Scribd Logo
Upload

Welcome to Scribd!

  • (R. Kent Nagle, Edward B. Saff, Arthur David Snide

    Uploaded by

    Tarcísio Miguel
    9 views758 pages

    Original Title

    [R. Kent Nagle, Edward B. Saff, Arthur David Snide(B-ok.org)

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    9 views758 pages

    (R. Kent Nagle, Edward B. Saff, Arthur David Snide

    Uploaded by

    Tarcísio Miguel

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    of 758
    FUNDAMENTALS OF Differential Equations NAGLE SAFF SNIDER Fundamentals of Differential Equations FIFTH EDITION R. Kent Nagle Edward B. Saff University of South Florida Arthur David Snider University of South Florida A vy ADDISON-WESLEY An imprint of Addison Wesley Longman, Inc. Reading, Massachusetts * Menlo Park, California * New York * Harlow, England Don Mills, Ontario * Sydney * Mexico City + Madrid * Amsterdam Sponsoring Editor Laurie Rosatone Project Manager Rachel S. Reeve Associate Project Editor Ellen Keohane Senior Designer Barbara T. Atkinson Associate Production Supervisor Kim Ellwood Project Coordination Elm Street Publishing Services, Inc. Compositor Nesbitt Graphics, Inc. Art Academy Artworks. Senior Technology Specialist Joe Vetere Prepress Buyer Caroline Fell Print Buyer Evelyn Beaton Marketing Manager Carter Fenton Associate Marketing Manager Michael Boezi Cover Iilustration George V. Kelvin, from a computer graphic by Alar Toomre, MIT Library of Congress Cataloging-in-Publication Data Nagle, R. Kent Fundamentals of differential equations. — Sth ed. /R. Kent Nagle, Edward B. Saff, Arthur David Snider. Pp. cm. Includes index. ISBN 0-201-33868-8 1. Differential equations, I, Saff, E, B., 1944~ U. Snider, Arthur David, 1940- II, Title. QA37IN24_ 2000 515135—de21 99-1732 cr Reprinted with corrections, Copyright © 2000 by Addison Wesley Longman. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the publisher. Printed in the United States of America. 6789 10-QUF-0302 Dedicated to R. Kent Nagle He has left his imprint not only on these pages but upon all who knew him. He was that rare mathematician who could effectively communicate at all levels, imparting his love for the subject with the same ease to undergraduates, graduates, precollege stu- dents, public school teachers, and his colleagues at the University of South Florida. Kent was at peace in life—a peace that emanated from the depth of his under- standing of the human condition and the strength of his beliefs in the institutions of family, religion, and education. He was a research mathematician, an accomplished author, a Sunday school teacher, and a devoted husband and father. Kent was also my dear friend and my jogging partner who has left me behind still struggling to keep pace with his high ideals E. B. Saff New Sections on Qualita- tive Features New Section ‘on Vibrations Preface OUR GOAL Fundamentals of Differential Equations is designed to serve the needs of a one-semester course in basic theory as well as applications of differential equations. We have striven to cre- ate a flexible text that affords the instructor substantial latitude in designing a syllabus (sample syllabi are provided in this preface), in course emphasis (theory, applications and techniques, numerical methods), and in using commercially available computer software and/or the soft- ware utility programs specifically designed for this text. INNOVATIONS FOR THIS EDITION In response to requests of users and reviewers, and recognizing recent developments in teach- ing and learning, we offer the following New sections describe how qualitative information can be gleaned about the solutions to intractible autonomous equations by observing their direction fields and critical points on the phase line (for single equations) or the phase plane (for two-dimensional systems). With the assistance of suitable software, this approach provides a refreshing, almost recreational alter- native to the traditional analytic methodology as we discuss applications in nonlinear mechan- ies, ecosystems, and epidemiology. Motivation for Chapter 4 on linear differential equations is provided in a new introductory sec- tion describing the mass-spring oscillator, We exploit the reader's familiarity with common vibratory motions to anticipate the exposition of the theoretical and analytical