangular matrix L lacks structure. To put it in the given form we need to make the same interchanges in rows as with the lower triangular matrix. L1=[1 0 0. [L. When a matrix lacks structure we say that it is permuted. A=[2 −3 1. −1/3 1 0. let us interchange the first and third rows.U]=lu(A) L = 2/3 -1/3 1 U = 3 0 0 -8 7/3 0 4 -2/3 -1 1 1 0 1 0 0 We observe that while the upper triangular matrix U has the proper structure. let us invoke the help lu command. Only part of the display is shown below. To find out how MATLAB performs LU factorization. Third Edition Copyright © Orchard Publications 14−13 . [L. 1 0 L' = – 1 ⁄ 3 1 2⁄3 1 0 0 1 (14.23) The new matrix L' has now the proper structure. a product Numerical Analysis Using MATLAB® and Excel®. −1 5 −2. we must interchange the first and third rows. 0 0 −1]. To put L in the proper structure.

X can be rectangular. for example. 3 −8 4]. Third Edition Copyright © Orchard Publications . 0 0 1 P = 0 1 0 1 0 0 (14. and permutation matrix P so that P*X = L*U. [L. the identity matrix 1 0 0 I = 0 1 0 0 0 1 If we interchange the first and third rows of the identity matrix I above.9.P] = lu(X) returns unit lower triangular matrix L.9.U.U.P]=lu(A) to decompose the matrix 2 –3 1 A = –1 5 –2 3 –8 4 into a lower and an upper triangular and show that P ⋅ A = L ⋅ U . we obtain the permutation matrix P below.P]=lu(A) L = 1 -1/3 2/3 0 1 1 0 0 1 14−14 Numerical Analysis Using MATLAB® and Excel®. −1 5 −2. [L.U. Solution: This is the same matrix as in Example 14.Chapter 14 Orthogonal Functions and Matrix Factorizations of lower triangular and permutation matrices) in L.10 Use the MATLAB function [L. The permutation matrix P is an identity matrix that is permuted so that the rows of this matrix indicate the interchanges. upper triangular matrix U. Thus. Consider.24) and matrix P indicates the same interchanges as with the lower triangular matrix in Example 14. Example 14. so that X = L*U. A=[2 −3 1.

% This file is saved as ExchRows.. Also. X(j. Third Edition Copyright © Orchard Publications 14−15 ..:). define the matrix or vector % X and the indices i and j in MATLAB's Command Window % as X=[.:)..The LU Factorization U = 3 0 0 P = 0 0 1 0 1 0 1 0 0 -8 7/3 0 4 -2/3 -1 We observe that the lower triangular matrix has now the proper structure and the P matrix displayed by MATLAB is the same as in (14.:) = temp. PA=P*A. % The function ExchRows interchanges rows i and j % of a matrix or vector X % function X = ExchRows(X. LU=L*U PA = 3 -1 2 LU = 3 -1 2 -8 5 -3 4 -2 1 -8 5 -3 4 -2 1 We observe that P ⋅ A = L ⋅ U with the first and second rows interchanged when compare with the given matrix A . The MATLAB matrix left division operator x = A\b uses the L ⋅ U factorization approach. interchanges rows i and j j of a vector or matrix X. X(i.:) = X(j. i = {first row # to be interchanged}.].m % To run this program.i.24). The user−defined function ExchRows below. % j = {row # to be interchanged with row i}.j) % temp = X(i. and Numerical Analysis Using MATLAB® and Excel®.

let us explain the use of MATLAB’s built-in function max(v) where v is a row or a column vector. Solution: At the MATLAB command prompt we enter X=[−2 5 −4 9. ExchRows(X.m]=max(abs(v)) 14−16 Numerical Analysis Using MATLAB® and Excel®. let v=[2 −1 3 −5 7 −9 −12]'.m]=max(v) Amax = 7 m = 5 [Amax. First. j = 3. max(v) ans = 7 [Amax. As an example. and for matrices is a row vector containing the maximum element from each column. Third Edition Copyright © Orchard Publications . i = 1. Example 14.i. performs Gauss elimination with row pivoting.j) and MATLAB outputs X = 7 -3 -2 4 -5 -6 5 -9 3 8 -4 -8 2 1 9 -1 The user−defined function GaussElimPivot below. 4 −9 −8 −1].j)at the command prompt.Chapter 14 Orthogonal Functions and Matrix Factorizations % then type ExchRows(X.m user−defined function above to interchange rows 1 and 3. 7 −5 3 2. −3 −6 8 1.11 Given that the matrix X is defined as –2 5 X = –3 –6 7 –5 4 –9 –4 8 3 –8 9 1 2 –1 use the ExchRows.i.

k).k)). and b % is a column vector that contains the known values % on the right hand side. z = ExchRows(z. end n = length(b). A(i.k. % A is a matrix that contains the coefficients of % the system of equations.The LU Factorization Amax = 12 m = 7 % This user−defined function file solves A*x=b by % the Gauss elimination with row pivoting method.k.m] = max(abs(A(k:n. A = ExchRows(A. Third Edition Copyright © Orchard Publications 14−17 .k). z = zeros(n. % Set up scale factor array for i = 1:n.k+1:n) b(i) = b(i) − alpha*b(k). if Amag < eps. b=b'.k+1:n)*b(k+1:n))/A(k. z(i) = max(abs(A(i.m). error('Matrix is singular').m).m).k+1:n) − alpha*A(k.k) ~= 0 alpha = A(i.1:n)))./z(k:n)).k. m = m + k − 1.b) if size(b.1). x is a column vector that % will display the computed unknown values.k)/A(k. end if m ~= k b = ExchRows(b. function x = GaussElimPivot(A. end % The statements below exchange rows if required for k = 1:n−1 [Amag. end % Elimination steps for i = k+1:n if A(i. end Numerical Analysis Using MATLAB® and Excel®.2) > 1. end end end % Back substitution phase for k = n:−1:1 b(k) = (b(k) − A(k.k+1:n) = A(i.

1429 -8.0000 1.0000 9.0000 -3.12 Given that –2 5 A = –3 –6 7 –5 4 –9 –4 8 3 –8 x1 9 –3 1 .0000 -5.0000 -8.0000 9.0000 -5.0000 -8.0000 2.0000 A = 7.9298 -9.5714 -6.1429 -5.1429 3.b).7143 2.1429 -9.0000 -5.1429 3.2857 -3.8571 9.Chapter 14 Orthogonal Functions and Matrix Factorizations % Enter the values of A and b at the MATLAB's % command window and type GaussElimPivot(A.0000 -2.0000 2. x Example 14.8571 9. x and MATLAB outputs the following: A = 7.2857 -4.8571 9.2857 0. −3 −6 8 1. x = x2 x3 2 8 –1 5 x4 use the GaussElimPivot user−defined function above to compute the values of the vector x.1429 5. GaussElimPivot(A.0000 -2.5714 -2.b).0000 -2.0000 1.8571 10.0000 4. Solution: At the MATLAB command prompt we enter A=[−2 5 −4 9.0000 A = 7.0000 -3.1429 2.0000 -1. b = 2 .0000 -3.5714 -6.2857 -3.0000 9.0000 -8.0000 1.1429 3.0000 -8.0000 -3.0000 3.5714 -9.0000 A = 7. Third Edition Copyright © Orchard Publications .0000 -2.0000 4.0000 1.0000 -8. 7 −5 3 2.3860 -2.0000 -9.7143 3. 4 −9 −8 −1]. b = [−3 2 8 5]'.1429 14−18 Numerical Analysis Using MATLAB® and Excel®.0000 4.0000 9.0000 4.1429 3.5714 -1.0000 3.

k.0000 1. function [A.1429 -5.0000 4.5439 10.7193 0.8571 10. z(i) = max(abs(A(i.0000 -8. returns A = L*U % and the row permutation vector permut n = size(A. A = ExchRows(A.permut] = LUdecomp(A) % LU decomposition of matrix A. end % Exchange rows if necessary for k = 1:n−1 [Amag.9298 0.k.0000 A = 7.3860 -3.7451 -1.1429 -6. if Amag < eps error('Matrix is singular') end if m ~= k z = ExchRows(z.The LU Factorization A = 7.0000 -3.1176 2.1889 -0.1).0980 0.0000 x = 0. permut = (1:n)'.0000 -2.1176 -0.m).1:n))). performs LU decomposition.0000 -8.9298 -16./z(k:n)).5714 -1.m).1373 Check with MATLAB’s left division: x=b\A x = 0.m] = max(abs(A(k:n.2857 0. and returns matrix A as A=L*U and the row permutation vector permut.0980 3.7193 3. z = zeros(n.k)).8571 -3.5439 2. Numerical Analysis Using MATLAB® and Excel®.0000 -2. m = m + k − 1.5714 -6.0000 -3. for i = 1:n. Third Edition Copyright © Orchard Publications 14−19 .0000 4.7451 -5.2857 -16.0000 1.1429 3.0000 9.0000 9.1).1373 The user−defined function LUdecomp below.1429 3.

13 Given that –2 5 A = –3 –6 7 –5 4 –9 –4 8 3 –8 9 1 2 –1 use the LUdecomp user−defined function above to decompose matrix A and show how the given rows were permuted.8571 -3.m).2857 permut = 3 2 4 1 Check: -5.4386 3. end end end Example 14. −3 −6 8 1.0000 -0.0000 -8. A(i.0556 2.1429 0. [A.k) = alpha.k+1:n) − alpha*A(k. Solution: At the MATLAB command prompt we enter A=[−2 5 −4 9.0000 1.7193 -0.k+1:n) = A(i. 7 −5 3 2. 4 −9 −8 −1].5439 10.0000 9. end % Elimination pass for i = k+1:n if A(i.2857 -16.k).4286 0. A(i.7544 -0.Chapter 14 Orthogonal Functions and Matrix Factorizations permut = ExchRows(permut.1889 14−20 Numerical Analysis Using MATLAB® and Excel®.k)~=0 alpha = A(i.k.k)/A(k.5714 -0. Third Edition Copyright © Orchard Publications .k+1:n).permut] = LUdecomp(A) A = 7.

4386 3. b = b'.0612 0.0000 0 0 0 P= 1 0 0 0 L*U 0 1. % function x = LUsolPivot(A.4490 0 0 0 0 0 1 0 0 1.0000 -0.0000 -0.0000 1. % if size(b) > 1. for i = 1:n.0000 -8.0816 -0.0000 -0.1).0000 0.P]=lu(A) L = 1.U.2632 0 1 0 0 0 0 1 0 ans = 7.5714 -0.2857 -16.4347 10.1429 0. end n = size(A.0408 U = 7.9796 -3.5439 10.The LU Factorization [L. x = b. % In this user−defined function matrix A and column % vector b are entered in MATLAB's command window % and “permut” holds the row permutation data.0417 3.0556 2.b.1376 0.7193 -0.7544 -0.0761 -5.0000 2.6612 0 0 0 0 1.0000 9.1889 The user−defined function LUsolPivot listed below.4286 0.0000 1.0000 -8.0000 9.permut) % % The six statements below rearrange vector b and % stores it in vector x.m and will be used in the user−defined function matInvert that follows. x(i)= b(permut(i)).4694 -15. is saved as LUsolPivot.8571 -3.2857 -5. Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications 14−21 .

i).1).permut] = LUdecomp(A).k). 14−22 Numerical Analysis Using MATLAB® and Excel®.permut). and store results in Ainv % replacing the corresponding vector using the % user−defined function LUsolPivot saved previously.k+1:n)*x(k+1:n))/A(k.14 Invert the matrix A below with the user−defined function matInvert. % for i = 1:n Ainv(:. % This user−defined function inverts a matrix A % defined in MATLAB's command prompt using LU % decomposition % function Ainv = matInvert(A) % n = size(A. end The user−defined function matInvert below. inverts matrix A with LU decomposition.1:k−1)*x(1:k−1). Third Edition Copyright © Orchard Publications . % % The last three statements solve for each vector % on the right side.i) = LUsolPivot(A. % Creates identity matrix of size n. % Assigns the size of A to n. end for k = n:−1:1 x(k) = (x(k) − A(k. end Example 14.Ainv(:.Chapter 14 Orthogonal Functions and Matrix Factorizations end % % The next six statements perform forward and % backward substitution % for k = 2:n x(k) = x(k)− A(k. % The statement below performs LU decomposition % using the user−defined function LUdecomp(A) saved % previously % [A. % Ainv = eye(n).

7 −5 3 2.0416 -0.0719 0.1174 0.0447 0.0078 0.1174 0.0783 -0. 4 −9 −8 −1]. −3 −6 8 1.6 The Cholesky Factorization A matrix is said to be positive definite if x ⋅A⋅x>0 T T (14.0683 -0.26) Relation (14. ans = -0. It is a special case of LU factorization and requires fewer computations than the LU factorization method of the previous section.0610 0.0447 0.0610 0. Under those conditions. Third Edition Copyright © Orchard Publications 14−23 .0369 0.0208 0.0201 -0. Ainv = matInvert(A) and MATLAB returns Ainv = -0.0055 Check with MATLAB’s built−in inv(A) function.0389 0. there exists an upper triangular matrix G with positive diagonal elements such that G ⋅G = A T (14.0201 -0.26) is referred to as Cholesky factorization.0242 -0.0078 0. A = A .0416 -0. that is.0055 14.0208 0.0242 -0.0783 -0.0389 0.0981 -0. Let us invoke the MATLAB help chol command to see how MATLAB performs this factorization.0981 inv(A) -0.0719 0.0013 -0.25) for every x ≠ 0 and A is symmetric.The Cholesky Factorization –2 5 A = –3 –6 7 –5 4 –9 –4 8 3 –8 9 1 2 –1 Solution: In MATLAB’s command prompt we enter A=[−2 5 −4 9. Numerical Analysis Using MATLAB® and Excel®.0369 0.0683 -0.0013 -0.

If X is positive definite. EYE(M. K = 0 is the main diagonal. Example m = 5. then p is 0 and R is the same as above. DIAG(V) is the same as DIAG(V. RANDN. If X is not positive definite. TRIL.N) or EYE([M. 14−24 Numerical Analysis Using MATLAB® and Excel®. We will consider an example using the Cholesky factorization after we review the MATLAB functions eye(n) and diag(v. with two output arguments. Third Edition Copyright © Orchard Publications . TRIU. DIAG(X) is the main diagonal of X. See also ONES.Chapter 14 Orthogonal Functions and Matrix Factorizations CHOL Cholesky factorization. help eye EYE Identity matrix. EYE(N) is the N-by-N identity matrix. then p is a positive integer.0) and puts V on the main diagonal. CHOL(X) uses only the diagonal and upper triangle of X.N]) is an M-by-N matrix with 1's on the diagonal and zeros elsewhere.K) when V is a vector with N components is a square matrix of order N+ABS(K) with the elements of V on the K-th diagonal. DIAG(DIAG(X)) is a diagonal matrix. The lower triangular is assumed to be the (complex conjugate) transpose of the upper. See also SPDIAGS.k) as defined by MATLAB. diag(-m:m) + diag(ones(2*m. then R = CHOL(X) produces an upper triangular R so that R'*R = X. If X is positive definite. ZEROS. K > 0 is above the main diagonal and K < 0 is below the main diagonal. EYE(SIZE(A)) is the same size as A. never produces an error message.p] = CHOL(X). an error message is printed.1) + diag(ones(2*m. help diag DIAG Diagonal matrices and diagonals of a matrix. DIAG(V.-1) produces a tridiagonal matrix of order 2*m+1. RAND.1).1). But if X is not positive definite.K) when X is a matrix is a column vector formed from the elements of the K-th diagonal of X. [R. DIAG(X.

00 0 0 0 -0.The QR Factorization Example 14.25 −0.25 0 -0. A=5*eye(5)+diag(B.7 The QR Factorization The QR factorization decomposes a matrix A into the product of an orthonormal matrix and an upper triangular matrix.15 Use MATLAB to compute the Cholesky factorization of matrix A as defined below.R]=qr(A) produces an n × n matrix whose columns form an orthonormal or unitary* matrix Q and an upper triangular matrix R of the same size as matrix A . the matrix product G ⋅ G = A is satisfied. 14.75 −1.45 2.00 5. In other words.22 2. A1=G'*G Solution: Execution of the MATLAB script above displays the following: A = 5.25 5.75 5.00 -1.50 5. A can be factored as A = QR (14.75 0 T 0 0 -0.19 0 0 0 -1. B=[−0.11 2.21 0 0 0 -0.00 -1. G=chol(A).25 0 0 0 -0.00 -0.00 -0.23 0 0 0 -0.00 We observe that A1 = A . −1)+diag(B.24 0 0 0 0 A1 = 5.00 -0.22 0 0 0 -0.25 5. format bank.75 5.00].00 0 0 0 -1.34 2.00 5. The MATLAB function [Q.75 0 0 0 G = 2.27) * An n × n matrix A is called unitary if ( A∗ ) = A T –1 where A∗ is the complex conjugate matrix of A .50 −0. that is. Numerical Analysis Using MATLAB® and Excel®.50 0 -0.50 5. 1).00 0 0 -0.00 -0.00 -0.50 0 0 0 -0. Third Edition Copyright © Orchard Publications 14−25 .00 0 0 -0.

By itself.28) and multiplying both sides of (14. The column permutation E is chosen so that abs(diag(R)) is decreasing.R]=qr(A) is described as follows: help qr QR Orthogonal-triangular decomposition. TRIU(QR(A)) is R.R] = QR(A. [Q.R. a system described by Ax = b becomes QRx = b (14.28) by Q ⋅ Q = I we obtain Rx = Q b T T (14. [Q.R] = QR(A) produces an upper triangular matrix R of the same dimension as A and a unitary matrix Q so that A = Q*R.0) and [C. Third Edition Copyright © Orchard Publications .R. The full version of QR does not return C. returns C = Q'*B. The column permutation E is chosen so that abs(diag(R)) is decreasing. R = QR(A) returns only R. If A is m-by-n with m > n.0) produces an "economy size" decomposition in which E is a permutation vector. but Q is often nearly full. QR(A) is the output of LAPACK'S DGEQRF or ZGEQRF routine.R] = QR(A) returns both Q and R.0) produces the "economy size" decomposition.Chapter 14 Orthogonal Functions and Matrix Factorizations Then. 14−26 Numerical Analysis Using MATLAB® and Excel®. The least squares approximate solution to A*x = b can be found with the Q-less QR decomposition and one step of iterative refinement: x r e x = = = = R\(R'\(A'*b)) b .0) produce economy size results.E] = QR(A) produces a permutation matrix E. so that Q*R = A(:. where B has as many rows as A.E] = QR(A.R] = QR(A. an upper triangular R and a unitary Q so that A*E = Q*R.A*x R\(R'\(A'*r)) x + e. Note that R = chol(A'*A). Q.B).R] = QR(A.B.E). [Q. then only the first n columns of Q are computed. [Q. C.29) The MATLAB [Q. R = QR(A.

74 0 0 x = 4.28 1.00 Check=A\b -0.00 Let us verify that the matrix Q is unitary. −1 5 −2.43 7. Solution: A=[2 −3 1.R]=qr(A).00 -0. [Q.58 -4.62 -0.58 Check = 4.00 -0.14 4.24 0 -0. b=[2 4 5]'.16 Given that 2 –3 1 2 A = – 1 5 – 2 and b = 4 3 –8 4 5 solve Ax = b using the MATLAB function [Q.35 -3. x=R\Q'*b T Q = -0. since the elements are real numbers.77 0. Of course.00 1.43 7.00 T –1 or Q ⋅ Q = I .00 0.14 4. T Numerical Analysis Using MATLAB® and Excel®. 3 −8 4]. the complex conjugate of Q is also Q and thus we only need to show that Q = Q Q*Q' ans = 1.53 0.58 0.80 R = -3.00 -0.15 9.00 1.54 0.00 -0.The QR Factorization Example 14.00 0. Third Edition Copyright © Orchard Publications 14−27 .R]=qr(A) and x = R\Q b .27 -0.58 0.

V]=svd(A).4286 7. of the same dimension as A and with nonnegative diagonal elements in decreasing order. so we select the vector X which produces the minimum error. then pinv(A)=inv(A).1429 4.8 Singular Value Decomposition The Singular Value Decomposition (SVD) method decomposes a matrix A into a diagonal matrix S . MATLAB does this with either the left division operator ( \ ) or with the non−negative least−squares function lsqnonneg(A. and unitary matrices U and V so that A = U⋅S⋅V T (14.0000 Underdetermined systems have infinite solutions and MATLAB selects one but no warning message is displayed.30) The matrices U .b). −1 5 −2. if A is square. Of course. This function returns the vector X that minimizes norm(A*X−b) subject to X ≥ 0 provided that the elements of A and b are real numbers. there is no vector X which can satisfy the entire system of equations. X=lsqnonneg(A. systems with more equations than unknowns as in applications where we need to find the least square distance in linear regression. and V .b) returns X = 4. decomposed from a given matrix A . As we’ve learned in Chapter 4. can be found with the MATLAB function [U. b=[2 4 5]'. The function pinv(A) displays the pseudoobtaininverse of a m × n (non−square) matrix A . 3 −8 4]. In an overdetermined system. For example. 14.* that is.17 Decompose the matrix * We defined overdetermined and underdetermined systems in Chapter 8 14−28 Numerical Analysis Using MATLAB® and Excel®. the MATLAB function inv(A) produces the inverse of the square matrix A and an error message is displayed if A is not a square matrix. S. S .Chapter 14 Orthogonal Functions and Matrix Factorizations QR factorization is normally used to solve overdetermined systems. Example 14. A=[2 −3 1. Third Edition Copyright © Orchard Publications .

0863 0.0000 V*V' -0.0000 0.0000 -0.5675 0 0 0.8050 -0.2122 -0.4605 0 0 V = -0.Singular Value Decomposition 2 –3 1 A = –1 5 –2 3 –8 4 into two unitary matrices and a diagonal matrix with non−negative elements. We also verify that the matrices U and V are unitary as shown below.0000 1.8929 As expected.0000 0.4420 0.S.0000 -0.0000 -0.3150 0.5184 0. the diagonal elements of the triangular S matrix are non−negative and in decreasing values. S.0000 -0.3116 0. A=[2 −3 1.0000 1.0000 0. −1 5 −2.5924 -0. [U. V]=svd(A) function.0000 1. 3 −8 4].0325 0 1.0000 0.5028 0.8632 -0. Third Edition Copyright © Orchard Publications 14−29 .0000 ans = 1.6521 0.3972 -0.V]=svd(A) U = -0. Solution: We will use the MATLAB [U.0000 -0.1782 0 -0.2440 0.9463 -0.0000 1.0000 -0.4731 -0.0000 Numerical Analysis Using MATLAB® and Excel®.0000 -0. U*U' ans = 1.8228 S = 11.

is defined as X = x 1 + x 2 + … + x n . • A matrix is said to be positive definite if x ⋅ A ⋅ x > 0 for every x ≠ 0 and A is symmetric. Third Edition Copyright © Orchard Publications . the basis is called orthonormal basis.R]=qr(A) produces an n × n matrix whose columns form an orthonormal or unitary matrix Q and an upper triangular matrix R of the same size as matrix A .Chapter 14 Orthogonal Functions and Matrix Factorizations 14. that is. A simpler procedure is the Gram−Schmidt orthogonalization procedure which we will discuss on the next section. the matrix A is said to be orthogonal and A ⋅ A = I where A is the transpose of A and I is the identity matrix. denoted as X . The MATLAB function [Q.U]=lu(A) decomposes the matrix A into a lower triangular matrix L and an upper triangular matrix U . Under those conditions. there exists an upper triangular matrix G with positive diagonal elements such that G ⋅ G = A . these vector are said to be orthogonal to each other. A unique 2 2 2 unit vector U is obtained by dividing each component of X by the magnitude X and this process is referred to as normalization. This process is referred to as the Cholesky factorization. • If the column (or row) vectors of a square matrix A are mutually orthogonal unit vectors. The MATLAB function [L. • The QR factorization decomposes a matrix A into the product of an orthonormal matrix and an upper triangular matrix. A = A . • The magnitude of a vector X . • The LU factorization method decomposes a matrix A into a lower triangular matrix L and an upper triangular matrix U so that A = L ⋅ U . • We can find an orthonormal set of eigenvectors in a 2 × 2 matrix easily from the eigenvalues but the computations for finding orthonormal sets of eigenvectors for larger size ( 3 × 3 or higher) matrices using the above procedure becomes quite involved. • Two families of curves with the property that each member of either family cuts every member of the other family at right angles are said to be orthogonal trajectories of each other. T T T T T 14−30 Numerical Analysis Using MATLAB® and Excel®. • A basis that consists of mutually orthogonal vectors is referred to as an orthogonal basis. • The inner (dot) product of two vectors X = [ x 1 x 2 x 3 … x n ] and Y = [ y 1 y 2 y 3 … y n ] is a scalar defined as X ⋅ Y = x 1 y 1 + x 2 y 2 + x 3 y 3 + … + x n y n • If the dot product of two vectors X 1 and X 2 is zero. If these vectors are also unit vectors.9 Summary • Orthogonal functions are those which are perpendicular to each other.

and V . and unitary matrices U and V so that A = U ⋅ S ⋅ V . of the same dimension as A and with nonnegative diagonal elements in decreasing order. decomposed from a given matrix A . T Numerical Analysis Using MATLAB® and Excel®. S.Summary • The Singular Value Decomposition (SVD) method decomposes a matrix A into a diagonal matrix S . V]=svd(A). S . Third Edition Copyright © Orchard Publications 14−31 . can be found with the MATLAB function [U. The matrices U .

Chapter 14 Orthogonal Functions and Matrix Factorizations 14. 7. Given the vectors X 1 = [ 2 – 1 ] and X 2 = [ 1 – 3 ] .80 4. 6.00 – 0.k) to define and display the matrix A shown below.00 0 0 – 1. 14−32 Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications .00 4. use the Gram−Schmidt orthogonalization procedure to find two vectors Y 1 and Y 2 to form an orthonormal basis.10 Exercises 1.00 0 – 1.00 – 1.8 to decompose the system of equations x 1 + 2x 2 + 3x 3 = 14 – 2x 1 + 3x 2 + 2x 3 = 10 5x 1 – 8 x 2 + 6x 3 = 7 T T 2 2 2 2 3 into an upper triangular matrix U and a lower triangular matrix L . Show that the curve x + 3y = k 1 and the curve 3y = k 2 x where k 1 and k 2 are constants. use the MATLAB Cholesky factorization function to obtain the matrix G and verify that G ⋅ G = A . Use a suitable function to verify your results. 5. Use the Gaussian elimination method as in Example 14.80 0 A = – 0. 4. 2.20 4. 4. Verify your answers with MATLAB. Find the orthogonal trajectories of the curves of the family 2x + y = kx 3. Using the MATLAB functions eye(n) and diag(v. Use MATLAB to find another set of an orthonormal basis with the vectors given in Exercise 3. Use the appropriate MATLAB function to decompose the system of equations A = 1 0 –1 3 and b = 0 1 2 5 –2 –3 4 9 T into an upper triangular matrix R of the same dimension as A and a unitary matrix Q so that Q ⋅ R = A .20 0 0 – 1. are orthogonal to each other.00 Then.

T Numerical Analysis Using MATLAB® and Excel®. Use the appropriate MATLAB function to decompose the matrix A given as A = 1 0 –1 0 1 2 –2 –3 4 into a diagonal matrix S of the same dimension as A and with non−negative diagonal elements in decreasing order and unitary matrices U and V so that U ⋅ S ⋅ V = A .Exercises 8. Third Edition Copyright © Orchard Publications 14−33 .

14−34 Numerical Analysis Using MATLAB® and Excel®.( kx ) ----. 2x + y = kx (1) 2 2 Implicit differentiation of (1) yields d 2 d d . x + 3y = k 1 (1) 3y = k 2 x 3 2 2 (2) Implicit differentiation of (1) yields dy . 3y k 2 = ----3 x and by substitution into (4) we obtain 3y dy -.( 2x 2 ) + ----dx dx dx dy . 2.Chapter 14 Orthogonal Functions and Matrix Factorizations 14.⋅ x2 = 3 y ----.y = ----.= ----3 x dx x We observe that (5) is the negative reciprocal of (3) and thus the given curves are orthogonal to each other.= 0 2x + 6y ----dx or 1 dy -. Third Edition Copyright © Orchard Publications .= k 2 x (4) dx 3 2 From (2).= -------------.(3) -⋅x ----.= k 4x + 2y ----dx and solving for dy ⁄ dx .11 Solutions to End−of−Chapter Exercises 1.(2) dx 2y From (1).= – -3 y dx Differentiation of (2) yields 3k 2 x 2 dy ----. dy k – 4x ----.= ------------.(5) .

(3) k = ------------------x 2 2 and by substitution into (2) 2x + y ------------------. we need to find the curves whose slopes are given by the negative reciprocal of (4). Then. dy = udx + xdu and by substitution into (6) ( 2x – u x ) ( udx + xdu ) = 2x udx 2 2 2 2 Division of both sides of the above by x yields ( 2 – u ) ( udx + xdu ) = 2udx 2 2 Collecting like terms and simplifying we obtain x ( 2 – u ) du = u dx Separating the variables we obtain 2 (2 – u ) dx .du – du --------.+ ----.du ----.Solutions to End−of−Chapter Exercises 2x + y .= ---3 x u u and by integrating these terms we find –1 -+C ln x + ln u = ----2 u By substitution of u = y ⁄ x we obtain Numerical Analysis Using MATLAB® and Excel®.= -------------------------------.= -----------------------dx 2y 2xy 2 2 Now.= ------------------2 2 dx 2x – y We rewrite (5) as ( 2x – y ) dy = 2xydx (6) 2 2 and we let y = ux . Third Edition Copyright © Orchard Publications 14−35 .(4) ----. we need to find the family of the curves of 2xy . that is.= -----------------3 3 u x u u 2 2 3 or 2 dx du .– 4x 2 2 dy x – 2x + y .du = ---.(5) dy ----.

B=orth(A) B = -0.5257 14−36 Numerical Analysis Using MATLAB® and Excel®. A=[2 −1. From (14.⋅ [ 2 – 1 ] T = [ 1 – 3 ] T – -.⋅ C⎞ = ln Cy = – x ⋅ -ln x + ln y -------.= [ –1 ⁄ 5 U 2 = -------Y2 – 1 ⁄ 5] T –2 ⁄ 5 ] T and denoting the matrix whose elements are the unit vectors as A .8507 I=B*B' -0.= [2 ⁄ 5 U 1 = -------Y1 Y2 .Chapter 14 Orthogonal Functions and Matrix Factorizations 2 y x.8507 0.14) Y1 = X1 = [ 2 –1 ] T T T Y1 ⋅ X2 [ 2 –1 ] ⋅ [ 1 –3 ] . 2⁄ 5 –1 ⁄ 5 –1 ⁄ 5 ⋅ 2 ⁄ 5 –2 ⁄ 5 –1 ⁄ 5 –1 ⁄ 5 = –2 ⁄ 5 4⁄5+1⁄5 –2 ⁄ 5 + 2 ⁄ 5 –2 ⁄ 5 + 2 ⁄ 5 = 1 0 1⁄5+4⁄5 0 1 4. we have: A = T 2⁄ 5 –1 ⁄ 5 –1 ⁄ 5 –2 ⁄ 5 We verify that A ⋅ A = I as shown below. 1 −3].5257 -0.⋅ [ 2 – 1 ] T Y 2 = X 2 – ----------------T T Y1 ⋅ Y1 [ 2 –1 ] ⋅ [ 2 –1 ] 5 T [2 + 3] .15) Y1 .+ ln C = ln ⎛ ⎝ x ⎠ 2 x y and thus the family of curves orthogonal to the given family is x = – y ln Cy 2 2 3.⋅ Y 1 = [ 1 – 3 ] T – -----------------------------------------------. Third Edition Copyright © Orchard Publications .⋅ [ 2 –1 ]T = = [ 1 – 3 ] – ---------------[4 + 1] 5 = [ 1 –3 ] – [ 2 –1 ] = [ –1 2 ] T T T and from (14.

Therefore. after simplification x 1 + 2x 2 + 3x 3 = 14 7x 2 + 8x 3 = 38 ( 81 ⁄ 7 ) x 3 = 243 ⁄ 7 (3) Thus. we multiply the second equation in (2) by – 18 ⁄ 7 and we subtract it from the third in (2). the lower triangular matrix is 1 0 L = –2 1 5 – 18 ⁄ 7 0 0 1 The elements of the upper triangular matrix are the coefficients of the unknowns in (3) and thus U = 1 2 3 0 7 8 0 0 81 ⁄ 7 Numerical Analysis Using MATLAB® and Excel®. 5 .0000 1. x 1 + 2x 2 + 3x 3 = 14 7x 2 + 8x 3 = 38 – 18 x 2 – 9 x 3 = – 63 (2) Next. 1. x 3 = 243 ⁄ 81 = 3 x 2 = ( 38 – 8x 3 ) ⁄ 7 = 2 x 1 = ( 14 – 3x 3 – 2x 2 ) = 1 The multipliers that we’ve used are – 2 . and L 32 respectively. we multiply the first equation by 5 and we subtract it from the third in (1). L 31 . and – 18 ⁄ 7 .0000 x 1 + 2x 2 + 3x 3 = 14 – 2x 1 + 3x 2 + 2x 3 = 10 (1) 5x 1 – 8 x 2 + 6x 3 = 7 Multiplying the first equation of (1) by – 2 and subtracting it from the second in (1) we obtain the second equation in (2) and thus x 1 is eliminated. Then.Solutions to End−of−Chapter Exercises I = 5. Likewise.0000 -0. These are the elements L 21 .0000 -0. Third Edition Copyright © Orchard Publications 14−37 . Then.

0 1 2. B=[−0. 0 0 81/7].51 1.5345 0. 0 7 8. 5 −18/7 1].3586 0.20 0 -0.00 -1. A=L*U A = 1 -2 5 6.8944 R = -2.00 -1.00 -1. 2 3 -8 3 2 6 format bank.6833 -1.20 4.20 0 0 -1.00 Q = -0.00 0 0 -1. x=R\Q'*b. 1).96 0 0 -0. −1)+diag(B. A1=G'*G A = 4. A=4*eye(4)+diag(B.00 4.0 −1.62 1.5976 0.7171 -0. A=[1 0 −1.1381 0. Check=A\b -0. b=[3 5 9]'. [Q.00 0 -0. Third Edition Copyright © Orchard Publications .80 4.8 −1. U=[1 2 3.80 0 0 G = 2.40 1.2]. G=chol(A).2673 14−38 Numerical Analysis Using MATLAB® and Excel®. QQT=Q*Q'.2361 0 0 -2. we use MATLAB to verify that L ⋅ U = A L=[1 0 0.90 0 0 -1.00 -0.6733 0 4.80 4.Chapter 14 Orthogonal Functions and Matrix Factorizations Now.00 4.00 0 0 -0.4472 0 0.4781 2.0249 -0.20 4.93 0 0 -1.00 -0.00 -1.80 0 0 7. −2 1 0.R]=qr(A).8018 0. −2 -3 4].00 0 0 0 A1 = 4.

0000 0.0000 1.1574 0 0 0.0000 1. A=[1 0 −1.7922 UUT = 1.Solutions to End−of−Chapter Exercises QQT = 1.0000 Numerical Analysis Using MATLAB® and Excel®.7000 -0.0000 0.0000 -0.0000 -0.V]=svd(A).0000 1. UUT=U*U'.0000 0.0000 1. VVT=V*V' 0.0000 0. [U.0000 0.1977 0.0000 -0.7500 -2.3795 -0.2493 -0.0000 -0.5985 0 0 V = -0.2410 0 2.0000 1.9577 S = 5.0000 -0.0000 0.0000 0.2093 0.9154 0.0000 x = 6.0000 U = -0.0000 0.5000 3.0000 0.5583 0.0000 0.4779 0.9480 -0.0000 0.5000 3.0000 -0.0000 1. 0 1 2.9556 -0.S.2076 0.0413 0 0.3184 -0. −2 −3 4].2464 0. Third Edition Copyright © Orchard Publications 14−39 .7500 -2.0000 VVT = 1.8187 0.1345 0.7500 8.7500 Check = 6.

It is beyond the scope of this book to derive the infinite series which are approximations to the solutions of these differential equations. and telephone equations. positive or negative integer. Then.Chapter 15 Bessel. Applying the method of Frobenius to (15. acoustics.2). cannot be solved in terms of familiar functions as those which we encountered in ordinary differential equations with constant coefficients. we will accept the solutions without proof. under certain conditions. these are discussed in advanced mathematics textbooks. theory of elasticity and others. They are solutions of differential equations with variable coefficients and. we replace n with – n .1). transmission lines. they are used in frequency modulation. denoted as J n ( x ) . are used in engineering.1). Therefore.2) This series is referred to as Bessel function of order n where n is any positive real number or zero. and the most common are the Method of Frobenius and the Method of Picard. These are special functions that find wide applications in science and engineering. If in (15. 15. The usual procedure is to derive solutions in the form of infinite series. Differential equations with variable coefficients. thermodynamics. Bessel functions are solutions of the differential equation x2 dy dx 2 2 +x dy 2 2 + (x – n )y = 0 dx (15.1) depends on the value of n .1) where n can be any number. we obtain the relation Numerical Analysis Using MATLAB® and Excel®. aeronautics. fractional. such as (15. Third Edition Copyright © Orchard Publications 15−1 . satisfy the orthogonality principle. in the electrical engineering field. we obtain the infinite power series Jn ( x ) = ∞ k=0 ∑ x⎞ k ( – 1 ) ⋅ ⎛ -⎝2⎠ n + 2k 1 ⋅ --------------------------------------k! ⋅ Γ ( n + k + 1 ) n≥0 (15. the form of the general solution of (15. or even a complex number. For instance. and Chebyshev Functions T his chapter is an introduction to some very interesting functions. Legendre.1 The Bessel Function The Bessel functions.

7). that when n is zero or even. Legendre.3) and the function J –n ( x ) is referred to as the Bessel function of negative order n . J n ( x ) is an even function of x .10) Also. If we differentiate the series of J 0 ( x ) in (15.+ --------------------. … (15.6) reduces to the following series: x x x x J 0 ( x ) = 1 – --------------------. and compare with the series of J 1 ( x ) in (15.+ ----------------------.2) reduces to Jn ( x ) = or k=0 ∑ ∞ x ⎞ n + 2k 1 k ⋅ --------------------------( – 1 ) ⋅ ⎛ -⎝2⎠ k ! ⋅ ( n + k )! n = 0.+ ----------------------.– --------------------.+ -------------------------–… 2 4 6 8 10 2 ⋅ 2! 2 ⋅ 1! ⋅ 3! 2 ⋅ 2! ⋅ 4! 2 ⋅ 3! ⋅ 5! 2 ⋅ 4! ⋅ 6! 2 4 6 8 10 3 5 7 9 2 4 6 8 (15. For the special case where n is a positive integer or zero.– ----------------------.5) n 2 4 x ⎧ x x 1 – --------------------------------------------------------------------------------------J n ( x ) = -------------+ ⎨ n 2 4 2 ⋅ n! ⎩ 2 ⋅ 1! ⋅ ( n + 1 ) 2 ⋅ 2! ⋅ ( n + 1 ) ( n + 2 ) 6 ⎫ x – ----------------------------------------------------------------------. 1 and 2 .8). we see that d J ( x ) = –J1 ( x ) dx 0 (15.9) We observe from (15. 1.4) and (15. 2. Third Edition Copyright © Orchard Publications .8) (15. and Chebyshev Functions J–n ( x ) = k=0 ∑ ( –1 ) ∞ k x ⎞ – n + 2k 1 ⋅ ⎛ -⋅ ------------------------------------------⎝2⎠ k! ⋅ Γ ( – n + k + 1 ) (15.+ … ⎬ 6 ⎭ 2 ⋅ 3! ⋅ ( n + 1 ) ( n + 2 ) ( n + 3 ) (15.Chapter 15 Bessel.6) For n = 0. Γ ( n + k + 1 ) = ( n + k )! (15.+ ---------------------–… 2 2 4 2 6 2 8 2 2 ⋅ ( 1! ) 2 ⋅ ( 2! ) 2 ⋅ ( 3! ) 2 ⋅ ( 4! ) x x x x x J 1 ( x ) = -.7) through (15. we will find that 15−2 Numerical Analysis Using MATLAB® and Excel®. (15.9).+ -----------------------–… 3 5 7 9 2 2 ⋅ 1! ⋅ 2! 2 ⋅ 2! ⋅ 3! 2 ⋅ 3! ⋅ 4! 2 ⋅ 4! ⋅ 5! x x x x x J 2 ( x ) = -------------. and odd when n is odd.– ----------------------. if we multiply the series for J 1 ( x ) by x and differentiate it.– ----------------------.– ----------------------.7) (15.

besselj(2.2239 b.1149 Numerical Analysis Using MATLAB® and Excel®.2240 1 . J 2 ( 1 ) Solution: a. we obtain besselj(0.9). With MATLAB. From (15.– -----------------.+ --------------------------. From (15. J 1 ( 3 ) c.3391 ans = 0.3391 c.– 27 J1 ( 3 ) = 3 2 16 384 18432 1474560 or from math tables.3). J 2 ( 1 ) = 0.= 0. J 0 ( 2 ) = 0.= 1 64 .2).– ----------------.+ -------256 .2239 ans = 0.11) Example 15.+ 243 -.+ ----------.1149 . besselj(1. We can use the MATLAB besselj(n.x) function or the Excel BESSELJ(x.– --------. 43. 19683 .The Bessel Function d { xJ 1 ( x ) } = xJ 0 ( x ) dx (15. correct to four decimal places.8). the values of a.– ----. J 1 ( 3 ) = 0.n) function for the above computations.1 Compute.= 0.– -----------------.= -------1.= 0. From (15. Third Edition Copyright © Orchard Publications 15−3 .+ 16 J0 ( 2 ) = 1 – 4 192 4 36 576 4 64 64 × 36 256 × 576 or from math tables.7).1152 J 2 ( 1 ) = -8 96 1536 184320 17694720 or from math tables. J 0 ( 2 ) b. 1 1 1 1 1 .+ ----------------------.+ -----------------------.3400 2187.1) ans = 0.

all J n ( x ) are infinite series and thus. text(2.z). J 1 ( x ) . tables of some of the roots of J 0 ( x ) and J 1 ( x ) are shown in math tables.5 J2(x) Jn(x) 0 -0. it is a very difficult and tedious task to compute all roots of these series. first sideband. 'J0(x)'). respectively. besselj(2. 0.x). represent the carrier. z = besselj(2. 0. Legendre. Fortunately.95. 0.Chapter 15 Bessel.00.00: 0.25. The functions J 0 ( β ) .1)= 0.x). The x −axis crossings in the plot of Figures 15. x = 0. title('Bessel Functions of the First Kind').2 show the first few roots of the J 0 ( x ) .2. 'J2(x)') The plots for J 0 ( x ) .1149 The MATLAB script below plots J 0 ( x ) .3391 besselj(1.x).x.x. and J 2 ( x ) series.v.85. 'J1(x)'). * In Frequency Modulation (FM).35. and Chebyshev Functions and with Excel. ylabel('Jn(x)'). text(0. J 2 ( β ) and so on.0)= 0. Third Edition Copyright © Orchard Publications .1. grid. x is denoted as β and it is called modulation index.1 and 15.05: 10.* 1 J0(x) Bessel Functions of the First Kind J1(x) 0.60. second sideband etc. J 1 ( x ) . Plots of J 0 ( x ) . J 1 ( x ) and J 2 ( x ) are shown in Figure 15. text(4. J 1 ( β ) .2)= 0.5 0 1 2 3 4 5 x 6 7 8 9 10 Figure 15. J 1 ( x ) and J 2 ( x ) using MATLAB We can also use Excel to plot these series as shown in Figure 15. plot(x.2239 besselj(3.w. 15−4 Numerical Analysis Using MATLAB® and Excel®. and J 2 ( x ) . xlabel('x').20.1. w = besselj(1. However. v = besselj(0. The definition of a Bessel function of the first kind will be explained shortly.

55 0. The most noteworthy are: 1.9120 0. They have no complex roots 2. that is.12) Numerical Analysis Using MATLAB® and Excel®.60 0.0050 0. J 0 ( x ) and J 1 ( x ) . 1 and 2 x 0. the so−called half−order Bessel functions J 1 ⁄ 2 ( x ) and J ( – 1 ⁄ 2 ) ( x ) .8971 0.2647 0.0249 0. and so on.70 0. 4.30 0.0250 0.The Bessel Function Plot of Bessel Function Jn(x) for n = 0.45 0.9844 0. for example.9500 0.35 0.5201 of J 0 ( x ) .50 0. 3 ⁄ 2 .0000 0.20 0. 5 ⁄ 2 .0012 0. For instance.9900 0.75 J0(x) 1.0112 0.0499 0.8812 0. then J n ( x ) can be expressed in a finite form of sines and cosines.2. the roots of these equations separate each other.5 Jn(x) J2(x) 0.0588 0.15 0.0437 0.1 which shows the first 5 positive roots of these equations.0306 0. are almost periodic with period almost 2 π . This is observed on Table 15.10 0.8642 J1(x) 0.25 0.0000 0.1483 0.0995 0. J 1 ( x ) and J 2 ( x ) using Excel The equations J 0 ( x ) = 0 and J 1 ( x ) = 0 exhibit some interesting characteristics.1960 0.1240 0. If we let n = 1 ⁄ 2 in (15. Each has an infinite number of distinct real roots 3. the difference between consecutive roots approaches the value of π. we obtain J1 ⁄ 2 ( x ) = 2. such as 1 ⁄ 2 .2194 0.3492 J2(x) 0.5 0 2 4 x 6 8 10 Figure 15.2867 0. Consider.9994 0.0 -0. As the roots become larger and larger.3290 0.9776 0.0510 0.1723 0.65 0.0 J0(x) J1(x) 0. Between two consecutive roots of one of these equations lies one and only one root of the other equation.8317 of J 1 ( x ) lies between the roots 2.2423 0.0748 0. these series behave like the cos x and sin x functions.9604 0. and the differences between consecutive roots.3081 0.0078 0. In other words.4048 and 5.9696 0.sin x ----πx (15.9975 0.9258 0.0000 0.05 0.00 0. Plots of J 0 ( x ) . Third Edition Copyright © Orchard Publications 15−5 .0152 0.9385 0.0671 Bessel Functions of the First Kind 1.0003 0.2).9944 0. If n is half of an odd integer. that is.0197 0.40 0. we observe that the first root 3.0028 0.0369 0.

the values of --⎞ a.8 ⋅ -----.1394 14.4706 13.8 ⋅ -----.4048 3. π -⎞ = J ( – 1 ⁄ 2 ) ⎛ -⎝4⎠ 2 1 2 2 cos ( π ⁄ 4 ) = -.9309 16. besselj(−0.1378 11. J 1 ⁄ 2 ⎛ ⎝ ⎠ π 4 --⎞ b.3).5201 3.1502 7.1839 Differences … … … … Likewise.13) Example 15.1 The first few roots of J 0 ( x ) and J 1 ( x ) J0 ( x ) = 0 Roots 2.1735 3. if we let n = 1 ⁄ 2 in (15.sin ( π ⁄ 4 ) = -. Third Edition Copyright © Orchard Publications .pi/4) 15−6 Numerical Analysis Using MATLAB® and Excel®.6366 -----------------2 π π π(π ⁄ 4) Check with MATLAB: besselj(0.cos x πx (15.5.13).1579 J1 ( x ) = 0 Differences Roots 3.Chapter 15 Bessel. π -⎞ = J 1 ⁄ 2 ⎛ -⎝4⎠ 2 1 2 4.7915 3.12).= 0.1469 10.8317 3.0156 3.1153 5.6537 3.3237 3.= -.6366 ----------------2 π π 2π π(π ⁄ 4) b. Legendre.= ----. Using (15. Using (15.= -2 . and Chebyshev Functions TABLE 15. we obtain J( –1 ⁄ 2 ) ( x ) = 2 ----.= 0. J ( –1 ⁄ 2 ) ⎛ ⎝ ⎠ π 4 Solution: a.2 Compute.5.pi/4).1336 8. correct to four decimal places.

are referred to as Bessel functions of the first kind.The Bessel Function ans = 0. certain differential equations resemble the Bessel equation. and will be explained in the next paragraph. that the numbers n = 0. Also. we see that m=0 ∑ ∞ ⋅ (x ⁄ 2 ) ( –1 ) ------------------------------------------------------------------( n + m )! ⋅ Γ ( m + 1 ) = ( –1 ) n n+m – n + 2n + 2m m=0 ∑ ∞ ⎛x --⎞ ⎝2⎠ n + 2m ( –1 ) -------------------------------------------. we will prove that J–n ( x ) = ( – 1 ) Jn ( x ) for n = 1. a Bessel function J –n ( x ) for n > 0 . if we let k = n + m in the second summation. Other Bessel functions. Third Edition Copyright © Orchard Publications 15−7 . and thus their solutions are called Modified Bessel functions. or Weber functions. then. – 1. the first summation in the above relation is zero for k = 0. we recall from Chapter 13.2). 2.14) Proof: From (15.= ( –1 )n Jn ( x ) Γ ( n + m + 1 )! ⋅ m! m and thus (15. … n (15. … yield infinite values in Γ ( n ) . after simplification and comparison with (15. or Hankel functions. Also. If n is an integer. 3.6366 The Bessel functions which we have discussed thus far.15) Now. As mentioned earlier. 1.6366 ans = 0. 2.5). Numerical Analysis Using MATLAB® and Excel®. – 2. These are additional solutions of the Bessel’s equation.14) has been proved. J –n ( x ) = = k=0 n–1 k=0 ∑ ∑ ∞ ( –1 ) ⋅ ( x ⁄ 2 ) -----------------------------------------------k! ⋅ Γ ( – n + k + 1 ) k k – n + 2k ( –1 ) ⋅ ( x ⁄ 2 ) ------------------------------------------------+ k! ⋅ Γ ( – n + k + 1 ) – n + 2k k=n ∑ ∞ ( –1 ) ⋅ ( x ⁄ 2 ) -----------------------------------------------k! ⋅ Γ ( – n + k + 1 ) k – n + 2k (15. or Neumann functions. can be obtained by replacing n with – n in (15.3). denoted as Y n ( x ) and referred to as Bessel functions of the second kind. n – 1 . ….

sin x + B ----.+ C 2 ----------x x (15.17) We will show that the general solution of (15.y = 0 dx ⎝ 4⎠ (15. Third Edition Copyright © Orchard Publications . we obtain 2 2 y = A ----. … (15. 15−8 Numerical Analysis Using MATLAB® and Excel®. Example 15. the general solution of the Bessel equation is y = AJ n ( x ) + BJ – n ( x ) n ≠ 0.19) can be written as sin x cos x y = C 1 ---------. the functions J n ( x ) and J –n ( x ) are not linearly independent as we have seen in (15. relation (15. (15. for this case.17). 3.13) into (15. … and so on.1).17) is y = AJ 1 ⁄ 2 ( x ) + BJ –1 ⁄ 2 ( x ) (15.20) Since the two terms on the right side of (15. 2. 1. we obtain x 2 dy dx 2 2 +x dy ⎛ 2 1 ⎞ + x – -. The following example illustrates the fact that when n is not an integer or zero. and Chebyshev Functions It is shown in advanced mathematics textbooks that.20) are linearly independent. therefore.3 Find the general solution of Bessel’s equation of order 1 ⁄ 2 . Solution: By the substitution n = 1 ⁄ 2 in (15.16) is not the general solution.16) For n = 1. that is. if n is not an integer.18) By substitution of (15.cos x πx πx (15. (15. 3.19) and letting C 1 = A 2 ⁄ π and C 2 = B 2 ⁄ π . the Bessel functions of the second kind are introduced to obtain the general solution. for this case.Chapter 15 Bessel. y represents the general solution of (15.14). 2. Legendre. and for this reason. J n ( x ) and J –n ( x ) are linearly independent.12) and (15.18). these two series produce only one solution.16) is the general solution.

The function e x⎛ -.23) cos ( x sin φ ) = J 0 ( x ) + 2 sin ( x sin φ ) = 2 ∞ k=1 ∑ J2k ( x ) cos 2k φ (15. when the average of their cross product is zero within some specified limits. ∑ Jn ( x ) n = –∞ 2 ∞ = 1 ∞ (15. Also.* For instance. we can obtain several interesting properties for integer values of n . they constitute an orthogonal system. The descriptions.21) and we say that J 0 ( ax ) and J 0 ( bx ) are orthogonal in the interval 0 ≤ x ≤ 1 . ∫0 xJ0 ( ax ) J0 ( bx ) dx onal with the variable x . ∫0 π ⎧ π J n ( x ) n = even cos n φ ⋅ cos ( x sin φ ) dφ = ⎨ n = odd ⎩ 0 n = even ⎧0 sin n φ ⋅ sin ( x sin φ ) dφ = ⎨ ⎩ π J n ( x ) n = odd ∫0 π (15. Some of these are given below without proof. Using this function. then. They are also orthog- = ∑ Jn ( x ) t n = –∞ ∞ n (15.22) is referred to as the generating function for Bessel functions of the first kind of integer order. can be evaluated at specified values either with MATLAB or Excel.The Bessel Function The Bessel functions of the second kind. Numerical Analysis Using MATLAB® and Excel®.24) J 2k – 1 ( x ) sin ( 2k – 1 ) φ ∑ k=1 where the subscript 2k denotes that the first relation is valid for even values of k . More detailed discussion and proofs can be found in advanced mathematics textbooks. syntax. third kind. and examples for each can be found by invoking help bessel for MATLAB. One very important property of the Bessel’s functions is that within certain limits. and help for Excel. Third Edition Copyright © Orchard Publications 15−9 . if a and b are distinct roots of J 0 ( x ) = 0 . indicates that the second relation is valid for odd values of k .t–1 --⎞ 2⎝ t⎠ 1 = 0 (15. and others.25) * Two functions constitute an orthogonal system. J 0 ( a ) = 0 and J 0 ( b ) = 0 . whereas 2k – 1 in the second.