aspects of linear equations. Not only does this model provide an anchor for the discourse on constant coefficient equations, but a liberal interpretation of its features enables us to predict the qualitative behavior of variable-coefficient and nonlinear equations as well. vii viii Preface Many New Figures New Group Projects Expanded Review of Linear Alge- braic Equations and Matrices New Treatment of Taylor Polynomials Revised Exercises Website Support System In keeping with the increased emphasis on qualitative methods, about 40 new figures have been added. Thanks again to the feedback we received from instructors using the earlier editions, five new group projects have been introduced at the end of the chapters. These projects include: Chapter 2 * Two Snowplows Chapter 4 + An Alternative to the Method of Undetermined Coefficients + Apollo Reentry Chapter 5 + Hamiltonian Systems Chapter 7 + Determining System Parameters ‘The chapter on matrix methods for linear systems (Chapter 9) has been expanded with two (optional) introductory sections reviewing the theory of linear algebraic systems and matrix algebra. In Chapter 8 the reader is gracefully initiated into the theory of power series solutions with an exposition of Taylor polynomial approximants to solutions, deferring the sophisticated issues of convergence to later sections. For instructors’ convenience, some of the basic “skills-building” exercises have been modified (to render previous answers obsolete), and a selection of more challenging exercises has been added. As a useful companion to the text, students and instructors are provided access to a specially designed Website that not only includes helpful links keyed to the sections of the text but also provides a utility for solving many of the numerical problems. This utility, which requires only a standard browser with Java and JavaScript activated, is student-friendly, requires little learn- ing time, and will entice students to conduct numerical experiments. It includes facility for direction field plots, Euler’s methods, Runge-Kutta methods (single equations and systems), phase plane diagrams, eigenvalues and eigenvectors, and standard computational algorithms such as Newton’s method for solving equations, Simpson’s rule for numerical integration, and plot/table construction for one or more functions. Also, the user will find recommendations for syllabi, an index of applications, study tips, and suggestions for incorporating Interactive Dif- ferential Equations modules (described in the “Supplements” section below). Point your Web browser to hup:/Avww.awlonline.com/nagle to access the Website. Preface ix PREREQUISITES While some universities make linear algebra a prerequisite for differential equations, many schools (especially engineering) only require calculus. With this in mind, we have designed the text so that only Chapter 6 (Theory of Higher-Order Linear Differential Equations) and Chap- ter 9 (Matrix Methods for Linear Systems) require more than high school linear algebra. More- over, Chapter 9 contains review sections on matrices and vectors as well as specific references for the deeper results used from the theory of linear algebra. We have also written Chapter 5 so as to give an introduction to systems of differential equations—including methods of solving, phase plane analysis, applications, numerical procedures, and Poincaré maps—that does not require a background in linear algebra. SAMPLE SYLLABI As a rough guide in designing a syllabus related to this text, we provide three sample syllabi that can be used for a 15-week course that meets three hours per week: the first emphasizes methods, computations, and applications; the second, theory and methods; and the third, meth- ods and partial differential equations. Chapters 1, 2, and 4 provide the core for any course. The rest of the chapters are, for the most part, independent of each other. ‘Theory and Methods, Methods Methods and ‘Computations, (linear algebra Partial Differential and Applications prerequisite) Equations Week Sections Sections Sections 1 1.1, 1.2,13 11,12, 13 11,1.2,13 2 1.4, 15,22 2.2,2.3 15,22 3 23,24 2.4, 4.2 23,24 4 3.2, 3.4, 3.5 43,44 32,34 5 3.6,4.1,4.2 45,4.6,4.7 42,43 6 3, 4.5 48,49 4.5,4.6,4.7 7 4.6.4.7, 4.8 5.2, 5.3, 6.1 48,49 8 4.9,4.10, 4.11 6.2, 6.3 5.2, 5.3 9 4.12, 5.1, 