Applying the method of Frobenius. we obtain two independent solutions y 1 and y 2 as follows. then (15. Here.30) and this series is absolutely convergent* for x > 1 .x + --------------------------------------------------------------------x – … y 2 = a 1 x – --------------------------------3! 5! (15.26) appear in frequency modulation.x + --------------------------------------------------------------------x – … + a 1 x – --------------------------------3! 5! (15. n( n + 1) (n – 2)n(n + 1)(n + 3) 4 .x 2 + -----------------------------------------------------x –… y 1 = a 0 1 – ------------------2! 4! ( n – 1 )( n + 2) 3 ( n – 3)( n – 1)( n + 2)( n + 4 ) 5 . and if it is zero or a positive integer. and Chebyshev Functions and 1 J n ( x ) = -π π ∫0 cos ( n φ – x sin φ ) dφ ∑ Jn ( β ) n = –∞ 2 ∞ (15. Third Edition Copyright © Orchard Publications . while y 2 is an odd function. For example.+ n(n + 1 )y = 0 – 2x ----( 1 – x ) --------2 dx dx 2 (15. as in the Bessel equation.x 2 + -----------------------------------------------------x –… y = a 0 1 – ------------------2! 4! (n – 1)(n + 2) 3 (n – 3)(n – 1)(n + 2)(n + 4) 5 .27) known as Legendre’s equation.2 Legendre Functions Another second−order differential equation with variable coefficients is the equation dy 2 d y .29) where a 0 and a 1 are constants. it reduces to 1 2 -A P ave = -2 C 15.23) through (15.Chapter 15 Bessel. n is a constant.27) has polynomial solutions of special interest. Then.28) (15. We observe that y 1 is an even function of x . the average power is shown to be 1 2 -A P ave = -2 C and with (15. the general solution of (15.26) Relations (15. Legendre.23). 15−10 Numerical Analysis Using MATLAB® and Excel®.27) is y = y 1 + y 2 or n( n + 1) (n – 2)n(n + 1)(n + 3) 4 .

whenever n is zero or a positive integer.33) From (15. The Legendre polynomials are defined as Pn ( x ) = ( –1 ) n⁄2 n(n + 1) 2 1 ⋅ 3 ⋅ 5 ⋅ … ⋅ (n – 1) x +… ⋅ ------------------------------------------------.Legendre Functions The parameter n is usually a positive integer. the general solution of Legendre’s equation contains a polynomial solution which is denoted as P n ( x ) . If. they will not be discussed in this text.31) for n = 0 or n = even integer Pn ( x ) = ( –1 ) (n – 1) ⁄ 2 (n – 1)(n + 2) 3 1⋅3⋅5⋅…⋅n -x + … ⋅ ------------------------------------------------. the second term contains only a finite number of terms. or (15. If n is zero. if it is odd. when we replace the terms of this series by their absolute value. The infinite series solution Q n ( x ) is referred to as Legendre functions of the second kind.x – --------------------------------3! 2 ⋅ 4 ⋅ 6 ⋅ … ⋅ (n – 1) for n = odd integer (15. Third Edition Copyright © Orchard Publications 15−11 . i.33).32) and these are also referred to as surface zonal harmonics. The even and odd functions of (15. These become infinite as x → ± 1 and their applications to science and engineering problems are very limited.e.1 – ------------------2! 2⋅4⋅6⋅…⋅n (15.. this series is said to be absolutely convergent. reaches a limit. the first term on the right side of (15. and an infinite series solution which is denoted as Q n ( x ) . ∑ un ( x0 ) n=1 ∞ = u 1 ( x 0 ) + u 2 ( x 0 ) + … also Numerical Analysis Using MATLAB® and Excel®. or an even positive integer.for n = even and N = ( for n = odd where N = -2 2 (15.n – 2k --------------------------------------------------x n 2 k! ( n – k )! ( n – 2k )! k n – 1) n ---------------. we find that the resulting series converges. we obtain the following first 6 Legendre polynomials. Accordingly. * Assume that the infinite series ∑ un ( x0 ) n=1 ∞ = u 1 ( x 0 ) + u 2 ( x 0 ) + … converges.32) can be combined to a single relation as Pn ( x ) = k=0 ∑ N ( – 1 ) ⋅ ( 2n – 2k )! .31) and (15.30) contains only a finite number of terms.32). Therefore.31) and (15.

34) The relation n n 1 d .37) as 2 ⋅ n! ⋅ P n ( x ) and thus (15.4 Find the potential difference (voltage) v at a point P developed by a nearby dipole* in terms of the distance between the point P and the dipole. Legendre. separated by a small distance. (x – 1) = 2 n k=0 n ∑ ( – 1 ) ⋅ n! 2n – 2k ---------------------------x k ! ⋅ ( n – k )! k (15.= P0 ( x ) + P1 ( x ) t + P2 ( x ) t + … + Pn ( x ) t = 2 1 – 2xt + t ∑ Pn ( x ) t n=0 ∞ n (15.( 35x 4 – 30x 2 + 3 ) P 4 ( x ) = -8 P1 ( x ) = x 1 . From the binomial theorem. Another important identity involving Legendre polynomials. * A dipole is a pair of electric charges or magnetic poles. by comparison with (15.36) and differentiation of (15. and Chebyshev Functions P0 ( x ) = 1 1 .35) is proved.35) is known as Rodrigues’ formula. and offers another method of expressing the Legendre polynomials. Third Edition Copyright © Orchard Publications .( 3x 2 – 1 ) P 2 ( x ) = -2 1 .( 5x 3 – 3x ) P 3 ( x ) = -2 1 . is the generating function for Legendre polynomials which is defined as 2 n 1 ------------------------------. we recognize (15.( 63x 5 – 70x 3 + 15x ) P 5 ( x ) = -8 (15.38) We will illustrate the use of the Legendre polynomials with the following example. usually fed from the center.⋅ -------. and the angle which point P makes with the center of the dipole.36) with respect to x n times yields n d ------. consisting of two equal rods extending outward in a straight line. a dipole is an antenna. of equal magnitude but of opposite sign or polarity.33).Chapter 15 Bessel.( x2 – 1 ) = n dx n k=0 ∑ N ( – 1 ) ⋅ n! ( 2n – 2k )! n – 2k -x ----------------------. 15−12 Numerical Analysis Using MATLAB® and Excel®.37) Now.( x2 – 1 ) P n ( x ) = -------------n n 2 ⋅ n! dx (15. Alternately. We prove (15. Example 15.⋅ -----------------------k! ( n – k )! ( n – 2k )! n k (15.35) as follows.

we will denote the quantity 1 ⁄ ( 4 πε 0 ) as k and thus we rewrite (15. Third Edition Copyright © Orchard Publications 15−13 .40) Next. By the law of cosines. Figure for Example 15. From electromagnetic field textbooks we find that q ⎛ 1 1⎞ .– ---⎞ = kq ( r 2 – r 1 ) ⎝r r ⎠ 2 1 (15.3. we need to express r 1 and r 2 in terms or d and r.4 Let the potential at point P be V P .41) (15.43) or * Permittivity is a measure of the ability of a material to resist the formation of an electric field within it.Legendre Functions Solution: Let the charges q and – q of the dipole be a distance 2d apart with the origin 0 as the midpoint as shown in Figure 15.--V P = ----------.= r d 2d cos θ ----.3.39) where ε 0 is the permittivity* of the vacuum.– --4 πε 0 ⎝ r 2 r 1 ⎠ (15. r1 = d + r – 2 dr cos ( 180 ° – θ ) = r2 = 2 2 2 2 d + r + 2dr cos θ 2 2 (15. For simplicity.+ 1 – -----------------2 r r 2 (15. P r1 r θ −q d 0 d q r2 Figure 15.42) and d + r – 2 dr cos θ Dividing both sides of (15.39) as –1 –1 1 1 V P = kq ⎛ --. we obtain r2 --.42) by r . Numerical Analysis Using MATLAB® and Excel®.

We do this by expressing these terms in the form of the generating function of (15. (15.48). Now.Chapter 15 Bessel. the point P is sufficiently far from the origin. Then. we can express (15.+ 1 – ----------------r ⎝ r2 r ⎠ –1 ⁄ 2 (15. and y = d ⁄ r . This requirement is satisfied since x and y.44) In all practical applications. we want to relate the terms inside the parentheses of (15.48) However. we assume that r > d .47) as kq V P = ----r ∑ n=0 ∞ d ⎞n [ P n ( cos θ ) – P n ( – cos θ ) ] ⎛ -⎝r ⎠ (15.50) 15−14 Numerical Analysis Using MATLAB® and Excel®.44) we obtain ⎛ d2 ⎞ cos θ + 1 – 2d -----------------⎟ ⎜ ----2 r ⎠ ⎝ r –1 ⁄ 2 = ( 1 – 2xy + y ) 2 –1 ⁄ 2 = ∑ Pn ( x ) y n=0 ∞ n (15. To find a similar expression for r 1 . and therefore.38).49) and for r » d . then.44).45) holds only if x < 1 and y < 1 . we obtain kq V P = ----r ∑ [ Pn ( x ) y n=0 ∞ – Pn ( –x ) y ] (15.46) into (15. and Chebyshev Functions ⎛ d2 ⎞ 1 –1 2d cos θ . and thus ( 1 + 2xy + y ) 2 –1 ⁄ 2 –1 = ∑ Pn ( –x ) y n=0 n n ∞ n (15. are both less than unity.46) By substitution of (15. as defined. P n ( – cos θ ) = – P n ( cos θ ) and the odd powers in (15.45) We recall that (15. all even powers vanish.45). We let x = cos θ .cos θ * V P ≈ -------2 ⎝r ⎠ r 1 r (15. kq V P = ----r ∞ n=0 ∑ d ⎞ 2n + 1 2P 2n + 1 ( cos θ ) ⎛ -⎝r ⎠ (15. to a Legendre polynomial. if n is even in (15.⎜ -----⎟ r 2 = -.P ( cos θ ) ⎛ -. Third Edition Copyright © Orchard Publications . we simply replace x with – x in (15. thus.45) and (15.47) Since x = cos θ . by substitution into (15.49) can be approximated as d⎞ 2kdq 2kq . and y = d ⁄ r . P n ( – cos θ ) = P n ( cos θ ) .48) are duplicated.= -----------.40). Legendre. But when n is odd.

is 2n + 1 B n = -------------2 ∫– 1 f ( x ) P n ( x ) d x 1 (15. be derived without the use of Legendre polynomials as follows: For r » d . and using (15.40) we obtain kq 2kqd 1 1 .– --. the distances r 1 .57) when we discuss the trigonometric Numerical Analysis Using MATLAB® and Excel®. = cos θ .( r 2 – r 1 ) = -----------V P = kq ⎛ --.50).⎞ = -----------2 ⎝r r ⎠ r1 ⋅ r2 r 2 1 (15.4. It is a vector directed from the negative charge towards the positive charge.52) is the same as (15. Derivation of the voltage developed by a dipole Another interesting relation that can be used to find the Legendre polynomial series of a function f ( x ) for x < 1 . Then. Third Edition Copyright © Orchard Publications 15−15 . we have used the identity P1 ( cos θ ) form of the Legendre polynomials. p = 2q d (15.cos θ . that is.4. and r 2 can be approximated by parallel lines as shown in Figure 15.50) can.Legendre Functions The term 2dq is the magnitude of the so−called dipole moment.51) The relation of (15.52) We observe that (15. It is denoted with the letter p . r. This will be seen shortly in (15. of course.53) * Here. the negative and positive charges look like a single point charge. P r1 × r2 = r 2 r2 r 2 – r 1 = 2d cos θ r r1 θ θ −q d 0 d q Figure 15.

54). f (x) 1 −1 0 −1 1 x Figure 15.( 3x 2 – 1 ) dx + -( – 1 ) ⋅ -2 2 –1 0 . For this example.= 0 --+1 2 2 1 0 3 ( – 1 ) ⋅ x ⋅ dx + -2 –1 3 2 -x 1 ⋅ x ⋅ dx = – -4 1 3 2 -x + -–1 4 1 2 3 3 3 = -. and Chebyshev Functions Then.( 5x 3 – 3x ) dx + 7 -( – 1 ) ⋅ -2 2 –1 0 .54) The example below illustrates how this relation is being used.Chapter 15 Bessel. Third Edition Copyright © Orchard Publications .( 35x 4 – 30x 2 + 3 ) dx + -( – 1 ) ⋅ -8 2 –1 0 .34) for the polynomials of P n ( x ) . Example 15. we will substitute these into (15.( 5x ∫0 1 ⋅ -2 1 1 3 -– 3x ) dx = – 7 8 4 ∫ 1 9 . then.5. Waveform for Example 15.( 3x ∫0 1 ⋅ -2 1 – 1 ) dx = 0 ∫ 1 .53). We will also use (15.( 35x ∫0 1 ⋅ -8 1 – 30x + 3 ) = 0 2 15−16 Numerical Analysis Using MATLAB® and Excel®.5 Compute the Legendre polynomial series representing the waveform of Figure 15.= -4 4 2 5 B 2 = -2 7 B 3 = -2 9 B 4 = -2 ∫ 1 5 .5 Solution: We will first compute the coefficients B n from (15.5. Legendre. 1 B 0 = -2 3 B 1 = -2 ∫ ∫ 1 ( – 1 ) ⋅ 1 ⋅ dx + -2 –1 0 0 ∫0 1 ⋅ 1 ⋅ d x ∫0 1 1 1 -x = – -2 0 –1 0 1 -x + -2 1 0 = –1 -.+ -. a function f ( x ) can be expanded as f ( x ) = B0 P0 ( x ) + B1 P1 ( x ) + B2 P2 ( x ) + … + Bn Pn ( x ) = Bn Pn ( x ) ∑ n=0 ∞ (15.

as we did in Example 15. the algebraic form of the Legendre polynomials is usually the most useful. However.54) and (15.Legendre Functions – 11 B 5 = -------2 ∫ 1 11 -. P 0 cos θ = 1 P 1 cos θ = cos θ 3 cos 2 θ + 1 P 2 cos θ = --------------------------4 5 cos 3 θ + 3 cos θ P 3 cos θ = --------------------------------------8 35 cos 4 θ + 20 cos 2 θ + 9 P 4 cos θ = ---------------------------------------------------------64 cos 5 θ + 35 cos 3 θ + 30 cos θ ------------------------------------------------------------------------P 5 cos θ = 63 128 (15.2 cos ( n – 4 ) θ + … + --------2 ⋅ 4 2 ⋅ 4 ⋅ … ⋅ ( 2n – 4 ) (15. Also.x – -= -8 2 16 8 2 7113 3 5 3 .x + --------x . its expansion contains only odd Legendre polynomials.( 63x 5 – 70x 3 + 15x ) . using (15.4. In many applications. there are times when we want to express the polynomials in terms trigonometric functions.34) we obtain 3 -P (x) – 7 -. Therefore. the trigonometric forms are most convenient with the cylindrical and spherical coordinate systems.57) The Legendre polynomials in algebraic form.56) From (15.⋅ -.x – -------= -------64 128 128 (15. Third Edition Copyright © Orchard Publications 15−17 .( 63x 5 – 70x 3 + 15x ) dx + ----8 2 –1 0 . It is shown in advanced mathematics textbooks that ⋅ 3 ⋅ … ⋅ ( 2n – 3 ) 1 ⋅ 3 ⋅ … ⋅ ( 2n – 1 ) 1 -------------------------------------------. Numerical Analysis Using MATLAB® and Excel®.P ( x ) + 11 -----P (x) f ( x ) = -2 1 8 3 16 5 7 1 11 1 3 .x – ----= -( 5x – 3x ) + -------⋅ ( 63x – 70x + 15x ) 16 128 2 525 3 693 5 525 .55) We observe that the waveform of f ( x ) is an odd function and.⋅ -------------------------------------------.( 5x 3 – 3x ) + ----. as we found above.⋅ -.( 63x ∫0 -8 1 1 5 3 ----– 70x + 15x ) = 11 16 and so on.56) we obtain the first 6 Legendre polynomials in trigonometric form listed below.2 cos ( n – 2 ) θ -⋅1 2 cos n θ + -P n cos θ = -------------------------------------------2 ⋅ 4 ⋅ … ⋅ 2n 2 2 ⋅ 4 ⋅ … ⋅ ( 2n – 2 ) 1 ⋅ 3 1 ⋅ 3 ⋅ … ⋅ ( 2n – 5 ) . satisfy the orthogonality principle when m ≠ n as indicated by the following integral.

58) Similarly. the Legendre polynomials in trigonometric form satisfy the orthogonality principle when m ≠ n as indicated by the following integral.61) where C 1 and C 2 are arbitrary constants. Third Edition Copyright © Orchard Publications .60) reduces to (15.+ n ( n + 1 ) – ------------– 2x ( 1 – x ) --------.62) and (15. ∫0 π ⎧ 0 ⎪ P m ( cos θ ) P n ( cos θ ) sin θ dθ = ⎨ 2 ⎪ -------------⎩ 2n + 1 m≠n m = n (15.35). There is another class of Legendre functions which are solutions of the differential equation 2 dy 2 d y m ----.59) We must remember that all the Legendre polynomials we have discussed thus far are referred to as surface zonal harmonics.63) are also known as spherical harmonics. The general solution of (15. The functions P n ( x ) and Q n ( x ) are referred to as associated Legendre functions of the first and second kind respectively.60) and this is referred to as the associated Legendre differential equation.60) is y = C1 Pn ( x ) + C2 Qn ( x ) m m m m (15. We observe that if m = 0 .y = 0 2 2 dx dx 1–x 2 (15.Chapter 15 Bessel.63) Relations (15. that is. These are evaluated from Pn ( x ) = ( –1 ) ( 1 – x ) m m 2 m⁄2 d P (x) ⋅ --------m n dx d Q (x) ⋅ --------m n dx m m (15. the polynomials P n ( x ) . and Chebyshev Functions ⎧ 0 ⎪ P m ( x ) P n ( x ) dx = ⎨ 2 –1 ⎪ -------------⎩ 2n + 1 1 ∫ m≠n m = n (15.27). We will restrict our subsequent discussion to the associated Legendre functions of the first kind.62) and Qn ( x ) = ( –1 ) ( 1 – x ) m m 2 m⁄2 (15. (15. m 15−18 Numerical Analysis Using MATLAB® and Excel®. Legendre. and math tables include values of these as computed from Rodrigues’ formula of (15.

x) function that computes the associated Legendre functions of the first kind of degree n . legendre(2. P 2 ( x ) x = 0.34). m = 1 and m = 2 are: Numerical Analysis Using MATLAB® and Excel®. Check with MATLAB: disp('The values for m = 0.5) The values for m = 0. Example 15. 2. Third Edition Copyright © Orchard Publications 15−19 . m = 1 . ….5 As stated above. n = 2 and thus MATLAB will return a matrix whose rows correspond to the values of m = 0 . we obtain P2 ( x ) 1 x = 0. P 3 ( x ) x = – 0. the MATLAB legendre(n.x) function computes the associated Legendre functions of the first kind of degree n and order m = 0. n evaluated for each element of x .5 2 1⁄2 d. n evaluated for each element of x .⎛ 3x –1 --------------------⎞ dx ⎝ 2 ⎠ 2 x = 0. a. and order m = 0. a. 1. 2.62). that is. P2 ( x ) 1 x = 0. second. Pn ( x ) = ( –1 ) ( 1 – x ) m m 2 m⁄2 1 x = 0. and m = 2 .0. 1. Excel does not have any functions related to Legendre polynomials.5 2 1⁄2 = ( –1 ) ( 1 – x ) = –( 1 – x ) 2 1⁄2 1 d----P (x) dx 2 = –( 1 – x ) x = 0.5 c.5 ( 3x ) = – 1. and third rows respectively.5 2 3 b. for the first. we use the relation (15.25 d P (x) ⋅ --------m n dx m and the appropriate relations of (15.6 Find the following associated Legendre functions and evaluate as indicated.62). P 2 ( x ) Solution: For this example.Legendre Functions At present.5 3x – 1⎞ = ⎛ ---------------⎝ 2 ⎠ 2 = – 0.2990 For m = 0 in (15. MATLAB provides the legendre(n. Here. m = 1 and m = 2 are:'). For this example. ….125 x = 0.

5 x = – 0.1)=m'. n = 3 . fprintf('\n'). y=zeros(4. m 0 1 2 3 Legendre 0.4f \n'.2).6250 -9.0f\t %7.2990 2. y=zeros(3.6250 Here.2)=legendre(2. Check with MATLAB: m=0:3.2).2500 or more elegantly.5 2 d ⎛ 5x – 3x⎞ . y(:.1250 -1.25 = ( –1 ) ( 1 – x ) 3 2 3⁄2 d ------P (x) 3 2 dx 3 = –( 1 – x ) x = 0. In general. second. Legendre. y(:.3248 -5. m=0:2. y(:.-------------------= ( 1 – x ) ------2⎝ 2 ⎠ dx 2 2 3 x = – 0. fprintf('\n').2990 2.y') m 0 1 2 b.2)=legendre(3.7428 P2 ( x ) 3 x = 0.5). for the first.1)=m'.5 = ( –1 ) ( 1 – x ) 2 d -------P (x) 2 3 dx 2 x = – 0. Third Edition Copyright © Orchard Publications . and thus MATLAB will display a matrix whose rows correspond to the values of m = 0 . then P n ( x ) = 0 . fprintf('m\t Legendre \n'). m 2 3 15−20 Numerical Analysis Using MATLAB® and Excel®. Legendre -0.5).5 2 d ⎛ 15x – 3⎞ .4f \n'.4375 -0.Chapter 15 Bessel.2500 2 2⁄2 P3 ( x ) 2 x = – 0.1250 -1.0f\t %7. m = 2 . and Chebyshev Functions ans = -0. fprintf('%2. third and fourth rows respectively.25 2 3⁄2 d 2 ------( 3x – 1 ) 3 dx 3 x = 0.y') c.−0. y(:. fprintf('m\t Legendre \n'). m = 1 . if m > n . fprintf('%2.------------------= ( 1 – x ) ----dx ⎝ 2 ⎠ 2 = ( 1 – x ) ( 15x ) x = – 0.25 and since the third derivative of 3x – 1 is zero.5 = – 5.0. and m = 3 . it follows that P 2 ( x ) = 0 .

65) The orthogonality principle for these polynomials states that ∫0 Example 15. these are solutions of the differential equation x dy dx 2 2 + (1 – x) dy + ny = 0 dx (15.65). L 3 ( x ) Solution: Using Rodrigues’s formula of (15.Laguerre Polynomials 15. we obtain L0 ( x ) = e L1 ( x ) = e L2 ( x ) = e x d dx 0 0 (x e ) = e e 0 –x x –x = e = 1 0 x –x x –x –x d ( xe ) = e ( e – xe ) = 1 – x dx x d dx 2 2 3 ( x e ) = e e ( 2 – 4x + x ) = 2 – 4x + x 2 –x x –x 2 2 (15. Third Edition Copyright © Orchard Publications 15−21 .3 Laguerre Polynomials Another class of polynomials that satisfy the orthogonality principle. L 2 ( x ) d. L 0 ( x ) b.66) Compute the Laguerre polynomials a.67) x 3 –x 2 3 L 3 ( x ) = e d ( x e ) = 6 – 18x + 9x – x 3 dx The differentiation of the last two polynomials in (15.2).64) These polynomials are computed with the Rodrigues’ formula x n –x Ln ( x ) = e d ( x e ) n dx n (15.% Differentiate y twice with respect to x Numerical Analysis Using MATLAB® and Excel®. are the Laguerre polynomials L n ( x ) .67) was performed with MATLAB as follows: syms x y z y=x^2*exp(−x).7 ∞ e L m ( x ) L n ( x ) dx = 0 –x (15. z=diff(y. L 1 ( x ) c.

2).68) are referred to as Chebyshev polynomials of the first kind and are denoted as y = T n ( x ) .* The solutions of (15.Chapter 15 Bessel. another class of orthogonal functions known as Genenbauer or Ultraspherical functions use the notation Cn ( x ) and for this reason. y=x^3*exp(−x). Legendre. (a) 15−22 Numerical Analysis Using MATLAB® and Excel®. simplify(L3x) ans = 6-18*x+9*x^2-x^3 15. we will avoid notation C k ( x ) for the Chebyshev polynomials. * Some books use the notation Ck ( x ) for these polynomials. For x > 1 they increase or decrease more rapidly than any other polynomial of order n . Third Edition Copyright © Orchard Publications .4 Chebyshev Polynomials The Chebyshev polynomials are solutions of the differential equations dy 2 d y . This property is the basis for the Chebyshev approximation in the design of Chebyshev type electric filters.% Differentiate y twice % we cannot differentiate three times at once w=diff(z). these are denoted as y = U n ( x ) .+ n(n + 2 )y = 0 – 3x ----( 1 – x ) --------2 dx dx 2 (15. simplify(L2x) ans = 2-4*x+x^2 syms x y z w . Both kinds comprise a set of orthogonal functions. They exhibit equiripple amplitute characteristics over the range – 1 ≤ x ≤ 1 . We will plot some of these later in this section.69) are the Chebyshev polynomials of the second kind. and Chebyshev Functions L2x=exp(x)*z. 2. they oscillate with the same ripple. However. that is.68) and dy 2 d y .69) The solutions of (15. Two interesting properties of the T n ( x ) polynomials are: 1. within this range.% Differentiate one more time L3x=exp(x)*w. z=diff(y.+ n2 y = 0 – x ----( 1 – x ) --------2 dx dx 2 (15. We will restrict our discussion to the T n ( x ) polynomials.

70) (15.75) Now. we replace n with n + 1 and we obtain T n + 1 ( y ) = cos ( n + 1 ) y = cos ny cos y – sin ny sin y (15.2 Chebyshev polynomials expressed in powers of x n 0 1 2 3 4 5 6 Tn ( x ) 1 x 2x – 1 4x – 3x 8x – 8x + 1 16x – 20x + 5x 32x – 48x + 18x – 1 6 4 2 5 3 4 2 3 2 To show that the relation of (15. Third Edition Copyright © Orchard Publications 15−23 . we let x = cos y (15.76) Numerical Analysis Using MATLAB® and Excel®.73) and (15.74) Similarly.71).2. replacing n with n – 1 .Chebyshev Polynomials These polynomials are tabulated in reference books which contain mathematical functions. we can express T n ( x ) as polynomials in powers of x .71) Using (15.75). A good reference is the Handbook of Mathematical Functions. we obtain T n – 1 ( y ) = cos ( n – 1 ) y = cos ny cos y + sin ny sin y (15.70) can be expressed as a polynomial.72) (15. T n ( x ) = cos ( n cos x ) T n ( x ) = cos h ( n cos h x ) –1 –1 for x ≤ 1 for x > 1 (15. and making use of (15. They can also be derived from the following relations. Some are shown in Table 15. we add (15.72).70) or (15. we obtain T n + 1 ( y ) + T n – 1 ( y ) = 2 cos ny cos y = 2T n ( y ) x = 2xT n ( y ) (15. Dover Publications.73). in (15.73) and T n ( y ) = cos ny Next.74) with (15. TABLE 15.

T 3 ( x ) = 2xT 2 ( x ) – T 1 ( x ) = 4x – 2x – x = 4x – 3x 3 (15. from (15. T 2 ( x ) = 2xT 1 ( x ) – T 0 ( x ) = 2x – 1 2 (15. we can replace y with x to obtain T n + 1 ( x ) = 2xT n ( x ) – T n – 1 ( x ) Recurrence Relation (15.73) yields T0 ( y ) = T0 ( x ) = 1 (15. we use the recurrence formula of (15.2. we can prove the first 3 entries of Table 15.73) and (15.80) 3 and when n = 3 . Third Edition Copyright © Orchard Publications .2 with (15.70) by letting y = cos x . Thus. and Chebyshev Functions or T n + 1 ( y ) = 2xT n ( y ) – T n – 1 ( y ) Then.Chapter 15 Bessel. T 0 ( x ) = cos ( 0 ⋅ cos x ) = cos ( 0 ⋅ y ) = 1 –1 For n = 1 . For instance. (15. T 1 ( x ) = cos ( 1 ⋅ cos x ) = cos ( 1 ⋅ y ) = cos y = x –1 and for n = 2 .77).81) –1 Alternately. can now be verified by using a combination of the above relations. Legendre. Thus. 4 and so on.77) The polynomials in Table 15. 3. when n = 2 . we obtain T1 ( y ) = T1 ( x ) = x (15.78) For n = 1 . for n = 0 . T 2 ( x ) = cos ( 2 ⋅ cos x ) = cos ( 2 ⋅ y ) = cos 2y = 2 cos y – 1 ⎧ ⎪ ⎪ ⎨ ⎪ ⎪ ⎩ ⎧ ⎪ ⎪ ⎨ ⎪ ⎪ ⎩ = 2 cos ( cos x ) – 1 = 2 2 –1 –1 2 cos ( cos x ) x 2 –1 ⋅ cos ( cos x ) x –1 –1 or T 2 ( x ) = 2x – 1 Relation (15.71) can be derived from (15.72).79) To derive the algebraic expressions corresponding to n = 2.70) as follows: We recall that 15−24 Numerical Analysis Using MATLAB® and Excel®. for n = 0 .

x.Chebyshev Polynomials e +e cos α = ----------------------2 jα –j α (15.71). We can also use MATLAB to convert (15.01:1. Tnx2=cos(2*acos(x)).5]).70) and (15. xlabel('x').2 1. expand(cos(3*acos(x))). axis([−1. j α = cosh v . if n = 3 . Tnx1. we obtain T n ( x ) = cos [ n ( – j cosh x ) ] = cos ( – jn cosh x ) = cosh ( jnj cosh x ) = cosh [ j ( – jn cosh x ) ] = cosh ( n cosh x ) –1 –1 –1 –1 –1 and this is the same as (15.. Tnx4=cos(4*acos(x)). % Chebyshev polynomials % x=−1. title('Chebyshev Polynomials of the First Kind'). Tnx0. Tnx1=cos(1*acos(x)). grid. Tnx6=cos(6*acos(x)).71) to polynomials. Tnx6). Tnx3.. –1 –1 –1 –1 Numerical Analysis Using MATLAB® and Excel®.85) follows. syms x. x.2. making use of (15.2 −1.2:0. then α = cos v . x.85) Then.70).. Tnx4. j cos v = cosh v and (15.83) (15. plot(x. expand(cosh(3*acosh(x))) ans = 4*x^3-3*x ans = 4*x^3-3*x The MATLAB script below plots the T n ( x ) for n = 0 through n = 6 . For example.5 1. Tnx5.84) (15. cos x = – j cosh x * –1 –1 By substitution into (15. Tnx3=cos(3*acos(x)). Third Edition Copyright © Orchard Publications 15−25 .82) and e +e cos h α = ------------------2 α –α (15. and that cosh ( – t ) = cosh t . x.85). Tnx2. cos α = cos hj α and when x > 1 . x. x.. Tnx0=cos(0*acos(x)). ylabel('Tn(x)') % We could have used the gtext function to label the curves but it is easier with the Figure text % tool * Let cos α = cosh j α = v . Tnx5=cos(5*acos(x)).

5 T5 -1 T4 T6 T2 T3 -1.86) where ω is the operating radian frequency.6 shows the plot of the Chebyshev polynomials of the first kind T n ( x ) for n = 0 through n = 6 . if we want to design a second order Chebyshev low−pass filter. among other applications.8 -0. ω C is the cutoff frequency.8 1 Figure 15. Third Edition Copyright © Orchard Publications .5 -1 -0.6 -0.Chapter 15 Bessel. Chebyshev polynomials.4 -0. are used in the design of electric filters. we use the Chebyshev polynomial T 2 ( x ) = 2x – 1 2 and (15. Legendre.4 0.86) becomes α A ( ω ) = ----------------------------------------------------2 2 1 + ε [ 2 ( ω ⁄ ωC ) – 1 ] 2 (15. Plot of Chebyshev polynomials with MATLAB As mentioned earlier. Orchard Publications. The basic Chebyshev low−pass filter approximation is defined as α A ( ω ) = -----------------------------------------2 2 1 + ε T n ( ω ⁄ ωC ) 2 (15. ISBN 0−9744239−9−8. and Chebyshev Functions Figure 15. For example. 1.5 Chebyshev Polynomials of the First Kind 1 T0 T1 0.2 0.* The filters are described in terms of rational polynomials that approximate the behavior of ideal filters.2 0 x 0.5 Tn(x) 0 -0.6. 15−26 Numerical Analysis Using MATLAB® and Excel®.87) * For a thorough discussion on the design of analog and digital filters refer to Signals and systems with MATLAB Applications. and α and ε are other parameters that are used to specify the order and type of the electric filter.6 0.

+ ----------------------. Third Edition Copyright © Orchard Publications 15−27 .+ --------------------x -–… x .5 Summary • Differential equations with variable coefficients cannot be solved in terms of familiar functions as those which we encountered in ordinary differential equations with constant coefficients.– ----------------------. or even a complex number. • Bessel functions are solutions of the differential equation x2 dy dx 2 2 +x dy 2 2 + (x – n )y = 0 dx where n can be any number. • For n = 0. 1 and 2 the series reduce to x .Summary 15.– ----------------------.+ -----------------------–… 2 2 3 ⋅ 1! ⋅ 2! 2 5 ⋅ 2! ⋅ 3! 2 7 ⋅ 3! ⋅ 4! 2 9 ⋅ 4! ⋅ 5! x x x x x J 2 ( x ) = -------------.– --------------------x . • The series J –n ( x ) = k=0 ∑ ( –1 ) ∞ k x ⎞ – n + 2k 1 ⋅ ⎛ -⋅ ------------------------------------------⎝2⎠ k! ⋅ Γ ( – n + k + 1 ) is referred to as the Bessel function of negative order n . • The series Jn ( x ) = k=0 ∑ ∞ x ⎞ n + 2k 1 k ⋅ --------------------------------------( – 1 ) ⋅ ⎛ -⎝2⎠ k! ⋅ Γ ( n + k + 1 ) n≥0 where n is any positive real number or zero is referred to as Bessel function of order n . fractional.– ----------------------.– ----------------------. The general solution depends on the value of n .+ ----------------------. The usual procedure is to derive solutions in the form of infinite series.+ -------------------------–… 2 4 6 8 10 2 ⋅ 2! 2 ⋅ 1! ⋅ 3! 2 ⋅ 2! ⋅ 4! 2 ⋅ 3! ⋅ 5! 2 ⋅ 4! ⋅ 6! 2 4 6 8 10 3 5 7 9 2 4 6 8 • Two more useful relations are d J ( x ) = –J ( x ) 1 dx 0 Numerical Analysis Using MATLAB® and Excel®. positive or negative integer.+ --------------------J 0 ( x ) = 1 – --------------------2 2 4 2 6 2 8 2 2 ⋅ ( 1! ) 2 ⋅ ( 2! ) 2 ⋅ ( 3! ) 2 ⋅ ( 4! ) x x x x x J 1 ( x ) = -. and the most common are the Method of Frobenius and the Method of Picard.

They also can be found in math table books. certain differential equations resemble the Bessel equation.x) function or the Excel BESSELJ(x. the general solution of the Bessel equation is y = AJ n ( x ) + BJ – n ( x ) n ≠ 0. Third Edition Copyright © Orchard Publications . or Neumann functions exist. J 0 ( a ) = 0 and J 0 ( b ) = 0 . and can also be found with the MATLAB besselj(n. dy d y . … • If a and b are distinct roots of J 0 ( x ) = 0 . then. • Besides the above functions known as Bessel functions of the first kind. other Bessel functions. and Chebyshev Functions d { xJ 1 ( x ) } = xJ 0 ( x ) dx • Values of J n ( x ) can be calculated using the appropriate series given above. for this case. ∫0 xJ0 ( ax ) J0 ( bx ) dx • The differential equation 1 = 0 and thus we say that J 0 ( ax ) and J 0 ( bx ) are orthogonal in the interval 0 ≤ x ≤ 1 . or Hankel functions. 15−28 Numerical Analysis Using MATLAB® and Excel®. or Weber functions. 1. Also.n) function. J n ( x ) and J –n ( x ) are linearly independent. Legendre. denoted as Y n ( x ) and referred to as Bessel functions of the second kind. and thus their solutions are called Modified Bessel functions.cos x πx are known as half−order Bessel functions. is referred to as Legendre functions of the second kind. 3. • The infinite series solution of the Legendre functions. is known as Legendre’s equation.sin x and J ( – 1 ⁄ 2 ) ( x ) = πx 2 ----. • The Bessel functions J1 ⁄ 2 ( x ) = 2 ----. • If n is not an integer. denoted as Q n ( x ) .Chapter 15 Bessel. 2.+ n(n + 1 )y = 0 – 2x ----( 1 – x ) --------2 dx dx 2 2 where n is a constant.

is 2n + 1 B n = -------------2 ∫– 1 f ( x ) P n ( x ) d x ∑ Bn Pn ( x ) n=0 ∞ 1 and with this relation we can find a polynomial f ( x ) defined as f ( x ) = B0 P0 ( x ) + B1 P1 ( x ) + B2 P2 ( x ) + … + Bn Pn ( x ) = • The trigonometric form of the Legendre polynomials is 1 ⋅ 3 ⋅ … ⋅ ( 2n – 1 ) 1 1 ⋅ 3 ⋅ … ⋅ ( 2n – 3 ) .⋅ -------------------------------------------. Third Edition Copyright © Orchard Publications 15−29 .for n = odd where N = -2 2 and the first 6 Legendre polynomials are P0 ( x ) = 1 1 . and offers another method of expressing the Legendre polynomials.2 ⋅ (x – 1) P n ( x ) = -------------n n 2 ⋅ n! dx is known as Rodrigues’ formula.for n = even and N = ---------------. • The Legendre polynomial series of a function f ( x ) for x < 1 .2 cos ( n – 2 ) θ P n cos θ = -------------------------------------------2 ⋅ 4 ⋅ … ⋅ 2n 2 2 ⋅ 4 ⋅ … ⋅ ( 2n – 2 ) 1 ⋅ 3 1 ⋅ 3 ⋅ … ⋅ ( 2n – 5 ) . Numerical Analysis Using MATLAB® and Excel®.-------d .2 cos n θ + -.( 3x 2 – 1 ) P 2 ( x ) = -2 1 .Summary • The Legendre polynomials are defined as Pn ( x ) = k=0 ∑ N ( – 1 ) ⋅ ( 2n – 2k )! n – 2k ---------------------------------------------------x n 2 k! ( n – k )! ( n – 2k )! k (n – 1) n .2 cos ( n – 4 ) θ + … + --------2 ⋅ 4 2 ⋅ 4 ⋅ … ⋅ ( 2n – 4 ) and the first 6 Legendre polynomials in trigonometric form listed below.( 35x 4 – 30x 2 + 3 ) P 4 ( x ) = -8 P1 ( x ) = x 1 .⋅ -------------------------------------------.( 63x 5 – 70x 3 + 15x ) P 5 ( x ) = -8 • The relation n n 1 .( 5x 3 – 3x ) P 3 ( x ) = -2 1 .

Third Edition Copyright © Orchard Publications . Legendre.y = 0 ----. also known as spherical harmonics. satisfy the orthogonality principle when m ≠ n as indicated by the integral ∫ ⎧ 0 ⎪ P m ( x ) P n ( x ) dx = ⎨ 2 –1 ⎪ -------------⎩ 2n + 1 1 m≠n m = n • The Legendre polynomials in trigonometric form satisfy the orthogonality principle when m ≠ n as indicated by the integral ∫0 π ⎧ 0 ⎪ P m ( cos θ ) P n ( cos θ ) sin θ dθ = ⎨ 2 ⎪ -------------2n + 1 ⎩ m≠n m = n • The differential equation 2 dy 2 d y m . These functions. are evaluated from the relations Pn ( x ) = ( –1 ) ( 1 – x ) m m 2 m⁄2 m m d P (x) ⋅ --------m n dx m 15−30 Numerical Analysis Using MATLAB® and Excel®. The functions P n ( x ) and Q n ( x ) are referred to as associated Legendre functions of the first and second kind respectively.Chapter 15 Bessel.+ n ( n + 1 ) – ------------( 1 – x ) --------– 2x 2 2 dx dx 1–x 2 is referred to as the associated Legendre differential equation. and Chebyshev Functions P 0 cos θ = 1 P 1 cos θ = cos θ 3 cos 2 θ + 1 P 2 cos θ = --------------------------4 5 cos 3 θ + 3 cos θ P 3 cos θ = --------------------------------------8 35 cos 4 θ + 20 cos 2 θ + 9 P 4 cos θ = ---------------------------------------------------------64 cos 5 θ + 35 cos 3 θ + 30 cos θ ------------------------------------------------------------------------P 5 cos θ = 63 128 • The Legendre polynomials in algebraic form. The general solution of this equation is y = C1 Pn ( x ) + C2 Qn ( x ) m m where C 1 and C 2 are arbitrary constants.

6. • The T n ( x ) polynomials are derived from the relations T n ( x ) = cos ( n cos x ) T n ( x ) = cos h ( n cos h x ) –1 –1 for x ≤ 1 for x > 1 These polynomials exhibit equiripple amplitute characteristics over the range – 1 ≤ x ≤ 1 .x) function computes the associated Legendre functions of the first kind of degree n . Numerical Analysis Using MATLAB® and Excel®. These polynomials are satisfy the orthogonality principle. Both kinds comprise a set of orthogonal functions.+ n2 y = 0 – x ----( 1 – x ) --------2 dx dx 2 and dy 2 d y . • The solutions of the differential equation x dy dx 2 2 + (1 – x) dy + ny = 0 dx are known as Laguerre polynomials and are denoted as L n ( x ) . these are denoted as y = U n ( x ) . and order m = 0. Third Edition Copyright © Orchard Publications 15−31 . that is. They are computed with the Rodrigues’ formula x n –x Ln ( x ) = e d ( x e ) n dx n • The Chebyshev polynomials are solutions of the differential equations dy 2 d y . The solutions of the second are the Chebyshev polynomials of the second kind. …. they oscillate with the same ripple as shown in Figure 15. This property is the basis for the Chebyshev approximation in the design of Chebyshev type electric filters. within this range. n evaluated for each element of x .+ n(n + 2)y = 0 – 3x ----( 1 – x ) --------2 dx dx 2 The solutions of the first differential equation are referred to as Chebyshev polynomials of the first kind and are denoted as y = T n ( x ) . 1.Summary Qn ( x ) = ( –1 ) ( 1 – x ) m m 2 m⁄2 d Q (x) ⋅ --------m n dx m • The MATLAB legendre(n. 2.

Third Edition Copyright © Orchard Publications . P 0 ( x ) through P 4 ( x ) will be sufficient. i. f (x) 1 −1 1 x 15−32 Numerical Analysis Using MATLAB® and Excel®. The first 5 terms of P n ( x ) .5 m b. and Chebyshev Functions 15. P 3 ( x ) f. Compute the Legendre polynomial g ( x ) representing the waveform f ( x ) of the figure below.e. use MATLAB or Excel to plot g ( x ) and compare with f ( x ) .Chapter 15 Bessel. Use the appropriate Legendre polynomials P n ( x ) or P n ( x ) and Rodrigues’s formulas to compute the following. a. J 1 ⁄ 2 ( π ⁄ 6 ) d. Legendre. Use the appropriate series of the Bessel functions J n ( x ) to compute the following values using the first 4 terms of the series and check your answers with MATLAB or Excel. P 1 ( x ) d. P 2 ( x ) e. J 1 ( 2 ) c.. P 1 ( x ) 2 x = 0.75 c.6 Exercises 1. P 2 ( x ) 3 x = 0. J 0 ( 3 ) b. P 3 ( x ) 2 x = 0. J –1 ⁄ 2 ( π ⁄ 3 ) 2. and check your answers with MATLAB. a.25 3.5 x = – 0.5 x = 0. Then.25 x = 0.

5.2563 and this is much closer to the value obtained with MATLAB.+ ----. besselj(−0. x x x x J 1 ( x ) = -.= 0.– ----------. Numerical Analysis Using MATLAB® and Excel®.Solutions to End−of−Chapter Exercises 15.sin ( π ⁄ 6 ) = --------π 2 2π π π(π ⁄ 6) 6 1 6 2 cos ( π ⁄ 3 ) = -----. x x x J 0 ( x ) = 1 – --------------------.– -----------.7 Solutions to End−of−Chapter Exercises 1.⋅ -.= – 0.5513 -----------------. This is because our computation was based on the first 4 terms of the series.+ --------------------.2).3).– ------------------------------2 22 ⋅ 4 22 ⋅ 42 ⋅ 6 22 ⋅ 42 ⋅ 62 ⋅ 8 3 5 7 x=2 8 32 128 = 1 – ----.⋅ -.2601 ans = 0.pi/6).cos x ----πx = x = π⁄3 Check with MATLAB: besselj(0.= 2 --------.– --------------------2 2 4 2 6 2 2 ⋅ ( 1! ) 2 ⋅ ( 2! ) 2 ⋅ ( 3! ) 2 4 6 x=3 9 81 729 = 1 – -.= 0. Third Edition Copyright © Orchard Publications 15−33 . besselj(1.+ -------.5.3898 -----------------π 2 2π π(π ⁄ 3) d.pi/3) ans = -0.3008 4 64 2304 b. besselj(0.= 0.= -----.– -------------.5764 16 384 18432 c.5513 ans = 0. J–1 ⁄ 2 ( x ) = 2.= -----.+ ---------------------. J1 ⁄ 2 ( x ) = 2 ----. a. Had we taken also the fifth term our answer would have been – 0.sin x πx = x = π⁄6 12 1 3 3 2 .3898 We observe that the first value returned by MATLAB above is significantly different from that we obtained from the series.5767 ans = 0.

75 c.( 3x 2 – 1 ) ⎬ = 0 3 3⎨2 dx dx ⎩ ⎭ f. e.( 35x 4 – 30x 2 + 3 ) P 4 ( x ) = -8 m Pn ( x ) m P1 ( x ) = x 1 .5 ( 5 × ( 0.25 = 3.5 d -(x) = ( 1 – x ) ⋅ ------2 dx 2 = 0 x = 0.5 m = 0 This is because m > n .5 ( 3 × ( 0.⋅ ----.5 2 b. P3 ( x ) 2 x = 0.5 = x = 0.-.75 ) – 1 ) = 0.25 1 d 2 .25 = ( –1 ) ( 1 – x ) 2 2 2⁄2 d -------P (x) 2 3 dx 2 x = 0.3438 x = 0. We will use the relations of (15.25 2 ⎫ 2 d ⎧1 .25 1 .5 = ( –1 ) ( 1 – x ) 2 2 2⁄2 d -------P (x) 2 1 dx x = 0. P2 ( x ) 3 x = – 0.( 5x 3 – 3x ) = -2 = 0.( 63x 5 – 70x 3 + 15x ) P 5 ( x ) = -8 2 m⁄2 = ( –1 ) ( 1 – x ) d ⋅ --------P (x) m n dx m a.3359 x = 0.34) and (15.5160 15−34 Numerical Analysis Using MATLAB® and Excel®. P2 ( x ) x = 0.( 5x 3 – 3x ) ⎬ = ( 1 – x ) ------2⎨2 dx ⎩ ⎭ x = 0. P3 ( x ) x = 0. Legendre. In other words.( 3x 2 – 1 ) = -2 = 0.25 ) – 3 × 0. then P n ( x ) = 0 . Third Edition Copyright © Orchard Publications .4688 ( 30x ) x = 0.25 ) = – 0. P1 ( x ) x = 0. 2 P 1 ( x ) x = 0.25 2 2 3 d. and Chebyshev Functions 2.25 x = 0.Chapter 15 Bessel.62).( 15x 2 – 3 ) = ( 1 – 0.75 1 . They are repeated below for convenience.( 3x 2 – 1 ) P 2 ( x ) = -2 1 . 3 3 ⎫ d ⎧1 d ------.P 2 ( x ) = ------.( 5x 3 – 3x ) P 3 ( x ) = -2 1 .-.25 ) ⋅ -2 dx = 0. P0 ( x ) = 1 1 .5 We recall that if m > n .

9985 2 3. that is.x) function computes the associated Legendre functions of degree n and order m = 0. y(:.y') As in (d) above m > n and MATLAB returns Numerical Analysis Using MATLAB® and Excel®. fprintf('m\t Legendre \n'). m > n .. fprintf('\n'). fprintf('m\t Legendre \n').2). m=0:2. y=zeros(2. evaluated for each element of x.4f \n'.3359 1 0.2).1)=m'.0.0. y=zeros(2.5). fprintf('%2. m=0:3. fprintf('m\t Legendre \n').3125 c. 1. fprintf('m\t Legendre \n'). . y(:.2).0.1)=m'.2)=legendre(1.6160 d.Solutions to End−of−Chapter Exercises Check with MATLAB: a.5).4f \n'.y') m Legendre 0 0. fprintf('%2. e. y(:. y=zeros(3. y(:.−0.4f \n'.1)=m'.0. fprintf('\n'). m=0:1. y(:. fprintf('\n').y') m Legendre 0 0.75).8660 Here.0f\t %7. fprintf('%2. fprintf('\n'). the legendre(n.0f\t %7.5). y(:. m=0:2.5000 1 -0.0f\t %7.2).5156 3 -13... y(:.y') m Legendre 0 -0. m=0:1. Third Edition Copyright © Orchard Publications 15−35 . For this reason we’ve used m=0:1.8660 b. y(:.4f \n'.2)=legendre(1. n. For this example. fprintf('%2.5000 1 -0.2). fprintf('\n').3438 1 -1.25). y=zeros(3. y=zeros(4. m = 2 and n = 1 and the statement m=0:2 is not accepted.2)=legendre(2.0f\t %7.0f\t %7.2)=legendre(3. fprintf('m\t Legendre \n').1)=m'.4f \n'.2)=legendre(2.1)=m'. fprintf('%2.y') m Legendre 0 0.4882 2 1. y(:. y(:.

4f \n'.0f\t %7.⎞ ( 5x – 3x ) dx = -4⎝ 5 3 ⎠ 1 = 0 0 9 B 4 = -2 ∫0 1 1 9 .25).⋅ ---x ⋅ P 0 ( x ) ⋅ dx = -2 2 3 -⋅x ---x ⋅ P 1 ( x ) ⋅ dx = -2 3 1 = -4 = 1 -2 1 0 1 3 1 0 ∫0 1 1 5 . 1 B 0 = -2 3 B 1 = -2 5 B 2 = -2 7 B 3 = -2 ∫0 ∫0 1 1 x .– ---( 3x – x ) dx = -4⎝ 4 2 ⎠ 3 4 2 5 = ----16 1 ∫0 1 1 . y=zeros(4.– 3x ------.2990 2 2. y(:.3359 1 0.Chapter 15 Bessel.⎛ 5x ------.+ ------16 ⎝ 6 4 2 ⎠ 3 = – ----32 15−36 Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications .---------= ----.( 35x 4 – 30x 2 + 3 ) dx = ----x ⋅ -8 16 1 0 ∫0 ( 35x 5 – 30x + 3x ) dx 3 9 ⎛ 35x 6 30x 4 3x 2 ⎞ . fprintf('\n').1)=m'. fprintf('%2. and Chebyshev Functions m Legendre 0 -0.2)=legendre(3. m=0:3.9985 2 3.6160 2n + 1 B n = -------------2 ∫–1 f ( x ) Pn ( x ) dx ∫0 f ( x ) Pn ( x ) dx 2 1 0 1 1 For this exercise f ( x ) = 0 for – 1 ≤ x ≤ 0 and thus 2n + 1 B n = -------------2 Then.( 3x 2 – 1 ) dx = -x ⋅ -2 4 ∫0 1 1 5 ⎛ 3x 4 x 2 ⎞ .2500 f.2).------. fprintf('m\t Legendre \n').1250 1 1. Legendre.y') 3.( 5x 3 – 3x ) dx = 7 -x ⋅ -2 4 ∫0 5 3 7 .5156 3 -13.– ---------. m Legendre 0 -0. y(:.0.

– ------------.⋅ 1 + -= -16 2 2 32 8 4 1 x 15x 5 105x 90x 9 .2 0.3 0. xlabel(‘x’). We plot f ( x ) with the MATLAB script below.4 0. Also.2 0.9 1 Numerical Analysis Using MATLAB® and Excel®.( 35x 4 – 30x 2 + 3 ) . ylabel(‘f(x)’).7 0.( 3x 2 – 1 ) – ----. x=0:0.P ( x ) + 0 ⋅ P 3 ( x ) – -----P (x) . plot(x.Solutions to End−of−Chapter Exercises and by substitution into f ( x ) = B0 P0 ( x ) + B1 P1 ( x ) + B2 P2 ( x ) + … + Bn Pn ( x ) = we obtain n=0 ∑ Bn Pn ( x ) ∞ 1 5 3 1 .4 0.*x. f ( 1 ) = 248 ⁄ 256 and this value is close to unity.P ( x ) + ----. Third Edition Copyright © Orchard Publications 15−37 .^4).fx)./256.6 f(x) 0.*x+210. These values are close to zero.⋅ -.01:1.3 0. grid 1 0.5 x 0.8 0.⋅ -.1 0 0 0.P ( x ) + -f ( x ) = -2 1 16 2 32 4 4 0 5 1 1 3 1 1 .⋅ x + ----.+ -.1 0.+ ---------= -.– -------4 2 32 32 256 256 256 2 4 2 or 12 4 f ( x ) = -------( 15 + 128x + 210x – 105x ) 256 we note that f ( – 1 ) = – 8 ⁄ 256 and f ( 0 ) = 15 ⁄ 256 .– ----.6 0. fx=(15+128.*x.8 0.+ ---------.7 0.^2−105.5 0.9 0.

is subject to limited supplies.00 per 1000 Due to limited supplies of silicon.1. and in most cases are subject to certain conditions or limitations referred to as constraints. * A linear program is one in which the variables form a linear combination. These variables may or may not be independent of each another. and we will illustrate these with some simple but practical examples. Numerical Analysis Using MATLAB® and Excel®. dynamic programming.. its product mix at times of high consumer demand. Third Edition Copyright © Orchard Publications 16−1 . are linearly related. required to manufacture the μ Ps and RAMs and the profits for each are shown in Table 16. ABC Semiconductor can only buy 450 parts of Material A .00 per unit RAMs (1000s) 2 10 $20.1 Linear Programming In linear* programming we seek to maximize or minimize a particular quantity. phosphorus and boron. 16. These methods are topics discussed in detail in a branch of mathematics called operations research and it is concerned with financial and engineering economic problems. All other programs are considered non−linear. Our intent here is to introduce these methods with the basic ideas. referred to as the objective. A and B . Thus. which is dependent on a finite number of variables.e. Example 16. The material types.1 The ABC Semiconductor Corporation produces microprocessors ( μ Ps ) and memory ( RAM ) chips. and network analysis. and 1000 parts of Material B . This corporation needs to know what combination of μ Ps and RAMs will maximize the overall profit.Chapter 16 Optimization Methods T his chapter introduces three methods for maximizing or minimizing some function in order to achieve the optimum solution. TABLE 16.1 Parts of Material Types μ Ps Semiconductor Material A Semiconductor Material B Profit 3 5 $25. We will discuss linear programming.1 Data for Example 16.i.