5.2 12,73 72,73 10 5.3,5.4, 5.6 14,15 14,15 u 7.6,8.1, 8.2 7.6, 8.1, 8.2 12 83,85 8.3, 8.5, 8.6 13 8.6, 9.1, 9.2, 9.3 10.2, 10.3, 14 8.1, 8.2, 8.3 9.4,9.5 10.4, 10.5 15 8.5, 8.6 9.6,9.7 10.6, 10.7 x Preface Flexible Organization Optional Use of Computer Software Choice of Applications Group Projects Technical Writing Exercises Historical Footnotes Motivating Problem Chapter Summary and Review Problems Computer Graphics RETAINED FEATURES Most of the material is modular in nature to allow for various course configurations and emphasis (theory, applications and techniques, and concepts). ‘The availability of computer packages such as MATHCAD®, MATHEMATICA®, MATLAB®, and MAPLE® provides an opportunity for the student to conduct numerical experiments and tackle realistic applications that give additional insights into the subject. Consequently, we have inserted several exercises and projects throughout the text that are designed for the stu- dent to employ available software in phase plane analysis, eigenvalue computations, and the numerical solutions of various equations. Because of syllabus constraints, some courses will have little or no time for sections (such as those in Chapters 3 and 5) that exclusively deal with applications. Therefore, we have made the sections in these chapters almost completely independent of each other. To afford the instructor even greater flexibility, we have built in a variety of applications in the exercises for the theo- retical sections. In addition, we have included many projects that deal with such applications. At the end of each chapter are group projects relating to the material covered in the chapter. A project might involve a more challenging application, delve deeper into the theory, or introduce more advanced topics in differential equations. Although these projects can be tackled by an individual student, classroom testing has shown that working in groups lends a valuable added dimension to the learning experience. Indeed, it simulates the interactions that take place in the professional arena. ‘Communication skills are, of course, an essential aspect of professional activities. Yet few texts provide opportunities for the reader to develop these skills. Thus we have added at the end of most chapters a set of clearly marked technical writing exercises that invite students to make documented responses to questions dealing with the concepts in the chapter. In so doing, stu- dents are encouraged to make comparisons between various methods and to present examples that support their analysis. ‘Throughout the text historical footnotes are set off by colored daggers(‘ ). These footnotes typ- ically provide the name of the person who developed the technique, the date, and the context of the original research. Most chapters begin with a discussion of a problem from physics or engineering that motivates the topic presented and illustrates the methodology. All of the main chapters contain a set of review problems along with a synopsis of the major concepts presented. Most of the figures in the text were generated via computer. Computer graphics not only ensure greater accuracy in the illustrations, they demonstrate the use of numerical experimentation in studying the behavior of solutions. Proofs Linear Theory Num Algorithms Exercises Optional Sections Laplace Transforms Power Series Preface xi While more pragmatic students may balk at proofs, most instructors regard these justifications as an essential ingredient in a textbook on differential equations. As with any text at this level, certain details in the proofs must be omitted. When this occurs, we flag the instance and refer readers either to a problem in the exercises or to another text. For convenience, the end of a proof is marked by the symbol mend. We have developed the theory of linear differential equations in a gradual manner. In Chapter 4 (Linear Second Order Equations) we present the basic theory for linear second order equations and discuss various techniques for solving these equations. Higher-order equations are briefly mentioned in this chapter. A more detailed discussion of linear higher-order differential equa- tions is given in Chapter 6 (Theory of Higher-Order Linear Differential Equations). For a beginning course emphasizing methods of solution, the presentation in Chapter 