This line will pass through one of the three corners denoted as a .Chapter 16 Optimization Methods Solution: Since with Material A we can produce 3 μ Ps and 2 RAMs . the corporation is confronted with the following constraints: 3x + 2y ≤ 450 5x + 10y ≤ 1000 We now can state the problem as subject to the constraints Maximize z = 25 × μ P + 20 × RAMs 3x + 2y ≤ 450 5x + 10y ≤ 1000 (16.1) (16. and x and y must be integers. a graphical solution is possible. The x and y intercept corresponding to the above equations is shown in the plot of Figure 16.1 where the cross−hatched area indicates the feasible region. Plot of constraint lines for Example 16. therefore.1. * The feasible region is the area which includes all points ( x. y ≥ 0 . 16−2 Numerical Analysis Using MATLAB® and Excel®. We will solve this example graphically. y ) satisfying all constrains. For this example. and c . and with Material B 5 μ Ps and 10 RAMs .1 The equation of the straight line of the maximum profit is referred to as isoprofit line. Third Edition Copyright © Orchard Publications . b . x and y .* y 250 200 150 100 50 a 5x + 10y = 1000 c 50 100 150 200 250 x 3x + 2y = 450 Isoprofit line b Figure 16.2) Two additional constraints are x ≥ 0 . there are only two variables.

these values must be integers. so we accept the values x = 125 . Then. and the highest isoprofit line passes through point b . Numerical Analysis Using MATLAB® and Excel®.1). The coordinates of point b in Figure 16. z max = 25 × 125 + 20 × 37 = $3865 (16. all possible isoprofit lines have the same slope. are parallel to each another. 25 C .x + ----.= 154. then. we obtain the points 3865 x = ----------.Linear Programming The isoprofit line that we are interested is described by the equation z = 25 × μ P + 20 × RAM = cons tan t = 25x + 20y = C (16. and y = 37 . Then.3) We can express this equation in y = mx + b form.5) Using MATLAB for the solution of (16. Third Edition Copyright © Orchard Publications 16−3 . 5*x+10*y−1000) x = 125 y = 75/2 Of course. y = 0 .25x + k y = – ----20 20 (16.25 y = 3865 20 This is shown as a dotted line on the plot of Figure 16. Therefore.6) and the isoprofit line can be drawn from the equation 25x + 20y = 3865 (16.4) where k is the y −intercept.7) by first letting x = 0 . by substitution into (16.1.= 193.6 25 and ----------.= – 1.1 are found by simultaneous solution of 3x + 2y = 450 5x + 10y = 1000 (16.5) we obtain syms x y [x y]=solve(3*x+2*y−450. that is. that is.

Add B18<=1000. ac . Third Edition Copyright © Orchard Publications . bd . Add B17<=450.Chapter 16 Optimization Methods It was possible to solve this problem graphically because it is relatively simple. where the nodes a through h represent the states. Enter zeros in B12 and B13 2. Use Add-Ins if necessary to add it. Add B13=Integer.2 A 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 B C D E F Optimization .2. we must resort to algebraic methods such as the simplex method. This and other methods are described in operations research textbooks.3 represents a line graph. Figure 16. the remaining decisions must constitute an optimum policy with regard to the state resulting from the first decision. Spreadsheet for solution of Example 16. however. 16−4 Numerical Analysis Using MATLAB® and Excel®. and the choice of alternative paths when leaving a given state. and so on. Add B13>=0. is called a decision. We can find the optimum solution to this type of problems with Excel’s Solver feature. On the Solver Parameters screen enter the following: Set Target Cell: B15 Equal to: Max By Changing Cells: B12:B13 Click on Add and enter Constraints: B12=Integer.860 448 1000 Note: Contents of A12:A18 are typed-in for information only Figure 16. OK . The procedure is included in the spreadsheet of Figure 16.1 1. From the Tools drop menu select Solver .1 with Excel’s solver 16.Maximum Profit for Example 16. Add B12>=0.2 Dynamic Programming Dynamic Programming is based on R. 5. In B17 enter =3*B12+2*B13 and in B18 =5*B12+10*B13 4. Solve x(μPs)= 124 y(RAMs)= 38 Maximum Profit= Semiconductor Material A= Semiconductor Material B= $3. Bellman’s Principle of Optimality which states that: An optimum policy has the property that whatever the initial state and the initial decisions are. we cannot obtain the solution by graphical methods and therefore. In most cases. In B15 enter =25*B12+20*B13 3. The alternative paths are represented by the line segments ab .

( J ac + J ce + J eh ). ( J bg + J gh ) } (16. and h . Then. and the initial decision has been made to go to state b . For example. f . if the initial decision is to go from state a to c .8) The costs for the other possible paths are defined similarly. Line graph for a typical dynamic programming example We assume that all segments are directed from left to right. Third Edition Copyright © Orchard Publications 16−5 . there is a cost associated with each segment.9) can now be expressed in terms of (16.3. we say that the optimum path policy for this line graph is J min = min { ( J ac + J cf + J fh ). Let the minimum cost from state b to h be denoted as g b . in this case.12) This relation indicates that to obtain the minimum cost we must minimize: 1. and each has a value assigned to it which we will refer to as the cost. Accordingly.Dynamic Programming c e d a b h f g Figure 16. and it is usually denoted with the letter J .10) Likewise. the cost is J ah = J ac + J cf + J fh (16. the objective is to go from state a to state h with minimum cost.10) and (16. ( J ab + J bd + J dh ). Then. costs J ab and J ac . g c = min { ( J cf + J fh ). let us suppose that the initial state is a .11) as g a = J min = min { ( J ab + g b ). ( J ce + J eh ) } (16. ( J ac + g c ) } (16. if the entire path from a to h is to be optimum (minimum in this case).3. Thus.11) The optimum path policy of (16. the path from b to h must be selected optimally.9) Now. Numerical Analysis Using MATLAB® and Excel®. ( J ab + J bg + J gh ) } (16. The part which is related to the present decision. for the path a . c . For the line graph of Figure 16. the path from state c to h must be optimum. g b = min { ( J bd + J dh ). that is.

The line segments are directed from left to right and the costs are indicated beside each line segment. ( 5 + g k ) } = min { ( 3 + 5 ). ( 6 + g e ) } = min { ( 9 + 6 ).16) 16−6 Numerical Analysis Using MATLAB® and Excel®.14) (16.4. ( 5 + 4 ) } = 8 (e → h → m) (16.Chapter 16 Optimization Methods 2. ( 3 + 12 ) } = 15 (a → c) (16. and f have no alternative paths since the lines are directed from left to right. Therefore. we make the first decision at state e .13) Next. ( 6 + 8 ) } = 14 (b → e) g c = min { ( 4 + g e ). Then. ( 6 + 8 ) } = 12 (c → e) (16.15) The final decision is at state a and thus g a = min { ( 5 + g b ). The part which represents the minimum value of all future costs starting with the state which results from the first decision. g b = min { ( 9 + g d ). Third Edition Copyright © Orchard Publications . Line graph for Example 16.2 8 6 e 5 k 4 3 6 h 5 m Solution: We observe that at states h . d .2 Find the minimum cost route from state a to state m for the line graph of Figure 16. ( 3 + g c ) } = min { ( 5 + 14 ). we make decisions at states b and c . ( 6 + g f ) } = min { ( 4 + 8 ).4. d 9 b 5 a 3 c 4 6 f Figure 16. g e = min { ( 3 + g h ). Example 16. k .

Line graph showing the minimum cost for Example 16. J . or L ) to minimize his total travel time. A salesman must leave New York City and be in one of the airports H . All flights originate at A and fly eastward.5.5 d 9 b 5 a 3 c 4 6 f Figure 16.3 L 3 5 4 3 4 K 6 4 J 2 H 3 Numerical Analysis Using MATLAB® and Excel®.6.Dynamic Programming Therefore. Third Edition Copyright © Orchard Publications 16−7 . also in hours. the minimum cost is 15 and it is achieved through path a → c → e → h → m . K . and in how many hours after departure from A will he reach his destination? D 4 B 8 A 6 3 C 5 7 4 2 2 E 4 3 F 7 Figure 16. node A represents an airport in New York City and nodes B through L several airports throughout Europe and Asia. and the numbers beside the line segments indicated the flight times. Which airport should he choose ( H . or L at the shortest possible time. J .6. Line graph for Example 16.2 8 6 e 5 k 4 3 6 h 5 m Example 16. K . The numbers in squares show the hours he must travel by an automobile to reach his destination. as shown in Figure 16. The encircled numbers represent waiting times in hours at each airport.3 On the line graph of Figure 16.

19) (16. g K = 4 . the minimum cost (minimum time from departure to arrival at destination) is 23 hours and it is achieved through path A → B → E → J .7. g D = 2 + min { ( 5 + g H ). and F . ( 5 + g F ) } = 3 + min { ( 7 + 9 ). E . Then. g J = 2 . ( 5 + 13 ) } = 3 + 16 = 19 C→E (16.22) The final decision is made at A . ( 7 + g L ) } = 3 + min { ( 6 + 4 ). as shown in Figure 16.23) Therefore.18) (16. ( 7 + 3 ) } = 3 + 10 = 13 F → K or F → L (16. Third Edition Copyright © Orchard Publications . ( 4 + 2 ) } = 2 + 6 = 8 D→J g E = 4 + min { ( 3 + g J ). ( 6 + g C ) } = min { ( 8 + 15 ). ( 2 + g E ) } = 4 + min { ( 4 + 8 ). Line graph showing the minimum cost for Example 16. 5 4 3 E 4 3 F 7 L 3 4 K 6 4 J 2 H 3 D 4 B 8 A 6 3 C 5 7 4 2 2 Figure 16.7. and g L = 3 (16.Chapter 16 Optimization Methods Solution: The hours that the salesman must travel by automobile to reach his destination are g H = 3 . ( 4 + gK ) } = 4 + min { ( 3 + 2 ). where we find g A = min { ( 8 + g B ).20) The next decisions are made at B and C where we find that g B = 4 + min { ( 4 + g D ).21) (16. ( 4 + g J ) } = 2 + min { ( 5 + 3 ).3 16−8 Numerical Analysis Using MATLAB® and Excel®. ( 4 + 4 ) } = 4 + 5 = 9 E→J g F = 3 + min { ( 6 + g K ).17) The first decisions are made at D . ( 6 + 19 ) } = 23 A→B (16. ( 2 + 9 ) } = 4 + 11 = 15 B→E g C = 3 + min { ( 7 + g E ).

B and C.000.000 $6. Investments in each product are assumed to be multiples of $1.000 $8.000.000.000 Return on Investment How should the money be allocated so that company will realize the maximum profit? Solution: This problem can also be solved with linear programming methods but we will use the so called tabular form of solution.000.Dynamic Programming Example 16.000. we express the profits in millions.000 $4. when x units of dollars are invested in it.000 $2.000 $7.000.000.000 $2.000. For simplicity.000 $1.000 $5.000.000.4 A start−up.000.25) subject to the constraint Numerical Analysis Using MATLAB® and Excel®.000.000 $6.000 and the company may allocate all the money to just one product or split it between these three products. C (16.000 to invest in three different products A. has $4.000. subject to the constraint that the amount invested does not exceed four million dollars.000 $5. TABLE 16. Let pi ( x ) i = A. and we enter these in Table 16. The expected profits are shown in Table 16.000 $7.3 Modified Table 16.2. we want to maximize z = p A ( x ) + p B ( x ) + p C ( x ) (16.000 $1. high−technology company.000 $4.000. TABLE 16.2 x p(x) pA ( x ) pB ( x ) pC ( x ) 0 0 0 0 1 2 1 1 2 5 3 4 3 6 6 5 4 7 7 8 Our objective is to maximize the total profit z that represents the sum of the profits from each product. B. Third Edition Copyright © Orchard Publications 16−9 .3.24) denote the profits in millions from product i .000 $3.000.000.000 $3.2 Amounts invested and return on investment for each product Investments Amount Invested 0 Product A Product B Product C 0 0 0 $1.000.000. In other words is.

4. we have three or less.30) with decision dC ( 3 ) = 5 If we have only two units left.Chapter 16 Optimization Methods xA + xB + xC ≤ 4 (16. 4. 1. The possibilities that we allocate 0 or 1 or 2 or 3 or 4 units (millions) to Product C.. p C ( 1 ). p C ( 3 ). we let d j ( u ) be the decision that is being made to achieve the optimum value from v j ( u ) .31) (16. and u = 4 .28) with decision dC ( 4 ) = 8 that is. We let v j ( u ) denote the value of the optimum profit that can be achieved. 1. the optimum value v C ( 3 ) is found from v C ( 3 ) = max { p C ( 0 ). We start by allocating units (millions) to Product C (Stage C). p C ( 2 ) } = max { 0. In this case. If we invest the remaining three units in Product C. p C ( 2 ). and C for Product C. B and C respectively. the maximum appears in the fourth position since the left most is the zero position. do not have 4 units to invest in Product C.3. by a previous decision.26) where x A . i. Third Edition Copyright © Orchard Publications . p C ( 2 ). we must consider all possibilities. 8 } = 8 (16.29) (16. p C ( 3 ) } = max { 0.e. 5 } = 5 (16. x B . and x C . 1.. from Table 16.32) with decision dC ( 2 ) = 4 With only one unit left to invest. Also. 4 } = 4 (16. where the subscript j indicates the number or stage assigned to the product. p C ( 1 ). p C ( 1 ).27) (16. j = C . 5. At Stage C.33) 16−10 Numerical Analysis Using MATLAB® and Excel®. are the amounts to be invested in products A. and the corresponding returns are. and we invest them in Product C. v C ( 4 ) = max { p C ( 0 ). The next possibility is that one unit was invested in either Product A or Product B. i. The computations are done in three stages. p C ( 4 ) } = max { 0. we obtain the maximum from v C ( 2 ) = max { p C ( 0 ). 4 millions assumed to be allocated to Product C. p C ( 1 ) } = max { 0. 1 } = 1 (16. but since we do not know what units were allocated to the previous products A and B. one per product. we have v C ( 1 ) = max { p C ( 0 ). B for Product B. A for Product A.e. and u represents the number of money units.

7 + 0 } = 8 (16.Dynamic Programming with decision dC ( 1 ) = 1 Finally. 2. Third Edition Copyright © Orchard Publications 16−11 .35) (16. 1 + 5. 3. and so on.39) with decision dB ( 4 ) = 0 since the maximum value is the zero position term.36) with decision dC ( 0 ) = 0 With these values. Inserting the appropriate values. p B ( 2 ) +v C ( 4 – 2 ). it is possible that all four units were invested in Product C.37) This expression says that if zero units were invested in Product B.4. or if one unit was invested in Product B. we construct Table 16.4 Optimum profit and decisions made for Stage C u 0 Stage C Stage B Stage A vC ( u ) dC ( u ) vB ( u ) dB ( u ) vA ( u ) dA ( u ) 0 0 … … … … 1 1 1 … … … … 2 4 2 … … … … 3 5 3 … … … … 4 8 4 … … … … Next. with no units left to invest in Product C.34) (16. 6 + 1. we obtain v B ( 4 ) = max { 0 + 8. Numerical Analysis Using MATLAB® and Excel®. p B ( 4 ) + v C ( 4 – 4 ) } (16. p B ( 1 ) + v C ( 4 – 1 ).38) (16. v C ( 0 ) = max { p C ( 0 ) } = max { 0 } = 0 (16. again we must consider all possibilities. it is possible that 3 units were invested in Product C. we have v B ( 4 ) = max { p B ( 0 ) + v C ( 4 – 0 ). we consider Stage B. and since we do not know what units were allocated to Product A (Stage A). 1 and 0 . p B ( 3 ) + v C ( 4 – 3 ). TABLE 16. With j = B and u = 4. 3 + 4.

43) (16. p B ( 3 ) + v C ( 3 – 3 ) } v B ( 3 ) = max { 0 + 5.52) or v B ( 0 ) = max { 0 + 0 } = 0 with decision dB ( 0 ) = 0 Next.41) (16.Chapter 16 Optimization Methods Using a similar reasoning. we have v B ( 3 ) = max { p B ( 0 ) + v C ( 3 – 0 ).46) (16. 3 + 1. 1 + 0 } = 1 with decision dB ( 1 ) = 0 if we consider the zero position term. Also. These are shown in Table 16. 1 + 4. p B ( 1 ) + v C ( 1 – 1 ) } or v B ( 1 ) = max { 0 + 1. 1 + 1. p B ( 1 ) + v C ( 2 – 1 ).47) (16.42) (16. v B ( 2 ) = max { p B ( 0 ) + v C ( 2 – 0 ). or dB ( 1 ) = 1 if we consider the first position term. Third Edition Copyright © Orchard Publications .44) (16.48) (16. v B ( 0 ) = max { p B ( 0 ) + v C ( 0 – 0 ) } (16. p B ( 2 ) +v C ( 2 – 2 ) } or v B ( 2 ) = max { 0 + 4.50) (16. we update the previous table to include the Stage B values.5.51) (16. p B ( 2 ) +v C ( 3 – 2 ).40) or with decision (16. 16−12 Numerical Analysis Using MATLAB® and Excel®. p B ( 1 ) + v C ( 3 – 1 ). 6 + 0 } = 6 (16. 3 + 0 } = 4 with decision dB ( 2 ) = 0 v B ( 1 ) = max { p B ( 0 ) + v C ( 1 – 0 ).49) dB ( 3 ) = 3 Also.45) (16.

9 units. * Since this is the first stage. v A ( 4 ) considers all possibilities. p A ( 4 ) + v B ( 4 – 4 ) } (16.5 Updated table to include Stage B values u 0 Stage C Stage B Stage A vC ( u ) dC ( u ) vB ( u ) dB ( u ) vA ( u ) dA ( u ) * 1 1 1 1 1 … … 2 4 2 4 0 … … 3 5 3 6 3 … … 4 8 4 8 0 … … 0 0 0 0 … … Finally. and the given table shows that 2 units ($2. that is.6 indicates that the maximum profit is realized with v A ( 4 ) = 9 .000.000. and this is always the case since in deriving v A ( 4 ) and d A ( 4 ) . 6 + 1.000) invested in this product will return $5. The only entries are in the last column.Dynamic Programming TABLE 16.6.000. all 4 units can be allocated to the Product A or some of these can be allocated to Products B and C. p A ( 1 ) + v B ( 4 – 1 ). 2 + 6. p A ( 2 ) +v B ( 4 – 2 ). 5 + 4. and thus the maximum profit is $9. this tells us that we should allocate 2 units to Product A .55) with decision dA ( 4 ) = 2 We complete the table by entering the values of Stage A in the last two rows as shown in Table 16. To determine the investment allocations to achieve this profit. Third Edition Copyright © Orchard Publications 16−13 .53) or v A ( 4 ) = max { 0 + 8.000. we start with d A ( 4 ) = 2 . 7 + 0 } = 9 (16.000.54) (16. Therefore. Numerical Analysis Using MATLAB® and Excel®. all possibilities have been considered. Table 16. with j = A and u = 4 v A ( 4 ) = max { p A ( 0 ) + v B ( 4 – 0 ). p A ( 3 ) + v B ( 4 – 3 ).

Thus. two units to Product C to earn $4. BE .3 Network Analysis A network. is a set of points referred to as nodes and a set of lines referred to as branches. C . as defined here. A typical network 16−14 Numerical Analysis Using MATLAB® and Excel®. Figure 16.000. Third Edition Copyright © Orchard Publications . This tells us that we should not invest any units in Product B if only two units are left.6. To find out where we should invest these units. and from Table 16.6. D and E .Chapter 16 Optimization Methods TABLE 16.000 2. we should allocate: 1. we have d B ( 4 – 2 ) = d B ( 2 ) .8 is a network with 5 nodes A . BD . This indicates that we should invest the remaining two units to Product C where we can obtain a return of $4. In summary.000.000.000.000. B E A D C Figure 16. and by reference to the Table 16. AC .000. two units to Product A to earn $5. AD . and 6 branches AB .6 Updated table to include values for all stages u 0 Stage C Stage B Stage A vC ( u ) dC ( u ) vB ( u ) dB ( u ) vA ( u ) dA ( u ) 0 0 0 0 … … 1 1 1 1 1 … … 2 4 2 4 0 … … 3 5 3 6 3 … … 4 8 4 8 0 9 2 We now have two units left to invest in Products B and C. we consider the decision at Stage B. we see that d 2 ( 2 ) = 0 . d C ( 2 ) = 2 . Since two out of the four units have already been invested. The decision at Stage C yields d C ( 4 – 0 – 2 ) = d C ( 2 ) . B . and CD . zero units to Product B to earn $0 3.8.000 16. to obtain the maximum profit of $9.

they are considered to be two−way. B E A C D Figure 16. are directed but the others are not.11 is a tree network. the network of Figure 16.9 is also connected.8. if they have a direction assigned to them. the branches indicate possible paths. Third Edition Copyright © Orchard Publications 16−15 . the branches BD and BE in Figure 16. However.11. or two−way.10. A network is said to be connected. B E A D Figure 16.10 is not connected since the branch CD is removed. A network which is not connected A tree is a connected network which has n branches and n + 1 nodes. Example 16. one−way. Thus. we seek to find a tree which contains all nodes.8 is connected. B E A D C Figure 16. the network of Figure 16.Network Analysis Branches can be either directed (or oriented).5 Figure 16. the network shown in Figure 16. and the sum of the costs (shortest total distance) is a minimum. In such problems.12 represents a network for a project that requires telephone cable be installed to link 7 towns. For example. The towns are the nodes. A network which is connected The network of Figure 16. A tree network Network analysis is a method that is used to solve minimum span problems. that is.9. if there is a path (branch) connecting each pair of nodes. and the numbers beside the Numerical Analysis Using MATLAB® and Excel®. Thus. If no direction is assigned.

show the distance (not to scale) between towns in kilometers. By inspection.5 with first connection 16−16 Numerical Analysis Using MATLAB® and Excel®.14. We accept this branch as the first branch of the minimum span tree and we draw a solid line from Node A to Node B as shown in Figure 16. AD . B 3 A 4 C 5 4 10 5 D 5 F 8 6 7 2 E 5 G Figure 16. and AC .14. B 3 A 4 C 5 4 10 5 D 5 F 8 6 7 2 E 5 G Figure 16.13. AB . i. we redraw the given network with dotted lines as shown in Figure 16. AC = 4 } = AB = 3 (16. Network of Example 16. that is. and we arbitrarily choose A as the starting node. Find the minimal spanning tree. or from the expression min { AB = 3.13.5 with no connections We observe that there are 3 branches associated with node A .Chapter 16 Optimization Methods branches. AD = 5.e.12. Network for Example 16. B 3 A 4 C 5 4 10 5 8 E 6 7 5 F 2 5 G D Figure 16.56) we find that branch AB is the shortest. Network of Example 16.. the least amount of telephone cable required to link each town. Third Edition Copyright © Orchard Publications .5 Solution: For convenience.

5 with the second connection We continue by considering all branches associated with Nodes A . we consider all branches associated with Nodes A and B . Network of Example 16. AC = 4.16. and we find that the shortest is min { AD = 5. we will not have a tree network. DF = 5.59) and the network now is connected as shown in Figure 16. BE = 8 } = AC = 4 (16.15.58) and we add branch CD to the network shown in Figure 16. We find that the minimum of these is min { AD = 5. Numerical Analysis Using MATLAB® and Excel®. BE = 8. B 3 A 4 C 4 10 8 D 5 F E 6 7 2 5 G Figure 16. B 3 A 4 C 5 4 10 5 8 E 6 7 5 F 2 5 G D Figure 16. CF = 10 } = CD = 4 (16. Third Edition Copyright © Orchard Publications 16−17 . Network of Example 16.57) and thus. CF = 10 } = DF = 5 (16. and D and we find that the shortest is min { BE = 8. DE = 6.5 with the third connection Next. because Nodes B and D are already connected. AC is connected to the network as shown in Figure 16.17. BD = 5. otherwise.15. BD = 5. B and C .16. The dotted lines AD and BD have been removed since we no longer need to consider branch AD and BD . C . DG = 7.Network Analysis Next. CD = 4. considering all branches associated with Nodes B .

Network of Example 16.5 with all connections Figure 16.Chapter 16 Optimization Methods B 3 A 4 C 4 8 6 D 5 F 7 2 E 5 G Figure 16.19. We find that min { EG = 5.18 B 3 A 4 C 4 D 5 F 6 7 2 E 5 G Figure 16. DE = 6.19 shows that the minimum distance is 3 + 4 + 4 + 6 + 5 + 2 = 24 kilometers. Network of Example 16. FG = 2 } = FG = 2 (16. we obtain min { BE = 8.18. Third Edition Copyright © Orchard Publications .60) and the network is connected as shown in Figure 16.17. Network of Example 16.19. DG = 7.61) and the complete minimum span tree is shown in Figure 16. B 3 A 4 C 4 D 5 F 2 6 G E Figure 16. DG = 7 } = DE = 6 (16.5 with the fourth connection Continuing. 16−18 Numerical Analysis Using MATLAB® and Excel®.5 with the fifth connection The last step is to determine the shortest branch to Node G .

Summary 16. Numerical Analysis Using MATLAB® and Excel®. we seek to find a tree which contains all nodes. and the sum of the costs (shortest total distance) is a minimum. referred to as the objective.4 Summary • Linear programming is a procedure we follow to maximize or minimize a particular quantity. as defined in this chapter. Third Edition Copyright © Orchard Publications 16−19 . In such problems. is a set of points referred to as nodes and a set of lines referred to as branches. • Network analysis is a method that is used to solve minimum span problems. • Dynamic Programming is based on R. which is dependent on a finite number of variables. • A tree is a connected network which has n branches and n + 1 nodes. These variables may or may not be independent of each another. Bellman’s Principle of Optimality which states that an optimum policy has the property that whatever the initial state and the initial decisions are. the remaining decisions must constitute an optimum policy with regard to the state resulting from the first decision. • A network. and in most cases are subject to certain conditions or limitations referred to as constraints.

The Environmental Protection Agency (EPA) requires that all oil sold must not contain more than 10% lead. D 2 B 3 A 5 2 C 1 7 5 4 1 E 3 4 F 6 L 4 5 K 3 6 4 4 J 2 H 2 7 16−20 Numerical Analysis Using MATLAB® and Excel®. The line segments are directed from left to right and the costs are indicated beside each line segment. He can also buy Grade B oil which contains 15% lead for $20.00 per barrel from one oil refinery company. d 8 b 3 a 5 c 4 6 f 7 5 e 4 k 5 3 5 h 4 m 3.00 per barrel from another oil refinery company.5 Exercises 1. Repeat Example 16. Third Edition Copyright © Orchard Publications . 2. A large oil distributor can buy Grade A oil which contains 7% lead for $25. Use dynamic programming to find the minimum cost route from state a to state m for the line graph shown below.3 for the line graph shown below.Chapter 16 Optimization Methods 16. How many barrels of each grade of oil should he buy so that after mixing the two grades can minimize his cost while at the same time meeting EPA’s requirement? Solve this problem graphically and check your answers with Excel’s Solver.

The third row (zero hours) indicates the commission that he will receive if he just calls instead of visiting them. He will earn the commissions shown on the table below for various visiting times. Third Edition Copyright © Orchard Publications 16−21 .5 for the network shown below. Visit Time (Hours) 0 1 2 3 4 Customer 1 $20 $45 $65 $75 $83 2 $40 $45 $57 $61 $69 3 $40 $52 $62 $71 $78 4 $80 $91 $95 $97 $98 5.Exercises 4. and ignore travel time from customer to customer. A salesman has 4 hours available to visit 4 of his customers. Repeat Example 16. B 30 40 A 50 20 C 40 D 50 F 50 90 70 E Numerical Analysis Using MATLAB® and Excel®. Compute the optimal allocation of time that he should spent with his customers so that he will maximize the sum of his commissions. Consider only integer number of visiting hours.