4 is sufficient and Chapter 6 can be skipped. Several numerical methods for approximating solutions to differential equations are presented along with program outlines that are easily implemented on a computer. These methods are introduced early in the text so that teachers and/or students can use them for numerical experi- mentation and for tackling complicated applications. Most algorithms discussed are imple- mented on the Website for this text. An abundance of exercises is graduated in difficulty from straightforward, routine problems to more challenging ones. Deeper theoretical questions, along with applications, usually occur toward the end of the exercise sets. Throughout the text we have included problems and proj- ects that require the use of a calculator or computer. These exercises are denoted by the symbol . The Website software specifically designed for use with this text greatly facilitates the solutions to these numerical problems. These sections can be omitted without affecting the logical development of the material. They are marked with an asterisk in the table of contents. As mentioned earlier, the sections in Chap- ters 3 and 5 are almost completely independent of each other. We provide a detailed chapter on Laplace transforms (Chapter 7) since this is a recurring topic for engineers. Our treatment emphasizes discontinuous forcing terms and includes a section on the Dirac delta function. Power series solutions is a topic that occasionally causes student anxiety. Possibly, this is due to inadequate preparation in calculus where the more subtle subject of convergent series is (fre- quently) covered at a rapid pace. Our solution has been to provide a thorough treatment of power series solutions that also includes a review of their properties as well as a discussion of real analytic functions. Unlike many texts, ours provides an extensive section on the method of Frobenius (Section 8.6) as well as a section on finding a second linearly independent solution. While we have given considerable space to power series solutions, we have also taken great care to accommodate the instructor who only wishes to give a basic introduction to the topic. An introduction to solving differential equations using power series and the method of Frobenius can be accomplished by covering the materials in Sections 8.1, 8.2, 8.3, and 8.6. xii Preface Partial Differential Equations Student's ‘Solutions Manual Instructor's Resource Guide Interactive Differential Equations (CD-ROM Companion Website MAPLE-Based Instructor's Guide An introduction to this subject is provided in Chapter 10, which covers the method of separa- tion of variables, Fourier series, the heat equation, the wave equation, and Laplace’s equation, Examples in two and three dimensions are included. SUPPLEMENTS By V. Maymeskul and E. B. Saff. Containing complete, worked-out solutions to selected exer- cises, this provides students with an excellent study tool. ISBN 0-201-33869-6 nal group projects. ISBN 0-201-66975-7 Contains short answers to all exercises and addi By Beverly West (Cornell University), Steven Swogatz (Cornell University), Jean Marie MeDill (California Polytechnic State University-San Luis Obispo), John Cantwell (St. Louis University), and Hubert Hohn (Massachusetts College of Art). A revised version of a popular software directly tied to the text. Focuses on helping students visualize concepts. Applications are drawn from engineering, physics, chemistry, and biology. Runs on Windows or Macintosh and is bundled free with every book. Provides additional resources for both instructors and students, including instructional utility software, helpful links keyed to sections of the text, suggestions for incorporating Interactive Differential Equations modules, suggested syllabi, index of applications, and study tips for stu- dents. Access: hutp:/www.awlonline.com/nagle By Kenneth Pothoven (University of South Florida). A collection of MAPLE® worksheets and projects to aid instructors in integrating MAPLE into their courses. Also available via the com- panion Website. ISBN 0-201-61840-0 ACKNOWLEDGMENTS The staging of this text involved considerable behind-the-scenes activity. A special note of gratitude is extended to Alar Toomre (Massachusetts Institute of Technology), who not only provided us with two new Group Projects along with a variety of challenging new problems for the exercises, but also generated the computer graphic appearing on the front cover. We want to thank Frank Glaser (California State Polytechnic University, Pomona) for many of the historical footnotes. We are also indebted to Herbert E. Rauch (Lockheed Research Lab- oratory) for help with Section 3.3 on heating and cooling of buildings, Project B in Chapter 3 on aquaculture, and other application problems. Our appreciation goes to Richard H. Elderkin (Pomona College), Jerrold Marsden (University of California, Berkeley), T. G. Proctor (Clemson University), and Philip W. Schaefer (University of Tennessee), who read and reread the manuscript for the original text, making numerous suggestions that greatly Preface xiii improved the book. We are also indebted to the many people who reviewed the manuscript for this new edition. REVIEWERS Tuncay Aktosun, North Dakota State University Dean R. Brown, Youngstown State University ‘Yunmei Chen, University of Florida Elias Deeba, University of Houston-Downtown, ‘Tyrone Duncan, University of Kansas Kanat Durgun, The Citadel Alejandro Engel, Rochester Institute of Technology David J. Kaup, Clarkson University Subrata Mukherjee, Cornell University Chuanxi Qian, Mississippi State University Philip W. Schaefer, University of Tennessee-Knoxville L. F Shampine, Southern Methodist University David J. Uherka, University of North Dakota In addition, we would like to offer our sincere thanks to all of the instructors who shared their own group projects with us for inclusion in this edition: Richard Bernatz, Luther College Randall K. Campbell-Wright, University of Tampa Alfred Clark, Jr., University of Rochester Leopoldo P. Franea, University of Colorado-Denver T.L. Pearson, Acadia University David Stapleton, University of Central Oklahoma Alar Toomre, Massachusetts Institute of Technology The newly developed companion Website for the text was specially designed by Link- Systems International, Inc. Its unique browser-driven utility package, which was developed by Emil Moskona, not only enhances the usefulness of the text but also provides a convenient tool for calculus/differential equation problem solving in general xiv Preface Finally, we want to thank the staff at Addison-Wesley for their dedicated assistance in bringing this edition to fruition. Special kudos go to mathematics editor Laurie Rosatone and her assis- tants Rachel Reeve and Ellen Keohane for their help, encouragement, and sensitivity, as well to Paul Lorezak, who carefully proofread the text and checked the formulas and exercises. The production phase was carefully guided by Kim Ellwood and Phyllis Crittenden, who admirably put up with our frequent outbursts of anxiety. E. B. Sarr, A. D. SNIDER TAMPA, FLORIDA, Contents CHAPTER I TRODUCTION I 1.1 Background 1 1.2. Solutions and Initial Value Problems 6 1.3 Direction Fields 16 *1.4 The Phase Line 4 *1.5 ‘The Approximation Method of Euler 31 Chapter Summary 37 Technical Writing Exercises 37 Group Projects for Chapter | 38 A. Taylor Series Method 38 B. Picard’s Method 39 C. Magnetic Dipole 40 CHAPTER 2 FIRST ORDER DIFFERENTIAL EQUATIONS 42 2.1. Introduction: Motion of a Falling Body 42 2.2 Separable Equations 45 *Denotes optional sections that can be omitted without compromising the logical flow. xv xvi Contents CHAPTER 3 2.3 Linear Equations 2.4 Exact Equations *2.5 Special Integrating Factors *2.6 Substitutions and Transformations Chapter Summary Review Problems Technical Writing Exercises Group Projects for Chapter 2 A. The Snowplow Problem B. Two Snowplows C. Asymptotic Behavior of Solutions to Linear Equations D. Torricelli's Law of Fluid Flow E, Clairaut Equations and Singular Solutions MATHEMATICAL MODELS AND NUMERICAL METHODS INVOLVING FIRST ORDER EQUATIONS 3.1 3.2 33 3.4 35 3.6 37 Mathematical Modeling Compartmental Analysis Heating and Cooling of Buildings Newtonian Mechanics Improved Euler’s Method Higher-Order Numerical Methods: Taylor and Runge-Kutta Professional Codes for Solving Initial Value Problems Group Projects for Chapter 3 A. Delay Differential Equations B. Aquaculture 63 3 11 86 87 88 89 89 89 90 90 92 93 93 95 107 14 124 135 144 148 148 149 CHAPTER 4 Contents C. Curve of Pursuit D. Aircraft Guidance in a Crosswind E, Stability of Numerical Methods