10 ⁄ 0. x = 0. x = 1 . x and y cannot be negative numbers. The objective is to minimize z = $25.15 = 2 ⁄ 3 and if y = 0 .00x + $20.07x + 0. if x = 0 .Chapter 16 Optimization Methods 16. (3).00y or. if x = 0 . Thus from (2). Each barrel to be sold must not contain more than 10 % lead and since Grade A contains 7 % and Grade B 15 % . x + y = 1 (3) Moreover. z = 25x + 20y (1) We want to minimize (1) because it represents a cost.10 8⁄7 9 ⁄ 7 10 ⁄ 7 x The isoprofit line passes through point b and its coordinates are found by simultaneous solu- 16−22 Numerical Analysis Using MATLAB® and Excel®. Likewise. we must have 0.15y = 0.6 Solutions to End−of−Chapter Exercises 1. To determine the feasible region we plot (2) and (3) where the x and y crossings are found by first setting x = 0 and then y = 0 . not a profit. Third Edition Copyright © Orchard Publications .07x + 0. and (4). Let x be the number of barrels of Grade A oil and y be the number of barrels of Grade B oil.10 (2) The oil of Grade A and Grade B used in each barrel to be sold must be equal to unity.10 ⁄ 0. for simplicity.15y ≤ 0. and if y = 0 . y = 1 . y = 0. y 1 Isoprofit line x+y = 1 2⁄3 a 1⁄3 b c 1⁄7 2⁄7 3⁄7 4⁄7 5⁄7 6⁄7 1 0. therefore x≥0 y ≥ 0 (4) The problem then can be stated as: Minimize (1) subject to the constraints of (2). from (3).07 = 10 ⁄ 7 . Thus.

Solutions to End−of−Chapter Exercises tion of (2) and (3).15*y−0.1563 Check with Excel’s Solver: A 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 B C D E F Optimization . Use Add-Ins if necessary to add it. For convenience.Minimum Cost for Exercise 16. x+y−1) x = 5/8 y = 3/8 Therefore. we use the following MATLAB script: syms x y.125 8 8 8 and this represents his cost per barrel. The isoprofit line is z = 25x + 20y = 23.1 1.125 ⁄ 20 = 1.10 Grade A + Grade B= 1. On the Solver Parameters screen enter the following: Set Target Cell: B15 Equal to: Min By Changing Cells: B12:B13 Click on Add and enter Constraints: B12>=0.374999 Minimum Cost= $23.y]=solve(0. In B17 enter =0.125 and the y −intercept is found by setting x in the equation above to zero and we find that y – intercept = 23. Add B17<=0.15*B13 and in B18 =B12+B13 4. Enter zeros in B12 and B13 2.000001 Note: Contents of A12:A18 are typed-in for information only Numerical Analysis Using MATLAB® and Excel®. In B15 enter =25*B12+20*B13 3. 5.10. Solve Grade A= 0. [x. Third Edition Copyright © Orchard Publications 16−23 .125 Lead Content= 0.= 185 z = 25 × --------.625002 Grade B= 0.= $23. the distributor should buy Grade A oil at the ratio x = 5 ⁄ 8 and Grade B at the ratio y = 3 ⁄ 8 and by substitution into (1) 5 .07*B12+0.+ 20 × 3 -. OK . From the Tools drop menu select Solver . Add B13>=0.07*x+0. Add B18=1.10.

Chapter 16 Optimization Methods 2. g J = 2 . 4 + 5 } = 7 Therefore. the best (shortest) path is a → b → e → h → m 3. g a = min { 3 + g b. the best route from state e to state m passes through state h . g b = min { 8 + g d. g K = 3 . 4 + g k } = min { 3 + 4. d 8 b 3 a 5 c 4 6 f 7 5 e 4 k 5 3 5 h 4 m Since the segments are directed from left to right. Third Edition Copyright © Orchard Publications . state e is the first node where a decision must be made and thus g e = min { 3 + g h. 5 + 11 } = 15 Thus. and g L = 4 16−24 Numerical Analysis Using MATLAB® and Excel®. Next. 5 + g c } = min { 3 + 12. 5 + 7 } = 12 and g c = min { 4 + g e. 6 + g f } = min { 4 + 7. D 2 B 3 A 5 2 C 1 7 5 4 1 E 3 4 F 6 L 4 5 K 3 6 4 4 J 2 H 2 7 The numbers in circles represent the waiting time at these nodes. Last stage: g H = 2 . 5 + g e } = min { 8 + 5. 6 + 7 } = 11 Finally.

The third row (zero hours) indicates the commission that he will receive if he just calls instead of visiting them. 1 + g F } = 2 + 15 = 17 from B to D from C to F Initial stage: g A = 0 + min { 3 + g B.4 It is convenient to rearrange the table as shown below. 4. and ignore travel time from customer to customer. minimum path is from A → B → D → J and the numerical minimum cost is 17 . Visit Time (Hours) 0 1 2 3 4 Solution We will follow the same method as in Example 16.Solution A salesman has 4 hours available to visit of his customers. 4 + g E } = 5 + 9 = 14 g C = 2 + min { 7 + g E.4 . 5 + g C } = 17 from A to B Therefore. 6 + g L } = 4 + 10 = 14 from D to J from E to J from F to K or F to L Next to initial stage: g B = 5 + min { 2 + g D. Compute the optimal allocation of time that he should spend with his customers so that he will maximize the sum of his commissions. A 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 B C D E F G H Exercise 16. Consider only integer number of visiting hours. 1 $20 $45 $65 $75 $83 Customer 2 $40 $45 $57 $61 $69 3 $40 $52 $62 $71 $78 4 $80 $91 $95 $97 $98 continued on the next page Numerical Analysis Using MATLAB® and Excel®.Solutions to End−of−Chapter Exercises Next stage to the left: g D = 1 + min { 6 + g H. 5 + g K } = 3 + 6 = 9 g F = 4 + min { 7 + g K. He will earn the commissions shown on the table below for various visiting times. Third Edition Copyright © Orchard Publications 16−25 . 4 + g J } = 1 + 6 = 7 g E = 3 + min { 4 + g J.

E25+C38. Third Edition Copyright © Orchard Publications . f3(1)+m4(3-1). f3(2)+m4(3-2).C26. f3(3)+m4(4-3). we compute the values of m3(u) and d3(u) m3(4)=max[f3(0)+m4(4-0).C25+B38) m3(0)=max[f3(0)+m4(0-0)] MAX(B25+B38) 80 0 120 0 160 0 1 91 1 132 1 172 0 2 95 2 143 1 183 0 3 97 3 153 2 193 0 4 98 4 162 3 202 0 248 2 98 with d4(4)= 97 with d4(3)= 95 with d4(2)= 91 with d4(1)= 80 with d4(0)= 4 3 2 1 0 162 with d3(4)= 3 153 with d3(3)= 2 143 with d3(2)= 1 132 with d3(1)= 1 120 with d3(0)= 0 continued on the next page.F26) m4(3)= max(B26.Chapter 16 Optimization Methods A 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 f(x) \ x f1(x) f2(x) f3(x) f4(x) B 0 20 40 40 80 C 1 45 45 52 91 D u 2 65 57 62 95 E 3 75 61 71 97 F 4 83 69 78 98 G H I m4(4)= max(B26.C25+E38.C25+D38.D26. f3(3)+m4(3-3)] MAX(B25+E38.C26. 16−26 Numerical Analysis Using MATLAB® and Excel®.F25+B38) m3(3)=max[f3(0)+m4(3-0). u 0 m4(u) d4(u) m3(u) d3(u) m2(u) d2(u) m1(u) d1(u) Next.D25+C38.E25+B38) m3(2)=max[f3(0)+m4(2-0). f3(1)+m4(4-1). f3(2)+m4(4-2).E26) m4(2)= max(B26.D26) m4(1)= max(B26. f3(4)+m4(4-4)] MAX(B25+F38. f3(2)+m4(2-2)] MAX(B25+D38.D25+B38) m3(1)=max[f3(0)+m4(1-0).D25+D38. f3(1)+m4(1-1)] MAX(B25+C38.E26.C26) m4(1)= max(B26) The values of m4(u) and d4(u) are entered in the table below.C26.C25+C38. f3(1)+m4(2-1).D26.

and for the remaining 2 hours he should spend 1 hour with Customer #3 and 1 hour with Customer #4. f1(3)+m2(4-3).E24+C40.C24+B40) m2(0)=max[f2(0)+m3(0-0)] MAX(B24+B40) These values are added to the table above.F25+B40) m2(3)=max[f2(0)+m3(3-0). u=4. 1 hour = Customer #4. 0 hours = Customer #3. f1(4)+m2(4-4)] MAX(B23+F42. f2(3)+m3(4-3). f2(1)+m3(3-1).D24+C40. 2 hours = Customer #2. Rows 40 and 41 Similarly. f2(1)+m3(1-1)] MAX(B24+C40. f2(1)+m3(2-1). Check: Customer #1. f2(2)+m3(4-2). the salesman should spend 2 hours with Customer #1 (d1(4)=2). 1 hour = Total = $65 $40 $52 $91 $248 202 with d2(4)= 0 193 with d2(3)= 0 183 with d2(2)= 0 172 with d2(1)= 0 160 with d2(0)= 0 248 with d1(4)= 2 Numerical Analysis Using MATLAB® and Excel®. Rows 42 and 43 Stage 1 is the last stage and there is only one state associated with it.F23+B42) These two values are the last entries into the table in Cells F44 and F45 The table (Rows 36 through 45) indicates that. we compute the values of m2(u) and d2(u) m2(4)=max[f2(0)+m3(4-0).Solutions to End−of−Chapter Exercises A 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 B C D E F G H I These values are now added to the table above. 0 hours with Customer #2 (d2(4)=0). f2(3)+m3(3-3)] MAX(B24+E40.C23+E42.C24+E40.C24+D40.E24+B40) m2(2)=max[f2(0)+m3(2-0). f2(1)+m3(4-1). f1(1)+m2(4-1). f2(2)+m3(3-2). Third Edition Copyright © Orchard Publications 16−27 . and thus m1(4)=max[f1(0)+m2(4-0).D24+D40. f2(2)+m3(2-2)] MAX(B24+D40.D24+B40) m2(1)=max[f2(0)+m3(1-0). f1(2)+m2(4-2).D23+D42. to achieve the maximum sum of commissions.C24+C40. f2(4)+m3(4-4)] MAX(B24+F40.E23+C42.

70 ] = 50 B 30 40 A 20 D 50 E C 40 F and thus the minimum distance is 20 + 30 + 40 + 40 + 50 = 180 kilometers 16−28 Numerical Analysis Using MATLAB® and Excel®. 90. 20 ] = 20 A→C A→B choose B → D C→F B→E min [ A → B. 40 ] = 40 min [ B → E. 50. D → E. 40 ] = 30 min [ B → D. B 30 40 A 50 20 C 40 D 50 F 50 90 70 E min [ A → B. C → D. F → E ] = min [ 50. D → E. C → F ] = min [ 30. Third Edition Copyright © Orchard Publications . 90. 50.Chapter 16 Optimization Methods 5. 50. C → D. C → F ] = min [ 50. B → E. 40 ] = 40 min [ B → E. D → F. 50. C → F ] = min [ 40. A → C ] = min [ 30.

a recursion is an expression. Third Edition Copyright © Orchard Publications A−1 .Appendix A Difference Equations in Discrete−Time Systems T his appendix is a treatment of linear difference equations with constant coefficients and it is confined to first− and second−order difference equations and their solution. each term of which is determined by application of a formula to preceding terms.1) is referred to as the forcing function. The three steps are as follows: 1.1 Recursive Method for Solving Difference Equations In mathematics. and this method is similar to the classical method of solving linear differential equations with constant coefficients. Higher−order difference equations of this type and their solution is facilitated with the Z−transform. The right side of relation (A. they do not provide a closed−form solution. * For an introduction and applications of the Z-transform please refer to Signals and Systems with MATLAB Computing and Simulink Modeling. yC ( n ) = k1 a1 + k2 a2 n n (A. This method is described in the next section. we can solve difference equations using the Method of Undetermined Coefficients.* A.1) is the same as that used for solving second−order differential equations. Third Edition. Numerical Analysis Using MATLAB® and Excel®. The general (closed-form) solution of relation (A. An example of a recursive method is Newton’s method† for solving non−linear equations. where a 1 and a 2 are also real constants. A. While recursive methods yield a desired result.2 Method of Undetermined Coefficients A second−order difference equation has the form y ( n ) + a1 y ( n – 1 ) + a2 ( n – 2 ) = f ( n ) (A. If a closed−form solution is desired. The solution of a difference equation is often obtained by recursive methods. Obtain the forced response (particular solution) y P ( n ) in terms of an arbitrary real constant k 3 . This difference equation expresses the output y ( n ) at time n as the linear combination of two previous outputs y ( n – 1 ) and y ( n – 2 ) . † Newton’s method is discussed in Chapter 2.1) where a 1 and a 2 are constants and the right side is some function of n . ISBN 0-9744239-9-8. Obtain the natural response (complementary solution) y C ( n ) in terms of two arbitrary real constants k 1 and k 2 .2) 2. that is. such as a polynomial.

Example A.Appendix A Difference Equations in Discrete-Time Systems that is. y ( n ) = yC ( n ) + yP ( n ) = k1 a1 + k2 a2 + yP ( n ) n n (A. As in the case with the solutions of ordinary differential equations with constant coefficients.4). that is. not from the complementary solution y C ( n ) . * Ordinary differential equations with constant coefficients are discussed in Chapter 5.4) using the given initial conditions. 3. we must multiply y P ( n ) by the lowest power of n that will eliminate the duplication. It is best to illustrate the Method of Undetermined Coefficients via examples. yP ( n ) = k3 a3 n (A. Third Edition Copyright © Orchard Publications . the most general form of the particular solution y P ( n ) is the linear combination of these terms.4) 4.1 Forms of the particular solution for different forms of the forcing function Form of forcing function Form of particular solutiona Constant an − a is a constant ab ±n k k − a constant k 0 + k 1 n + k 2 n + … + k k n − k i is constant Expression proportional to b k 1 cos ( n ω ) + k 2 sin ( n ω ) ±n 2 k − a and b are constants acos ( n ω ) or asin ( n ω ) a.3) where the right side of (A.1 Find the total solution for the second−order difference equation 5 -. we must remember that if f ( n ) is the sum of several terms.5) subject to the initial conditions y ( – 2 ) = 25 and y ( – 1 ) = 6 . It is important to remember that the constants k 1 and k 2 must be evaluated from the total solution of (A. Add the natural response (complementary solution) y C ( n ) and the forced response (particular solution) y P ( n ) to obtain the total solution.3) is chosen with reference to Table A. Solve for k 1 and k 2 in (A. Also.1.y ( n – 2 ) = 5 –n -y(n – 1) + 1 y ( n ) – -6 6 n≥0 (A. A−2 Numerical Analysis Using MATLAB® and Excel®.* TABLE A. if a term in y P ( n ) is a duplicate of a term in the complementary solution y C ( n ) .

11) and (A.9) are 1 a 1 = -2 1 a 2 = -3 and by substitution into (A.7) Substitution of y ( n ) = a n into (A.5).= 0 a – -6 6 (A.14) Numerical Analysis Using MATLAB® and Excel®.8) by a n – 2 yields 5 2 -a + 1 -.7) yields 5 n–1 1 n–2 n -a -a + -= 0 a – -6 6 (A.5 + ⎛ -.9) (A.10) The roots of (A.+ k 2 ⎛ -y C ( n ) = k 1 ⎛ -= k1 2 + k2 3 ⎝2⎠ ⎝3⎠ (A.Method of Undetermined Coefficients Solution: 1.⎞ 5–( n – 1 ) + k 3 ⎛ 1 -.⎞ 5 –( n – 2 ) = 5 –n k 3 5 – k 3 ⎛ -⎝6⎠ ⎝6⎠ Division of both sides by 5 –n yields 1⎞ 2 5⎞ .8) Division of (A.5) is 5 1 . Since the forcing function is 5 –n .13) The total solution is the addition of (A. Third Edition Copyright © Orchard Publications A−3 . We assume that the complementary solution y C ( n ) has the form yC ( n ) = k1 a1 + k2 a2 n n (A.11) 2.y ( n – 1 ) + --y(n – 2) = 0 y ( n ) – -6 6 n≥0 (A.6) we obtain 1 ⎞n 1 ⎞n –n –n . that is.13).5 = 1 k 3 1 – ⎛ -⎝6⎠ ⎝6⎠ or k 3 = 1 and thus yP ( n ) = 5 –n (A.6) The homogeneous equation of (A. 5 –n .12) and by substitution into (A. y ( n ) = yC ( n ) + yP ( n ) = k1 2 + k2 3 + 5 –n –n –n (A. we assume that the particular solution is yP ( n ) = k3 5 –n (A.

15) and (A.15) from which 2k 1 + 3k 2 = 1 (A. Third Edition Copyright © Orchard Publications .*3.16) Simultaneous solution of (A.^(−n)+5.14) reduces to 4k 1 + 9k 2 = 0 y ( –1 ) = k1 2 + k2 3 + 5 = 6 1 1 1 (A. Plot for the difference equation of Example A.1.6 1.1 A−4 Numerical Analysis Using MATLAB® and Excel®.1. grid The plot is shown in Figure A.18) To plot this difference equation for the interval 0 ≤ n ≤ 10 .14) we obtain the total solution as 3 ⎞ –n ⎛ 2 ⎞ –n –n .4 1.8 1.3 +5 y ( n ) = y C ( n ) + y P ( n ) = ⎛ -⎝ 3⎠ ⎝2⎠ n≥0 (A.4 0. (A. we use the following MATLAB script: n=0:1:10. 2 1. s y ( – 2 ) = 25 and y ( – 1 ) = 6 . yn=1. i.. (A.16) yields 3 k 1 = -2 2 k 2 = – -3 (A.2 0 0 1 2 3 4 5 6 7 8 9 10 Figure A.17) and by substitution into (A.^(−n).yn). stem(n.*2.6 0./3).Appendix A Difference Equations in Discrete-Time Systems To evaluate the constants k 1 and k 2 we use the given initial conditions.2 1 0.^(−n)−(2.14) reduces to y ( – 2 ) = k 1 2 + k 2 3 + 5 = 25 2 2 2 from which For n = – 1 .5.2 + – -.8 0. For n = – 2 .e.

20) The homogeneous equation of (A.26) contain common terms.Method of Undetermined Coefficients Example A.⎞ + k 2 ( 1 ) n = k 1 2 –n + k 2 y C ( n ) = k 1 ⎛ -⎝2⎠ (A.19) subject to the initial conditions y ( – 2 ) = 0 and y ( – 1 ) = 2 Solution: 1. To avoid the duplication.25) and (A. we choose the particular solution as yP ( n ) = k3 n + k4 3 –n (A.23) are -a1 = 1 2 a2 = 1 and by substitution into (A.19) is 3 -.22) by a n – 2 yields 3 2 -a + 1 -.2 Find the total solution for the second−order difference equation 3 -.21) yields 3 n–1 1 n–2 n -a -a a – -+ -= 0 2 2 (A.24) The roots of (A.21) Substitution of y ( n ) = a n into (A. Third Edition Copyright © Orchard Publications A−5 .26) However.1. We assume that the complementary solution y C ( n ) has the form yC ( n ) = k1 a1 + k2 a2 n n (A. the constants k 2 and k 3 . we would assume that the particular solution is yP ( n ) = k3 + k4 3 –n (A.22) Division of (A. we observe that both relations (A. Since the forcing function is 1 + 3 –n .y(n – 1) + 1 y ( n ) – -2 2 n≥0 (A.20) we obtain n 1 .27) and by substitution of (A.y ( n – 2 ) = 1 + 3 –n .= 0 a – -2 2 (A.y(n – 2) = 0 .23) (A.19) we obtain Numerical Analysis Using MATLAB® and Excel®.y(n – 1) + 1 y ( n ) – -2 2 n≥0 (A. in accordance with the first and third rows of Table A. that is.25) 2.27) into (A.

k – ⎛ -.k n + ⎛ -.30) and (A.k 3 –( n – 2 ) = 1 + 3 –n k 3 n + k 4 3 – ⎛ -⎝2 ⎠ 3 ⎝2⎠ 4 ⎝2⎠ 4 2 3 3⎞ 1 9⎞ 3⎞ 9⎞ –n . s y ( – 2 ) = 0 and y ( – 1 ) = 2 .32) and by substitution into (A. y P ( n ) = 2n + 3 –n (A.k – k 3 = 1 + 3 –n k 4 3 + ⎛ -⎝2 ⎠ 3 Equating like terms.k n – k 3 + ⎛ -.k 3 –( n – 1 ) + -.⎞ k – k3 = 1 ⎝2 ⎠ 3 k4 3 –n = 3 –n and after simplification. Third Edition Copyright © Orchard Publications . (A.27).29) reduces to y ( –1 ) = k1 2 + k2 – 2 + 3 = 2 1 1 from which 2k 1 + k 2 = 1 (A.31) Simultaneous solution of (A..25) and (A.28) The total solution is the addition of (A. (A.k 3 –n = 1 + 3 –n .k ( n – 1 ) – ⎛ -.30) For n = – 1 .29) reduces to y ( –2 ) = k1 2 + k2 – 4 + 9 = 0 2 from which 4k 1 + k 2 = – 5 (A. i. that is.29) To evaluate the constants k 1 and k 2 we use the given initial conditions.e.28).k 3 –n + -k 3 n + k 4 3 – ⎛ -⎝2 ⎠ 3 ⎝2 ⎠ 4 ⎝2⎠ 4 ⎝2 ⎠ 3 2 3 3⎞ –n .31) yields k1 = –3 k2 = 7 (A. k3 = 2 k4 = 1 By substitution into (A. we obtain ⎛3 -.k ( n – 2 ) + ⎛ -.Appendix A Difference Equations in Discrete-Time Systems 3⎞ 3⎞ 1 1⎞ –n . For n = – 2 .29) we obtain the total solution as A−6 Numerical Analysis Using MATLAB® and Excel®. y ( n ) = y C ( n ) + y P ( n ) = k 1 2 + k 2 + 2n + 3 –n –n (A.

*n+3.34) subject to the initial condition y ( – 1 ) = 5 Solution: 1. Plot for the difference equation of Example A.Method of Undetermined Coefficients y ( n ) = y C ( n ) + y P ( n ) = ( – 3 ) 2 + 7 + 2n + 3 –n –n n≥0 (A. yn=(−3).35) The homogeneous equation of (A. stem(n.37) Numerical Analysis Using MATLAB® and Excel®. Third Edition Copyright © Orchard Publications A−7 . we use the following MATLAB script: n=0:1:10.^(−n).2 Example A. We assume that the complementary solution y C ( n ) has the form yC ( n ) = k1 a n (A.2.*2.34) is y ( n ) – 0.9y ( n – 1 ) = 0 n≥0 (A.yn). grid 30 25 20 15 10 5 0 0 1 2 3 4 5 6 7 8 9 10 Figure A.9 ) n–1 n≥0 (A.9y ( n – 1 ) = 0.^(−n)+7+2.35) yields a – 0.5 + ( 0.33) To plot this difference equation for the interval 0 ≤ n ≤ 10 .3 Find the total solution for the first-order difference equation y ( n ) – 0.36) Substitution of y ( n ) = a n into (A.9a n n–1 = 0 (A.

42) The total solution is the addition of (A. we observe that both relations (A.5 + ( 0.9 = 0 a = 0.5 k 3 ( 0.34) we obtain k 2 + k 3 n ( 0.9 ) n (A.9 ) n n n –1 n –1 = 0.9 ) (n – 1) = 0.9 ) = 0. in accordance with the first and third rows of Table A.41).41) and by substitution of (A.9 ) n –1 n and after simplification. 10 n .9k 3 n ( 0.39) and (A.5 + ( 0.9k 3 n ( 0.43) A−8 Numerical Analysis Using MATLAB® and Excel®.9 ) n (A.1k 2 + k 3 n ( 0. Third Edition Copyright © Orchard Publications .5 + ( 0.9 ) = ( 0.9 ) 0.1k 2 + k 3 n ( 0.9k 3 ( 0.39) 2.9 ) ( 0.9 ) + ----9 (A.40) contain common terms. we choose the particular solution as y P ( n ) = k 2 + k 3 n ( 0.9 ) n – 1 .9 ) 0.Appendix A Difference Equations in Discrete-Time Systems Division of (A.9 ) + k 3 ( 0.38) and by substitution into (A. we obtain 0. k2 = 5 ----k 3 = 10 9 By substitution into (A.9 ) n (A.9k 3 ( 0.9 ) n–1 + 0.9 ) – 0.5 + ( 0.n ( 0.9 ) – 0. we would assume that the particular solution is y P ( n ) = k 2 + k 3 ( 0.9 ) n y P ( n ) = 5 + ----9 (A.37) by a n – 1 yields a – 0.9k 2 – 0.9k 3 ( n – 1 ) ( 0.9 ) –1 n Equating like terms.1k 2 = 0.5 + ( 0. that is.9 ) ( 0.9 ) – k 3 n ( 0.41) into (A. that is.9 + 0.9 ) n n n (n – 1) n (n – 1) = 0.9 ) n .9 ) n–1 n–1 = 0.9 ) n–1 0.1.42). Since the forcing function is 0.9 0.n ( 0.1k 2 + k 3 n ( 0. the constants k 1 ( 0. To avoid the duplication.9 (A.5 + ( 0.35) we obtain y C ( n ) = k 1 ( 0. 10 .39) and (A.9 ) n + 5 y ( n ) = y C ( n ) + y P ( n ) = k 1 ( 0.9 ) 0.40) However.9 ) – 0.9 ) n and k 3 ( 0.

^(n-1)+5. Plot for the difference equation of Example A.3 Example A.( – 1 ) ( 0.44) and by substitution into (A.9 ) n–1 n≥0 (A. For n = – 1 .45) To plot this difference equation for the interval 0 ≤ n ≤ 10 .46) Numerical Analysis Using MATLAB® and Excel®.43) we obtain the total solution as y ( n ) = ( 0.yn). i.e. yn=(n+1).9 ) +5 n–1 +5 y ( n ) = ( n + 1 ) ( 0. (A.8y ( n – 1 ) + 0.*(0.4 Find the total solution for the second−order difference equation y ( n ) – 1.9 ) n–1 + n ( 0.9).43) reduces to 10 –1 . grid 10 9 8 7 6 5 4 3 2 1 0 0 1 2 3 4 5 6 7 8 9 10 Figure A.. y ( – 1 ) = 5 . we use the following MATLAB script: n=0:1:10.81y ( n – 2 ) = 2 –n n≥0 (A.9 ) –1 + 5 = 5 y ( – 1 ) = k 1 ( 0.9 ) + ----9 10 ----. Third Edition Copyright © Orchard Publications A−9 .k – 100 -------.= 0 9 1 81 from which 10 k 1 = ----9 (A.3.Method of Undetermined Coefficients To evaluate the constant k 1 we use the given initial condition. stem(n.

49) Division of (A.8 ) 2 –n –( n – 1 ) + k 3 ( 0.9 and as in the case of ordinary differential equations.6 + 3.8 ) 2 + ( 0.81a n–2 = 0 (A. k 3 2 – k 3 ( 1.9 ) + k 2 n ( 0. we accept the complementary solution to be of the form y C ( n ) = k 1 ( 0.46) is y ( n ) – 1.53) –( n – 2 ) and by substitution into (A.50) (A.48) yields a – 1.= ----0. Since the forcing function is 2 –n .50) are repeated roots. 1.51) The roots of (A. we assume that the particular solution is yP ( n ) = k3 2 –n (A.81 ) 2 ] = 1 k 3 [ 1 – 3.47) The homogeneous equation of (A.49) by a n – 2 yields a – 1.52) 2. Third Edition Copyright © Orchard Publications .81 ) 2 = 2 –n Division of both sides by 2 –n yields k 3 [ 1 – ( 1. a 1 = a 2 = 0. We assume that the complementary solution y C ( n ) has the form yC ( n ) = k1 a1 + k2 a2 n n (A.8a + 0.81y ( n – 2 ) = 0 n≥0 (A.24 ] = 1 1 25 k 3 = --------.8y ( n – 1 ) + 0.64 16 2 A−10 Numerical Analysis Using MATLAB® and Excel®.46).Appendix A Difference Equations in Discrete-Time Systems subject to the initial conditions y ( – 2 ) = 25 and y ( – 1 ) = 6 Solution: No initial conditions are given and thus we will express the solution in terms of the unknown constants.48) Substitution of y ( n ) = a n into (A.8a n n–1 + 0.9 ) n n (A.81 = 0 2 (A. that is.

9 ) 2 n (A.8y ( n – 1 ) + 0.60) Numerical Analysis Using MATLAB® and Excel®. The next choice would be a term of the form k 3 n ( 0.81y ( n – 2 ) = ( 0.9 ) + k 2 n ( 0.57).2 y ( n ) = y C ( n ) + y P ( n ) = k 1 ( 0.57).9 ) + k 2 n ( 0. But this is of the same form as the first term on the right side of (A.4 where the forcing function is ( 0.1 indicates that a good choice for the particular solution would be k 3 ( 0.6 Find the particular solution for the first-order difference equation nπ ⎞ y ( n ) – 0.55) Example A.9 ) n instead of 2 –n where we found that the complementary solution is y C ( n ) = k 1 ( 0. Therefore.9 ) n n≥0 (A. 25 ⎞ –n n n . the particular solution has the form nπ π -⎞ + k 2 cos ⎛ n -----⎞ y P ( n ) = k 1 sin ⎛ ----⎝ 2⎠ ⎝ 2⎠ (A.9 ) n .52) and (A.9 ) + ⎛ ----⎝ 16 ⎠ (A.5y ( n – 1 ) = sin ⎛ ----⎝2⎠ n≥0 (A.54) The total solution is the addition of (A.54).Method of Undetermined Coefficients and thus –n 25 .58) Example A. Third Edition Copyright © Orchard Publications A−11 .57) Row 3 in Table A. that is.56) what would be the appropriate choice for the particular solution? Solution: This is the same difference equation as that of Example A.59) Solution: From Row 4 in Table A.1 we see that for a sinusoidal forcing function. the proper choice would be y P ( n ) = k 3 n ( 0.9 ) n n (A.9 ) n but this is of the same form as the second term on the right side of (A.⎞2 y P ( n ) = ⎛ ----⎝ 16 ⎠ (A.5 For the second−order difference equation y ( n ) – 1.

– 0.– -⎝ 2 ⎠ 2 2 and by substitution into (A.59) is nπ π 4 -⎞ -----⎞ – 2 .61) n π⎞ n π⎞ n π⎞ n π⎞ nπ⎞ .= sin ⎛ ----k 1 sin ⎛ ----⎝ 2⎠ ⎝ 2⎠ ⎝ 2⎠ ⎝ 2⎠ ⎝2⎠ (A.+ 0.+ k 2 cos ⎛ ----.cos ⎛ ----y P ( n ) = -⎝ ⎝ ⎠ 2⎠ 2 5 5 (A. nπ ⎞ nπ π sin ----.5k 2 = 1 0.– 0.– 0.63) (A. sin ⎛ θ – π -.5k 1 cos ⎛ ----. the particular solution of (A. Third Edition Copyright © Orchard Publications . we obtain k 1 – 0.sin ⎛ n -.+ k 2 cos ⎛ ----.– -.⎞ = – cos θ ⎝ 2⎠ π⎞ cos ⎛ θ – -.60) into (A.+ k 2 cos ⎛ ----.5k 1 + k 2 = 0 (A.= sin ⎛ ----k 1 sin ⎛ ----⎝ 2⎠ ⎝ 2⎠ ⎝2⎠ 2 2 nπ⎞ n π⎞ n π⎞ π π nπ π .5k 2 cos ------------------.Appendix A Difference Equations in Discrete-Time Systems and by substitution of (A.64) yields 4 k 1 = -5 2 k 2 = – -5 Therefore.– 0.5k 2 cos ----.5k 1 sin ------------------.5k 1 sin n ----.63) and (A.61) From trigonometry.– -.59) n π⎞ n π⎞ ( n – 1 )π ( n – 1 )π nπ ⎞ .– 0.= sin ⎛ ----k 1 sin ⎛ ----⎝ ⎝ 2⎠ ⎝ 2⎠ 2⎠ 2 2 2 2 (A.= – cos ⎛ ----⎝ 2 ⎠ 2 2 nπ nπ π -⎞ cos ----.64) and simultaneous solution of (A.= sin θ ⎝ 2⎠ Then.65) A−12 Numerical Analysis Using MATLAB® and Excel®.5k 2 sin ⎛ ----.– -.62) Equating like terms.= sin ⎛ ----.

1. and revise our models in either environment at any point.1 Simulink and its Relation to MATLAB The MATLAB® and Simulink® environments are integrated into one entity. This author feels that we can best introduce Simulink with a few examples.5 V . dv C i = i L = i C = C -------dt (B. Circuit for Example B. it is highly recommended that the novice to MATLAB reader reviews Chapter 1 which serves as an introduction to MATLAB.1 For this example. R L 1⁄4 H C 1Ω + vC ( t ) − + − i(t) 4⁄3 F vs ( t ) = u0 ( t ) Figure B.1) and by Kirchoff’s voltage law (KVL). simulate.1) into (B. Example B. and for this purpose.1 For the circuit of Figure B. and v c ( 0 − ) = 0. We will compute v c ( t ) . We invoke Simulink from within MATLAB. and thus we can analyze.2) yields Numerical Analysis Using MATLAB® and Excel®. B.+ vC = u0 ( t ) Ri L + L -----dt (B. We will introduce Simulink with a few illustrated examples. di L .Appendix B Introduction to Simulink® T his appendix is a brief introduction to Simulink.1. Some familiarity with MATLAB is essential in understanding Simulink. Third Edition Copyright © Orchard Publications B−1 .2) Substitution of (B. the initial conditions are i L ( 0 − ) = 0 .

Third Edition Copyright © Orchard Publications .+ 4 -------2 dt dt 2 2 (B. and the solution using Simulink follows.+ 3v C = 3u 0 ( t ) ---------. and thus the complete solution will consist of the sum of the forced response and the natural response.+ vC = u0 ( t ) .+ LC ---------. It is obvious that the solution of this equation cannot be a constant since the derivatives of a constant are zero and thus the equation is not satisfied. Therefore. are possible candidates since their derivatives have the same form but alternate in sign. It can be shown* that if k 1 e 1 and k 2 e 2 where k 1 and k 2 are constants and s 1 and s 2 are the roots of the characteristic equation of the homogeneous part of the given differential equation.Introduction to Simulink® d vC dv C .5) To appreciate Simulink’s capabilities. the total solution will –s1 t –s2 t v c ( t ) = natural response + forced response = v cn ( t ) + v cf ( t ) = k 1 e + k2 e + v cf ( t ) (B. First Method − Assumed Solution Equation (B. the natural response is the sum of the terms k 1 e be –s1 t –s t –s t – at and k 2 e –s2 t . for comparison. the solution cannot contain sinusoidal functions (sine and cosine) since the derivatives of these are also sinusoids. B−2 Numerical Analysis Using MATLAB® and Excel®.4) t>0 (B. non−homogeneous differential equation with constant coefficients.+ vC = u0 ( t ) RC -------2 dt dt 2 (B. decaying exponentials of the form ke where k and a are constants.5) is a second−order.---------.-------3 dt 2 3 dt dv C d vC .3) Substituting the values of the circuit constants and rearranging we obtain: 1 d v C 4 dv C -. three different methods of obtaining the solution are presented. Appendix B for a thorough discussion.+ -.6) The values of s 1 and s 2 are the roots of the characteristic equation * Please refer to Circuit Analysis II with MATLAB Applications.+ 3v C = 3 ---------.+ 4 -------2 dt dt 2 dv C d vC . ISBN 0−9709511−5−9. However. Also.

7) Solution of (B.7) yields of s 1 = – 1 and s 2 = – 3 and with these values (B. using v C ( 0 ) = 0. -------.12) Also.9) is a constant. we evaluate it at t = 0 .8)..= ---=0 C C (B.13).5). we differentiate (B.Simulink and its Relation to MATLAB s + 4s + 3 = 0 2 (B. and equate it with (B.9) we obtain 0 + 0 + 3k 3 = 3 or v Cf = k 3 = 1 (B.11) The constants k 1 and k 2 will be evaluated from the initial conditions.11) at t = 0 .+ 4 -------2 dt dt t>0 (B. Third Edition Copyright © Orchard Publications B−3 .6) is written as vc ( t ) = k1 e + k2 e –t –3 t + v cf ( t ) (B. the forced response will also be a constant and we denote it as v Cf = k 3 . we obtain v C ( 0 ) = k 1 e + k 2 e + 1 = 0. Thus.5 0 0 (B.14) we obtain Numerical Analysis Using MATLAB® and Excel®.5 V and evaluating (B.8) The forced component v cf ( t ) is found from (B.= --i L = i C = C -------dt dt C dv C -------dt iL ( 0 ) 0 = ----------.5 k 1 + k 2 = – 0. dv C -------dt = – k 1 – 3k 2 t=0 (B.9) Since the right side of (B.+ 3v C = 3 ---------.13) t=0 Next. 2 dv C d vC . i.13) and (B.14) By equating the right sides of (B. yields the total solution as v C ( t ) = v Cn ( t ) + v Cf = k 1 e + k 2 e –t –3 t +1 (B. First. By substitution into (B. and dv C dv C i L -.10) Substitution of this value into (B.11).e.

Third Edition Copyright © Orchard Publications .2.-.= -.25s C 3 ⁄ 4s + VC ( s ) − Vs ( s ) = 1 ⁄ s + − I(s) 0. for our example.25*exp(−3*t)+1.75 e + 0. gives k 1 = – 0.16).8).1 B−4 Numerical Analysis Using MATLAB® and Excel®.25e –t –3 t + 1 ) u0 ( t ) (B. the solution has been verified by MATLAB.5 ⁄ s + V (0) C − Figure B.12) and (B. By substitution into (B.2. we obtain the total solution as v C ( t ) = ( – 0.15) Simultaneous solution of (B.15).17) Second Method − Using the Laplace Transformation The transformed circuit is shown in Figure B. y1=diff(y0) % Define symbolic variable t % The total solution y(t).16) Check with MATLAB: syms t y0=−0. vc(t) % The first derivative of y(t) y1 = 3/4*exp(-t)-3/4*exp(-3*t) y2=diff(y0.75*exp(−t)+0.25 .Introduction to Simulink® – k 1 – 3k 2 = 0 (B. Using the expression for v C ( t ) in (B. Transformed Circuit for Example B. R 1 L 0.75 and k 2 = 0.2) % The second derivative of y(t) y2 = -3/4*exp(-t)+9/4*exp(-3*t) y=y2+4*y1+3*y0 % Summation of y and its derivatives y = 3 Thus. we find the expression for the current as dv C 4 ⎛ 3 –t 3 – 3t ⎞ – t – 3t .e – -i = i L = i C = C --------= e –e A -e ⎠ dt 3⎝ 4 4 (B.

please refer to Chapter 5.+ --------------.+ --------------V C ( s ) = ----------------------------------s (s + 1) (s + 3) s(s + 1)(s + 3) 2 Taking the Inverse Laplace transform‡ we find that v C ( t ) = 1 – 0. ISBN 09744239-9-8.75 s = –1 = 0. Signals and Systems with MATLAB Computing and Simulinl Modeling.+ v C = u 0 ( t ) ** Ri L + L -----dt * For derivation of the voltage division and current division expressions.25 – 0.Simulink and its Relation to MATLAB By the voltage division* expression.+ ------------------------------------------------s (s + 1) (s + 3) s(s + 1 )( s + 3) (B. this text. ‡ For an introduction to Laplace Transform and Inverse Laplace Transform.18) 0. † Partial fraction expansion is discussed in Chapter 12.5s + 2s + 3.18) 0.5s s(s + 3) 0. For distinction.75.= -------------------------------V C ( s ) = --------------------------------------------2 ( 1 + 0.5s + 2s + 31.= 1 -.+ 0. Signals and Systems with MATLAB Computing and Simulinl Modeling.5 -----.25e –t – 3t Third Method − Using State Variables di L .= -----------------------------------. Third Edition Copyright © Orchard Publications B−5 . ISBN 0-9744239-9-8. For a detailed discussion on State−Space and State Variables Analysis. 2 3 ⁄ 4s 0.† we let 2 r2 r3 0.5 -----.25 s = –3 and by substitution into (B.5 ------⎞ + 0.5s + 2s + 3 r 3 = --------------------------------s(s + 1) 2 2 2 = 1 s=0 = – 0.5s + 2s + 3. ISBN 0−9709511−2−4.25s + 3 ⁄ 4s ) ⎝ s s s ⎠ s s(s + 1)(s + 3) s ( s + 4s + 3 ) Using partial fraction expansion. Numerical Analysis Using MATLAB® and Excel®.5s + 2s + 3 r 1 = --------------------------------(s + 1)(s + 3) + 2s + 3 --------------------------------r 2 = 0.= r 1 --. in State−Space and State Variables Analysis. we will denote the Unit Step Function as u0 ( t ) .5 . please refer to Circuit Analysis I with MATLAB Applications. ** Usually.– 0. please refer to Chapters 2 and 3.⋅ ⎛1 . u ( t ) denotes any input.75e + 0.+ --------------0.

or dv C i L = C -------dt dv C · ·2 = 4 --x . (B.= ( –1 ) iL – vC + 1 4 dt or di L -----. Then.19) Next. * The notation x † The detailed solution of (B. we obtain di L 1 -.22).23) Solution† of (B.23) yields · (x dot) is often used to denote the first derivative of the function x . ·1 x x = –4 –4 1 + 4 u0 ( t ) · 3 ⁄ 4 0 x2 0 x2 (B. we define the state variables x 1 = i L and x 2 = v C . L* · 1 = di x -----dt (B.20) and dv C · 2 = -------x dt (B.= – 4i L – 4v C + 4 dt (B.-----. and (B. Signals and Systems with MATLAB Computing and Simulinl Modeling.22) Therefore. B−6 Numerical Analysis Using MATLAB® and Excel®.19). that is.20).= Cx x 1 = i L = C -------3 2 dt ·2 = 3 --x x 4 1 (B. from (B.Introduction to Simulink® By substitution of given values and rearranging. ISBN 0-9744239-9-8. and thus.21) Also. x · = dx ⁄ dt .23) is given in Chapter 5. we obtain the state equations · = – 4x – 4x + 4 x 1 1 2 · = 3 --x x 2 4 1 and in matrix form. Third Edition Copyright © Orchard Publications .

Third Edition Copyright © Orchard Publications B−7 .25) Modeling the Differential Equation of Example B. The right side is referred to as the Contents Pane and displays the blocks that reside in the library currently selected in the Tree Pane.= – 4 -------2 dt dt (B. x1 = iL = e –e (B.75 e + 0. In Figure B. The Simulink icon Upon execution of the Simulink command. It appears on the top bar on MATLAB’s Command prompt.26) A block diagram representing relation (B. the Commonly Used Blocks appear as shown in Figure B. we must first invoke MATLAB.3.* * Henceforth.26) above is shown in Figure B. all Simulink block diagrams will be referred to as models.25e –t (B.3.5. We will use Simulink to draw a similar block diagram. Let us express the differential equation of Example B.Simulink and its Relation to MATLAB x1 x2 = e –e –t –t – 3t – 3t 1 – 0.24) – 3t and x 2 = v C = 1 – 0. In the MATLAB Command prompt.75e + 0. Make sure that Simulink is installed in your system.25e –t – 3t Then.– 3v C + 3u 0 ( t ) ---------. we type: simulink Alternately. Figure B.1 with Simulink To run Simulink. Numerical Analysis Using MATLAB® and Excel®. the left side is referred to as the Tree Pane and displays all Simulink libraries installed. we can click on the Simulink icon shown in Figure B.1 as 2 dv C d vC .4.4.

On the Simulink Library Browser.4. we perform the following steps: 1. Third Edition Copyright © Orchard Publications .26) using Simulink.26) To model the differential equation (B. The Simulink Library Browser d vC ---------2 dt 2 u0 ( t ) 3 Σ ∫ dt −4 dv C -------dt ∫ dt vC −3 Figure B.Introduction to Simulink® Figure B. we click on the leftmost icon shown as a blank page on the top title bar.6. Block diagram for equation (B. B−8 Numerical Analysis Using MATLAB® and Excel®.5. A new model window named untitled will appear as shown in Figure B.

8. it can be recalled by clicking on the white page icon on the top bar of the Simulink Library Browser. and on the right side we scroll down until we see the unit step function shown as Step. and all saved files will have this appearance. The triangle on the right side of the unit step function block and the > symbols on the left and right sides of the gain block are connection points. 3.8). we observe that we need to connect an amplifier with Gain 3 to the unit step function block. We save file Equation_1_26 using the File drop menu on the Equation_1_26 model window (right side of Figure B.8. 4. and we drag it into the Equation_1_26 model window which now appears as shown in Figure B.7. See Figure B.6 is the model window where we enter our blocks to form a block diagram. If the Equation_1_26 model window is no longer visible.mdl. we click on the Sources appearing on the left side list.Simulink and its Relation to MATLAB Figure B. and draw a straight line to connect the two Numerical Analysis Using MATLAB® and Excel®. We select it. We point the mouse close to the connection point of the unit step function until is shows as a cross hair. Figure B. We save this as model file name Equation_1_26. With reference to block diagram of Figure B.6 where we choose Save as and name the file as Equation_1_26. The gain block in Simulink is under Commonly Used Blocks (first item under Simulink on the Simulink Library Browser).6. Model window for Equation_1_26.5. Third Edition Copyright © Orchard Publications B−9 .7. This is done from the File drop menu of Figure B. We choose the gain block and we drag it to the right of the unit step function.8. See Figure B. See Figure B.mdl file 2. The new model window will now be shown as Equation_1_26. The window of Figure B. Simulink will add the extension . With the Equation_1_26 model window and the Simulink Library Browser both visible. The Untitled model window in Simulink.

Third Edition Copyright © Orchard Publications . we change the gain from 1 to 3.9.* We double−click on the gain block and on the Function Block Parameters. Dragging the unit step function into File Equation_1_26 Figure B. and left−clicking on the destination block.8.9. See Figure B. B−10 Numerical Analysis Using MATLAB® and Excel®. Figure B. then holding down the Ctrl key.Introduction to Simulink® blocks. File Equation_1_26 with added Step and Gain blocks * An easy method to interconnect two Simulink blocks is by clicking on the source block to select it.

we click on the Gain block. Finally. Third Edition Copyright © Orchard Publications B−11 .11. we double−click on these gain blocks and on the Function Block Parameters window. From the Commonly Used Blocks of the Simulink Library Browser. We then connect the output of the of the gain block to the first input of the adder block as shown in Figure B. File Equation_1_26 complete block diagram Numerical Analysis Using MATLAB® and Excel®. and we copy and paste it twice. and on the Function Block Parameters window which appears. and we connect it to the output of the Add block. we change the gains from to −4 and −3 as shown in Figure B. we drag it into the Equation_1_26 model window.12. we add the Scope block which is found in the Commonly Used Blocks on the Simulink Library Browser. We flip the pasted Gain blocks by using the Flip Block command from the Format drop menu.Simulink and its Relation to MATLAB 5. We click on the text “Integrator” under the first integrator block.11. we change the text “Integrator 1” under the second Integrator to “Integrator 2” as shown in Figure B.10. We select it. and we change it to Integrator 1. The adder block appears on the right side of the Simulink Library Browser under Math Operations. We double click it. and we label these as Gain 2 and Gain 3. We repeat this step and to add a second Integrator block. File Equation_1_26 with added gain block 6. Next. Then. Figure B. and we drag it into the Equation_1_26 model window. To complete the block diagram.12. File Equation_1_26 with the addition of two integrators 7. we need to add a thee−input adder. Figure B.10. Figure B. we specify 3 inputs. we choose the Integrator block.

1. Figure B. This is done by specifying the simulation time to be 10 seconds on the Configuration Parameters from the Simulation drop menu. B−12 Numerical Analysis Using MATLAB® and Excel®.13.13. The waveform for the function v C ( t ) for Example B. 9.5 V are entered by − double clicking the Integrator blocks and entering the values 0 for the first integrator. is to use State− Space block from Continuous in the Simulink Library Browser. and then clicking on the Autoscale icon.Introduction to Simulink® 8. and v c ( 0 ) = 0. Figure B. and 0. We can start the simulation on Start from the Simulation drop menu or by clicking on the icon. Obtaining the function v C ( t ) for Example B.1 Another easier method to obtain and display the output v C ( t ) for Example B. we double click on the Scope block.5 for the second integrator.14. Third Edition Copyright © Orchard Publications . The initial conditions i L ( 0 − ) = C ( dv C ⁄ dt ) t=0 = 0 .1 with the State−Space block. To see the output waveform.14. We also need to specify the simulation time. we obtain the waveform shown in Figure B. as shown in Figure B.

reside in the MATLAB Workspace. ·1 x x = –4 –4 1 + 4 u0 ( t ) · 3 ⁄ 4 0 x2 0 x2 The output equation is or y = Cx + du y = [0 1] x1 x2 + [0 ]u We double−click on the State−Space block. We issue the command who to see those variables. This block writes its output to an array or structure that has the name specified by the block's Variable name parameter.14 writes its input to the workspace.15. Third Edition Copyright © Orchard Publications B−13 . From Equation B.23. Page B−6. The data and variables created in the MATLAB Command window. Figure B. and in the Functions Block Parameters window we enter the constants shown in Figure B.15. This gives us the ability to delete or modify selected variables. Numerical Analysis Using MATLAB® and Excel®. The Function block parameters for the State−Space block.Simulink and its Relation to MATLAB The simout To Workspace block shown in Figure B.

Example B. As before.27) where y ( t ) is the output representing the voltage or current of the network. Third Edition Copyright © Orchard Publications . a.2 A fourth−order network is described by the differential equation 4 y d y dy d y+a d . we double click on the Scope block.5. and the initial conditions are y ( 0 ) = y' ( 0 ) = y'' ( 0 ) = y''' ( 0 ) = 0 .16.1 with the State−Space block. We will express (B. x2=0. and u ( t ) is any input.Introduction to Simulink® The initials conditions [ x1 x2 ]' are specified in MATLAB’s Command prompt as x1=0. The state−space block is the best choice when we need to display the output waveform of three or more variables as illustrated by the following example.+ a 2 ------. and to see the output waveicon. we form. The waveform for the function v C ( t ) for Example B.+ a0 y ( t ) = u ( t ) .+ a 1 ------------3 -------3 2 4 dt dt dt dt 3 2 (B. Figure B.16.27) as a set of state equations B−14 Numerical Analysis Using MATLAB® and Excel®. to start the simulation we click clicking on the icon. and then clicking on the Autoscale obtain the waveform shown in Figure B.

x 2.31) and in matrix form ·1 x ·2 x ·3 x ·4 x 0 0 = 0 –a0 1 0 0 –a1 0 1 0 –a2 0 0 1 –a3 0 x2 + 0 u(t) x3 0 1 x4 x1 (B.28).+ y ( t ) = sin t ------.32) In compact form.29) In our set of state equations.28) subject to the initial conditions y ( 0 ) = y' ( 0 ) = y'' ( 0 ) = y''' ( 0 ) = 0 .28) is of fourth−order. a 1. a 2.34) Also. Third Edition Copyright © Orchard Publications B−15 . has the solution y ( t ) = 0. and using Simulink. we must define four state variables that will be used with the four first−order state equations.33) (B.+ 2 ------2 4 dt dt 2 (B. and x 4 . and we relate them to the terms of the given differential equation as x1 = y ( t ) dy x 2 = ----dt d y x 3 = -------2 dt 2 d y x 4 = -------3 dt 3 (B. we will select appropriate values for the coefficients a 3. the output is where y = Cx + du Numerical Analysis Using MATLAB® and Excel®.125 [ ( 3 – t ) – 3t cos t ] 2 (B. 1. we will display the waveform of the output y ( t ) . The differential equation of (B. and a 0 so that the new set of the state equations will represent the differential equation of (B.32) is written as · = Ax + bu x (B.Simulink and its Relation to MATLAB b. It is known that the solution of the differential equation 4 d y d y .30) We observe that · 1 = x2 x · 2 = x3 x · 3 = x4 x d y = x · 4 = –a0 x1 –a1 x2 – a2 x3 –a3 x4 + u ( t ) --------4 dt 4 (B. (B. therefore. x 3 . We denote the state variables as x 1.

36) 2. 0 1 · = x .34) is expressed as x1 y = [1 0 0 0] ⋅ x2 x3 x4 + [0 ]u( t) (B. Third Edition Copyright © Orchard Publications . the differential equation of (B. By inspection. 0 b= 0. 1 –a3 x1 x= x2 x3 x4 y ( t ) = x1 . and u = u ( t ) (B.27) will be reduced to the differential equation of (B. 0 1 · = x .38) Since the output is defined as in matrix form it is expressed as B−16 Numerical Analysis Using MATLAB® and Excel®. 0 b= 0.28) can be expressed in state−space form as ·1 x ·2 x ·3 x ·4 x 0 0 = 0 –a0 1 0 0 0 0 1 0 –2 0 0 1 0 0 + 0 sin t x3 0 1 x4 x2 x1 (B.28) if we let a3 = 0 a2 = 2 a1 = 0 a0 = 1 u ( t ) = sin t and thus the differential equation of (B.35) and since the output is defined as relation (B. and u = sin t (B.37) where ·1 x ·2 x ·3 x ·4 x 0 0 A= 0 –a0 1 0 0 0 0 1 0 –2 0 0 .Introduction to Simulink® ·1 x ·2 x ·3 x ·4 x 0 0 A= 0 –a0 1 0 0 –a1 0 1 0 –a2 0 0 . 1 0 x1 x= x2 x3 x4 y ( t ) = x1 .

we double−click on the state−space block and we enter the following parameter values in the Function Block Parameters: A: [0 1 0 0. and we save this model as Example_1_2.2 We now double−click on the Signal Generator block and we enter the following in the Function Block Parameters: Wave form: sine Time (t): Use simulation time Amplitude: 1 Frequency: 2 Units: Hertz Next. 0 0 0 1. Block diagram for Example B.39) We invoke MATLAB. We also add the Display block found under Sinks on the Simulink Library Browser. On the Simulink Library Browser we select Sources. 0 0 1 0. we start Simulink by clicking on the Simulink icon. −a0 −a1 −a2 −a3] B: [0 0 0 1]’ C: [1 0 0 0] D: [0] Initial conditions: x0 Numerical Analysis Using MATLAB® and Excel®.17. on the Simulink Library Browser we click on the Create a new model (blank page icon on the left of the top bar). Third Edition Copyright © Orchard Publications B−17 . we click and drag the State−Space block from the Continuous on Simulink Library Browser. Figure B.17.Simulink and its Relation to MATLAB x1 y = [1 0 0 0] ⋅ x2 x3 x4 + [ 0 ] sin t (B. We connect these four blocks and the complete block diagram is as shown in Figure B. we drag the Signal Generator block on the Example_1_2 model window. and we click and drag the Scope block from the Commonly Used Blocks on the Simulink Library Browser .

Display blocks found in the Sinks library.Introduction to Simulink® Absolute tolerance: auto Now. z 2 .3 Using Algebraic Constraint blocks found in the Math Operations library. we will create a model that will produce the simultaneous solution of three equations with three unknowns. x0=[0 0 0 0]’.40) B−18 Numerical Analysis Using MATLAB® and Excel®. We could have obtained the solution of Example B. Example B. we double click on the Scope block. Examples B. we obtain the waveform shown in Figure B. and z 3 in the system of the equations a1 z1 + a2 z2 + a3 z3 + k1 = 0 a4 z1 + a5 z2 + a6 z3 + k2 = 0 a7 z1 + a8 z2 + a9 z3 + k3 = 0 (B.2 The Display block in Figure B. The model will display the values for the unknowns z 1 .17 shows the value at the end of the simulation stop time. then clicking on the Autoscale icon.18. We change the Simulation Stop time to 25 .2 using four Integrator blocks by this approach would have been more time consuming. Figure B. Third Edition Copyright © Orchard Publications . a3=0.1 and B. and Gain blocks found in the Commonly Used Blocks library. we switch to the MATLAB Command prompt and we type the following: >> a0=1. a1=0.18. To see the output waveform. a2=2. Waveform for Example B. and we start the simulation by clicking on the icon.2 have clearly illustrated that the State−Space is indeed a powerful block.

k2=−7. z 2 = – 3 . By default. k1=−8. The Algebraic Constraint block constrains the input signal f ( z ) to zero and outputs an algebraic state z .Simulink and its Relation to MATLAB The model is shown in Figure B. a3=−1.19.3 Next. a2=−3. and z 3 = 5 . a4=1. a5=5. we observe the values of the unknowns as z 1 = 2 . k3=5. Figure B. the Initial guess parameter is zero.19. Model for Example B. a9=2. This enables us to specify algebraic equations for index 1 differential/algebraic systems (DAEs). The block outputs the value necessary to produce a zero at the input. a7=−6.19. After clicking on the simulation icon. we go to MATLAB’s Command prompt and we enter the following values: a1=2. a8=1.These values are shown in the Display blocks of Figure B. a6=4.. Numerical Analysis Using MATLAB® and Excel®. We can improve the efficiency of the algebraic loop solver by providing an Initial guess for the algebraic state z that is close to the solution value. Third Edition Copyright © Orchard Publications B−19 ... The output must affect the input through some feedback path.

we can group all blocks and lines in the model of Figure B. The model of Figure B. a3=4. ISBN 0−9744239−7−1. Introduction to Simulink with Engineering Applications.1. k3=9. Section 2. This model can be seen by typing thermo in the MATLAB Command prompt.2 Simulink Demos At this time.. k1=14. a7=−8. a6=9.. we choose Create Subsystem from the Edit menu. k2=−6.19 except the display blocks. † The contents of the Subsystem block are not lost.20 show the values of z 1 . Introduction to Simulink with Engineering Applications.. Third Edition Copyright © Orchard Publications . a2=−1. Figure B. a9=15. z 2 .20† where in MATLAB’s Command prompt we have entered: a1=5. B. ISBN 0−9744239−7−1. B−20 Numerical Analysis Using MATLAB® and Excel®. the reader can follow the steps delineated in The MathWorks Simulink User’s Manual to run the Demo Models beginning with the thermo model. the reader with no prior knowledge of Simulink. The Subsystem block replaces the inputs and outputs of the model with Inport and Outport blocks. * The Subsystem block is described in detail in Chapter 2. Page 2−2. Page 2−2. These blocks are described in Section 2.Introduction to Simulink® An outstanding feature in Simulink is the representation of a large model consisting of many blocks and lines.* For instance.1. Chapter 2. Then. It is highly recommended that the reader becomes familiar with the block libraries found in the Simulink Library Browser. a4=11. We can double−click on the Subsystem block to see its contents.20. should be ready to learn Simulink’s additional capabilities. to be shown as a single Subsystem block. and this model will be shown as in Figure B. a8=4.19 represented as a subsystem The Display blocks in Figure B. and z 3 for the values specified in MATLAB’s Command prompt. a5=6.

. −3 −6 8 1.1) and it is computed with the MATLAB function norm(A). We will introduce a reference against which the determinant can be measured to classify a matrix as a well− or ill−conditioned. C. is defined as A = 2 ∑ ∑ Aij 2 i=1j=1 n n (C. defined as –2 5 A = –3 –6 7 –5 4 –9 –4 8 3 –8 9 1 2 –1 Solution: At the MATLAB command prompt. An example is the two−dimensional Euclidean space denoted as R . (2.1 The Norm of a Matrix A norm is a function which assigns a positive length or size to all vectors in a vector space. The Euclidean norm assigns to each vector the length of its arrow.5)) are usually drawn as arrows in a two−dimensional cartesian coordinate system starting at the origin (0.g. 4 −9 −8 −1]. norm(A) and MATLAB outputs Numerical Analysis Using MATLAB® and Excel®. we enter A=[−2 5 −4 9. 7 −5 3 2.Appendix C Ill−Conditioned Matrices T his appendix supplements Chapters 4 and 14 with concerns when the determinant of the coefficient matrix is small.1 Using the MATLAB function norm(A). The elements of the Euclidean vector space (e. The Euclidean norm of a matrix A . denoted as A . Third Edition Copyright © Orchard Publications C− 1 . other than the zero vector.0). compute the Euclidean norm of the matrix A . Example C.

Page 4−22. compute the condition number of the matrix A defined as –2 5 A = –3 –6 7 –5 4 –9 –4 8 3 –8 9 1 2 –1 Solution: At the MATLAB command prompt.5539 C. Example C. that is.Appendix C Ill−Conditioned Matrices ans = 14. −3 −6 8 1. We recall from Chapter 4 that if the determinant of a square matrix A is singular. 4 −9 −8 −1]. if det ( A ) = 0 . Matrices with condition number close to unity are said to be well−conditioned matrices. Please refer to Chapter 4.2 Condition Number of a Matrix The condition number of a matrix A is defined as k( A) = A ⋅ A –1 (C. cond(A) and MATLAB outputs ans = 2. we enter A=[−2 5 −4 9. and those with very large condition number are said to be ill−conditioned matrices. Third Edition Copyright © Orchard Publications . C− 2 Numerical Analysis Using MATLAB® and Excel®. 7 −5 3 2.1) above.2) where A is the norm of the matrix A defined in relation (C.2 Using the MATLAB function cond(A). The condition number of a matrix A is computed with the MATLAB function cond(A).3724 This condition number is relatively close to unity and thus we classify matrix A as a well-conditioned matrix. the inverse of A is undefined.

a m1 x 1 + a m2 x 2 + … +a mn x n = b m where x 1.. x 2.-5 6 1 1 -..-7 8 1 1 -. 1 ⁄ n . a system with m linear equations and n unknowns can be written as a 11 x 1 + a 12 x 2 + … +a 1n x n = b 1 a 21 x 1 + a 22 x 2 + … +a 2n x n = b 2 . a 12.4) * In general. we classify such a matrix as ill−conditioned. a mn are the coefficients of the system. 1 ⁄ 1... i..e. almost singular.1 4 5 1 1 -. 1 ⁄ 2.-6 7 1 1 -.3) Shown below is an example of the 5 × 5 Hilbert matrix.-8 9 (C. The coefficient matrix is the m × n matrix with the coefficient a ij as the (i.... . . let us consider that the coefficient matrix* is very small.. C.3 Hilbert Matrices Let n be a positive integer.Hilbert Matrices Now. ..j)-th entry: a 11 a 12 … a 1n a 21 a 22 … a 2n … … … … a m1 a m1 a m1 a m1 Numerical Analysis Using MATLAB® and Excel®. Thus. 1 -1 1 -2 1 -3 1 -4 1 -5 1 -2 1 -3 1 -4 1 -5 1 -6 1 -3 1 -4 1 -5 1 -6 1 -7 1 --. … are unit fractions. x n are the unknowns and the numbers a 11. Third Edition Copyright © Orchard Publications C− 3 . .. that is. Accordingly. A Hilbert matrix is a matrix with unit fraction elements B ij = 1 ⁄ ( i + j – 1 ) (C. A unit fraction is the reciprocal of this integer. …. …. 1 ⁄ 3.

we are asked to find the values of x 1 and x 2 of the linear system 0. Example C.187 0.379 ⋅ x 1 = 0. Solution: Here.464 0.3 Compute the determinant and the condition number of the 6 × 6 Hilbert matrix using MATLAB. Solution: At the MATLAB command prompt.464 0. we enter det(hilb(6)) and MATLAB outputs ans = 5.4 Let Ax = b where A = 0. Third Edition Copyright © Orchard Publications . for the 6 × 6 Hilbert matrix.4951e+007 This is a large number and if the coefficient matrix is multiplied by this number. We can find the condition number of a matrix A with the cond(A) MATLAB function. cond(hilb(6)) ans = 1.Appendix C Ill−Conditioned Matrices MATLAB’s function hilb(n) displays the Hilbert n × n matrix. Let us consider another example.728 0.728 0.585 0.379 and b = 0.256 Compute the values of the vector x . Thus.585 0. Example C.256 C− 4 Numerical Analysis Using MATLAB® and Excel®. seven decimal places might be lost.187 x2 0.3673e-018 This is indeed a very small number and for all practical purposes this matrix is singular.

4 above should serve as a reminder that when we solve systems of equations using matrices.728 0.187 0. b=[0. x=b\A x = 2.585 0.9428 Check: A=[0. Example C. determinant = det(A) determinant = -0.8852]'. 0.9428 1. i.378.378.256]'. we conclude that this system of equations is ill-conditioned and the solution is invalid. x=[2.728 0.464].8852 b = 2.4341 3.464].0037 condition=cond(A) condition = 328.6265 Therefore. so let us check the determinant and the condition number of the matrix A .. we should check the determinants and the condition number to predict possible floating point and roundoff errors. b=A*x 1. and we use the left division operation. 0. Numerical Analysis Using MATLAB® and Excel®.585 0. we define A and b .e.0171 but these are not the given values of the vector b .Hilbert Matrices Using MATLAB. Third Edition Copyright © Orchard Publications C− 5 . A=[0.

ISBN 0-4866127-2-4 7. Natick. Using MATLAB Graphics 4. ISBN-13: 978−1−934404−01−2 2. ISBN 0-8493-0626-4 . ISBN 0-9744239-7-1 5. The following publications by The MathWorks. Third Edition. Handbook of Mathematical Functions. are listed below. Financial Toolbox 5. Using MATLAB 3. ISBN 0−9709511−5−9 4. 01760. Introduction to Simulink with Engineering Applications.mathworks. MA.com. www. 1. 3 Apple Hill Drive. Mathematics for Business. Circuit Analysis I with MATLAB Applications. Statistics Toolbox B. They are available from The MathWorks. 1. ISBN 0−9709511−2−4 3. Signals and Systems with MATLAB Computing and Simulink Modeling. and Technology with MATLAB and Excel Computations. ISBN 0-9744239-9-8 6. Third Edition. Other references indicated in footnotes throughout this text. Getting Started with MATLAB 2. are highly recommended for further study. CRC Standard Mathematical Tables. Circuit Analysis II with MATLAB Applications.References and Suggestions for Further Study A. Science.

Index Symbols % (percent) symbol 1-2 %d MATLAB symbol 2-5 %e MATLAB symbol 2-21 %s MATLAB symbol 2-21 %u MATLAB symbol 2-21 A abs(z) MATLAB function 1-24 absolute cell in Excel 2-19 absolutely convergent 15-10 Adams’ method 9-13 Add Trendline Excel feature 8-6 adjoint of a matrix . 2-15 autoscale icon in Simulink B-12 axis MATLAB command 1-17.see matrix diagonal of a matrix . 33 EXP(GAMMALN(n)) Excel function 13-5 expand(s) MATLAB function 7-13.see matrix algebraic constrain block in Simulink B-18 algebraic form of the Legendre polynomials 15-17 alternate form of the trigonometric Fourier series 6-25 alternating current 3-1 angle(z) MATLAB function 1-24 angular velocity 3-2 antidifference 7-12 approximations with Excel 2-7 associated Legendre differential equation 15-18 associated Legendre functions of the first and second kind 15-18 AutoFill Excel feature 2-9.see matrix collect(s) MATLAB function 7-15. 12-11 column vector 1-19 command screen in MATLAB 1-1 command window in MATLAB 1-1 commas in MATLAB 1-7 comment line in MATLAB 1-2 Commonly Used Blocks in Simulink B-7 complementary solution A-1 complex conjugate 1-4 complex numbers 1-2. 14-10 Bessel function of negative order n 15-2 Bessel function of n 15-1 Bessel functions 15-1 Bessel functions of the first kind 15-7 Bessel functions of the second kind 15-7 besselj(n. 7-8 default color in MATLAB 1-14 default line in MATLAB 1-15 default marker in MATLAB 1-14 default values in MATLAB 1-11 degree of differential equation 5-3 demo in MATLAB 1-2 determinant .see matrix diff(s) MATLAB function 2-6 difference equations A-1 difference operator 7-4.k) MATLAB function 14-24 diagonal elements of a matrix .n) Excel function 15-3 beta distribution 13-20 beta function 13-17 beta(m. 12-9 exponential form of complex numbers 3-14 exponential form of the Fourier series 6-29 eye(n) MATLAB function 4-6.see finite differences dipole 15-12 direct terms in MATLAB 12-1 directed network 16-15 disp(x) MATLAB function 2-5 Display block in Simulink B-18 display formats in MATLAB 1-31 distinct poles 12-2 distinct roots of characteristic equation 5-9 divided differences 7-1 division of complex numbers 3-13 dot division operator in MATLAB 1-21 dot exponentation operator in MATLAB 1-21 dot multiplication operator in MATLAB 1-21 dsolve(s) MATLAB function 9-5 dynamic programming 16-4 E Editor window in MATLAB 1-1.see matrix diag(v. 11-1 differences . 6-31 even symmetry 6-7 exit MATLAB command 2.b) MATLAB function 1-6 Cramer’s rule 4-18. 1-21 elements of a matrix .q) MATLAB function 1-6. 3-11 complex poles 12-5 complex roots of characteristic equation 5-9 cond(A) MATLAB function C-2 condition number of a matrix .n) MATLAB function 13-19 BETADIST Excel function 13-21 bisection method for root approximation 2-19 box MATLAB command 1-12 C Casorati’s determinant 11-2 Cayley-Hamilton theorem 5-30 characteristic (auxiliary) equation 5-8 characteristic equation of a second order difference equation 11-3 chart type in Excel 2-10 Chart Wizard Excel feature 2-10.see matrix eps in MATLAB 1-22. 8-4 cubic interpolarion in MATLAB 7-25 cubic spline interpolarion in MATLAB 7-25 curve fitting 8-1 curved regression 8-7. 8-6 Chebyshev polynomials 15-22 Chebyshev polynomials of the first kind 15-22 Chebyshev polynomials of the second kind 15-22 Cholesky factorization 14-23 classification of differential equations 5-2 clc MATLAB command 1-2 clear MATLAB command 1-2 coefficient matrix .see matrix cofactors . 1-27 equating the numerators procedure in partial fraction expansion 12-13 Erlang distribution 13-16 error function 10-4 error in method of least squares 8-3 Euclidean norm C-1 Euler’s identities 3-14 even functions 6-7. 1-27 Editor/Debugger in MATLAB 1-1. 8-14 cycle 3-2 cyclic frequency 3-3 D Data Analysis Toolpack in Excel 8-7 data points in MATLAB 1-14 decibel 12 decimal format 2-21 deconv(p. 1-22 B backward substitution. 14-24 eye(size(A)) MATLAB function 4-7 F factor(p) MATLAB function 12-3 factorial polynomials 7-6 Fibonacci numbers 11-7 figure window in MATLAB 1-14 finite differences 7-1 first divided difference 7-1 first harmonic 6-1 fixed point format 2-21 Flip Block command in Simulink B-11 forced response 5-8 forcing function A-1 format MATLAB command 1-31 format specifiers in MATLAB 2-21 IN1 .x) MATLAB function 15-3 BESSELJ(x.see matrix configuration parameters in Simulink B-12 congugate of a matrix .see matrix conj(A) MATLAB function 4-8 conj(x) MATLAB function 4-8 conjugate of a complex number 3-12 connected network 16-15 constraints in linear programming 16-1 Contents Pane in Simulink B-7 conv(a. 1-2 eigenvalues 5-30 eigenvectors 5-39 element-by-element division in MATLAB 1-21 element-by-element exponentiation in MATLAB 1-21 element-by-element multiplication in MATLAB 1-19.

see matrix lsqnonneg MATLAB function 14-28 LU factorization method 14-9 lu(A) MATLAB function 14-12. 13-10 GAMMADIST Excel function 13-16 GAMMALN Excel function 13-5 Gaussian elimination method 4-20.yi) MATLAB function 7-25 interp2(x.x0) MATLAB function 12-18 fzero(f. 15-25 H half-wave rectifier 6-19.see matrix Math Operations Library in Simulink B-11 MATLAB Demos 1-2 matrix (matrices) 4-1 adjoint of 4-22 coefficient C-3 cofactors of 4-11 condition number of C-2 conformable for addition 4-2 conformable for multiplication 4-4 conformable for subtraction 4-2 congugate of 4-8 determinant of 4-9 diagonal elements of 4-1 diagonal of 4-6 elements of 4-1 Hermitian 4-9 Hilbert C-3 identity 4-6 ill-conditioned 4-22.x) MATLAB function 15-19 Legendre’s equation 15-10 lims = MATLAB function 1-27 line graph 16-4 line spectrum 6-33 linear factor 1-9 linear interpolarion 7-25 linear programming 16-1 linear regression 8-2 linspace MATLAB command 1-14 ln (natural logarithm) 1-12 log (common logarithm) 1-12 log(x) MATLAB function 1-12 log10(x) MATLAB function 1-12 log2(x) MATLAB function 1-12 loglog MATLAB command 1-12. 1-15 Hermitian matrix .xi.’method’) MATLAB function 7-25 interpolation 7-1 interpolation with MATLAB 7-24 interval halving 2-19 inverse of a matrix . 11-2. A-1 nearest neighbor interpolarion 7-25 network analysis 16-14 Neumann functions 15-7 .see matrix homogeneous difference equation 11-2 Hz (Hertz) 3-3 I identity matrix .x) MATLAB function 2-7 G Gain block in Simulink B-18 gamma distribution 13-16 gamma function 13-1 gamma(n) MATLAB function 13-3.xi.see matrix isoprofit line 16-2 J j operator 3-11 L L’ Hôpital’s rule 13-2 lagging sinudoid 3-3 Lagrange’s interpolation method 7-17 Laguerre polynomials 15-21 leading sinusoid 3-3 least squares 8-2.yi.y. C-2 inverse of 4-23 lower triangular 4-6 main diagonal elements of 4-1 main diagonal of 4-1 minor of determinant 4-13 non-singular 4-22 scalar 4-6 singular 4-22 size of 4-7 skew-Hermitian 4-9 skew-symmetric 4-8 square 4-1 symmetric 4-8 trace of 4-2 transpose of 4-7 unitary 14-25 upper triangular 4-5 well-conditioned C-2 zero matrix .see matrix hilb(n) MATLAB function C-4 Hilbert matrix .y) MATLAB command 1-17 method of Frobenius 15-1 method of least squares 8-2. 8-7 Legendre functions 15-10 Legendre functions of the second kind 15-11 Legendre polynomials 15-11 Legendre polynomials in trigonometric form 15-18 legendre(n. 6-31 help in MATLAB 1-2.z. 14-9 Genenbauer 15-22 general solution of a differential equation 5-6 generalized factorial function 13-1 generating function for Bessel functions of the first kind of integer order 15-9 generating function for Legendre polynomials 15-12 Goal Seek Excel feature 2-16 Gram-Schmidt orthogonalization procedure 5-39.see matrix main diagonal of a matrix .y.xi) MATLAB function 7-24 interp1(x.see matrix ill-conditioned matrix . 14-14 M Maclaurin power series 6-41. 7-6 main diagonal elements . 2-4 fundamental frequency 6-1 fundamental theorem of integral calculus 7-12 fundamental theorem of sum calculus 7-13 fzero MATLAB function 1-27.see matrix MINVERSE Excel function 4-28 MMULT Excel function 4-28.see matrix matrix left division in MATLAB 4-26 matrix multiplication in MATLAB 1-19. 34 lower triangular matrix .z) MATLAB command 1-17 mesh(Z) MATLAB function 7-32 meshgrid(x.y. 1-27. 4-29 modified Bessel functions 15-7 modulation index 15-4 multiple poles 12-6 multiplication of complex numbers 3-12 N NaN in MATLAB 1-27 natural response 5-8. 8-3 least squares line 8-2 least-squares curve 8-2 least-squares parabola 8-2. 1-27 Milne’s method 9-15 minimum span problems 16-15 minor of determinant . 1-20 mesh(x.y.a.xi. 1-29 fzero(‘function’.forward substitution 14-10 Fourier analysis 3-2 Fourier series 6-1 fplot MATLAB command 1-27 fprintf MATLAB command 7-29 frequency 3-2 Frequency Modulation (FM) 15-4 frequency response 1-11 full-wave rectifier with even symmetry 6-24 function block parameters in Simulink B-10 function files in MATLAB 1-26. 14-7 Gregory-Newton Backward Interpolation Formula 7-21 Gregory-Newton Forward Interpolation Formula 7-19 grid MATLAB command 1-11 gtext MATLAB command 1-14.b) MATLAB function 10-5 interp1(x. 6-22 half-wave symmetry 6-7.y.’method’) MATLAB function 7-25 interp2(x. 8-3 method of Picard 15-1 method of undetermined coefficients 5-10.see matrix imag(z) MATLAB function 1-24 imaginary axis 3-10 imaginary number 3-10 IN2 improper integral 13-1 improper rational function 12-1 increments between points in MATLAB 1-14 in-phase sinusoids 3-3 input(‘string’) MATLAB command 2-2 int(f.z. A-1 method of variation of parameters 5-20 m-file in MATLAB 1-1 1-26.

6-19 trigonometric Fourier series 6-1 trigonometric relations 3-5 two-dimensional plots 7-32 type of a diferential equation 5-2 U ultraspherical functions 15-22 undetermined system 8-3 unit fraction C-3 unitary matrix . 8-11 power series 6-40 proper rational function 12-1 Q QR factorization 14-25 qr(A) MATLAB function 14-25 quad MATLAB function 10-10 quad8 MATLAB function 10-10 quadratic curve 8-1 quadratic factor 1-9 quit MATLAB command 1-2 R radian frequency 3-2 rational polynomial 1-8 rationalization of the quotient 3-13 real axis 3-10 real number 3-11 real(z) MATLAB function 1-24 recursion A-1 recursive method A-1 regression 8-1 regression analysis 8-7 relative cell in Excel 2-19 repeated poles 5-9.Newton’s divided difference interpolation method 7-15 Newton-Cotes 8-panel rule 10-10 non-homogeneous difference equation 11-2 non-homogeneous ODE 5-6 non-singular matrix . 11-2 X xlabel MATLAB command 1-12 XY (Scatter) in Excel 8-6 Y ylabel MATLAB command 1-12 Z zero matrix . 12. 6-49 start simulation in Simulink B-12 state equations 5-24 state transition matrix 5-28 state variables 5-24 State-Space block in Simulink B-12 Stirling’s asymptotic series for the G(n) function 13-9 string in MATLAB 1-17 subplot MATLAB command 1-18 surface zonal harmonics 15-11 svd(A) MATLAB function 14-28 sym.r) MATLAB function 1-24 poles 12-2 poly MATLAB function 1-4 polyder MATLAB function 1-7 polyfit(x. syms MATLAB symbolic expressions 12-4 symbolic expressions in MATLAB 12-3 Symbolic Math Toolbox in MATLAB 12-4 symmetric matrix .see matrix square waveform 6-9.6 residue(r.see matrix symmetry 6-7. 6-41. 1-12. 3-3 periodic waveform 3-2 phasor 3-2 plot area in Excel 6 plot MATLAB command 1-9. 1-19 Runge-Kutta method 5-24.see matrix skew-symmetric matrix .y) MATLAB function 10-3. 6-14. 6-16.p.see matrix V VLOOKUP Excel function 7-23 W Wallis’s formulas 13-16 Weber functions 15-7 well-conditioned matrix .see matrix trapezoidal rule 10-1 trapz(x. 10-5 Tree Pane in Simulink B-7 Trendline Excel feature 8-9 triangular waveform 6-11. 6-48. 1-8 rotating vector 3-5 round(n) MATLAB function 1-24 row vector 1-3. 6-44 Taylor series expansion method 9-1 text MATLAB command 1-14.k) MATLAB function 12-1 revolutions per second 3-5 Rodrigues’ formula 15-12. 6-31 odd symmetry 6-7 ODE (Ordinary Differential Equation) 5-3 ode23 MATLAB function 9-9 ode45 MATLAB function 9-9 one-dimensional wave equation 5-3 optimum path policy 16-5 order of a differential equation 5-3 ordinary differential equation 5-3 oriented network 16-15 orthogonal basis 14-5 orthogonal functions 6-2. 14-1. 14-4 orthonormal basis 14-5 out-of-phase sinusoids 3-3 overdetermined system 8-3 P parabolic curve 8-1 partial differential equation (PDE) 5-3 partial fraction expansion 12-1 PDE (Partial Differential Equation) 5-3 Pearson correlation coefficient 8-10 period 3-2.see matrix Singular Value Decomposition 14-28 Sinks library B-18 sinusoids 3-2 size of a matrix . 1-17 third harmonic 6-1 title(‘string’) in MATLAB 1-12 trace of a matrix . 6-14.see matrix upper triangular matrix .see matrix while end in MATLAB 2-4 Wronskian determinant 5-10.see matrix Scope block in Simulink B-12 script file in MATLAB 1-26 second divided difference 7-1 second harmonic 6-1 semicolons in MATLAB 1-7 semilogx MATLAB command 1-12 semilogy MATLAB command 1-12 simple differential equations 5-1 simplex method 16-4 Simpson’s rule 10-6 Simulation drop menu in Simulink B-12 simulation start icon in Simulink B-12 Simulink icon B-7 Simulink Library Browser B-8 singular matrix .see matrix skew-Hermitian matrix . 6-31 T Taylor series 5-24. 15-18 roots of polynomials 1-3 roots(p) MATLAB function 1-3. 1-15 plot3 MATLAB command 1-16 polar form of complex numbers 3-15 polar plot in MATLAB 1-24 polar(theta.y. 6-18 scalar matrix .see matrix solution of the homogeneous ODE 5-8 solutions of ODEs 5-6 space equations 5-24 spectrum analyzer 33 spherical harmonics 15-18 sprintf MATLAB command 2-5 sqrt MATLAB function 10-12 square matrix . 14-2 orthogonal system 15-9 orthogonal trajectories 14-2 orthogonal unit vectors 14-5 orthogonal vectors 5-39.n) MATLAB function 8-11 polynomial construction from known roots in MATLAB 1-4 polyval(p.x) MATLAB function 1-5.see matrix norm C-1 norm(A) MATLAB function C-1 numeric expressions in MATLAB 12-4 numerical evaluation of Fourier coefficients 6-36 O odd functions 6-8.see matrix transpose of a matrix . 9-5 S sawtooth waveform 6-10.see matrix zeros 12-2 zlabel MATLAB command 1-17 IN3 .