F. Period Doubling and Chaos G. Bang-Bang Controls LINEAR SECOND ORDER EQUATIONS 4.1 _ Introduction: The Mass-Spring Oscillator 4.2. Linear Differential Operators 4.3. Fundamental Solutions of Homogeneous Equations *4.4 Reduction of Order 4,5 Homogeneous Linear Equations with Constant Coefficients 4.6 Auxiliary Equations with Complex Roots 4.7 Superposition and Nonhomogeneous Equations 4.8 Method of Undetermined Coefficients 49 *4.10 4.11 4.12 Variation of Parameters Qualitative Considerations for Variable-Coefficient and Nonlinear Equations A Closer Look at Free Mechanical Vibrations A Closer Look at Forced Mechanical Vibrations Chapter Summary Review Problems Technical Writing Exercises Group Projects for Chapter 4 A. Undetermined Coefficients Using Complex Arithmetic B. An Alternative to the Method of Undetermined Coefficients C. Convolution Method xvii 150 151 152 153 155 156 156 161 168 178 183 191 200 204 213 218 230 240 248 250 251 252 252 253 254 xviii Contents CHAPTER 5 D. Linearization of Nonlinear Problems E, Nonlinear Equations Solvable by First Order Techniques F. Apollo Reentry G. Simple Pendulum H. Asymptotic Behavior of Solutions INTRODUCTION TO SYSTEMS AND PliASE PLANE ANALYSIS 5.1 Interconnected Fluid Tanks *5.2 Introduction to the Phase Plane 5.3. Elimination Method for Systems 5.4 Coupled Mass-Spring Systems 5.5 Electrical Circuits 5.6 Numerical Methods for Higher-Order Equations and Systems 5.7 Dynamical Systems, Poincaré Maps, and Chaos Chapter Summary Review Problems Group Projects for Chapter 5 A. Designing a Landing System for Interplanetary Travel B. Things That Bob C. Effects of Hunting on Predator-Prey Systems D. Periodic Solutions to Volterra-Lotka Systems E, Hamiltonian Systems E. Limit Cycles G. Strange Behavior of Competing Species—Part 1 H. Cleaning Up the Great Lakes 255 256 257 258 259 261 261 263 278 286 293 301 315 325 327 328 328 329 330 331 332 334 335 336 CHAPTER 6& CHAPTER 7 THEORY OF HIGHER~ORDER LINEA! Contents DIFFERENTIAL EQUATIONS 6.1 6.2 63 64 Basic Theory of Linear Differential Equations Homogeneous Linear Equations with Constant Coefficients Undetermined Coefficients and the Annihilator Method Method of Variation of Parameters Chapter Summary Review Problems Technical Writing Exercises Group Projects for Chapter 6 A. Justifying the Method of Undetermined Coefficients B. Transverse Vibrations of Beam LAPLACE TRANSFORMS WA 12 13 14 15 16 17 718 *19 Introduction: A Mixing Problem Definition of the Laplace Transform Properties of the Laplace Transform Inverse Laplace Transform Solving Initial Value Problems ‘Transforms of Discontinuous and Periodic Functions Convolution Impulses and the Dirac Delta Function Solving Linear Systems with Laplace Transforms Chapter Summary Review Problems Technical Writing Exercises xix 366 366 367 367 367 369 369 373 382 388 398 420 429 436 439 440 441 Contents Group Projects for Chapter 7 A. Duhamel’s Formulas B. Frequency Response Modeling C. Determining System Parameters CHAPTER 8 SERIES SOLUTIONS OF DIFFERENTIAL EQUATIONS 8.1 Introduction: The Taylor Polynomial Approximation 8.2 Power Series and Analytic Functions 8.3 Power Series Solutions to Linear Differential Equations 8.4 Equations with Analytic Coefficients *8.5 Cauchy-Euler (Equidimensional) Equations Revisited 8.6 Method of Frobenius 8.7 Finding a Second Linearly Independent Solution 8.8. Special Functions Chapter Summary Review Problems ‘Technical Writing Exercises Group Projects for Chapter 8 A. Spherically Symmetric Solutions to Schrédinger’s Equation ‘for the Hydrogen Atom B. Airy’s Equation C. Buckling of a Tower D. Aging Spring and Bessel Functions CHAPTER 9 MATRIX METHODS FOR LINEAR SYSTEMS 9.1 Introduction 9.2 Review 1: Linear Algebraic Equations 443, 443 444 446 447 447 453 462 473 479 483 495 507 520 521 522 523 523 524 524 525 527 527 532 CHAPTER IO Contents 9.3. Review 2: Matrices and Vectors 9.4 Linear Systems in Normal Form 9.5 Homogeneous Linear Systems with Constant Coefficients 9.6 Complex Eigenvalues 9.7 Nonhomogeneous Linear Systems 9.8 ‘The Matrix Exponential Function Chapter Summary Review Problems Technical Writing Exercises Group Projects for Chapter 9 ‘A. Uncoupling Normal Systems B. Matrix Laplace Transform Method C. Undamped Second Order Systems D. Strange Behavior of Competing Species—Part Il PARTIAL DIFFERENTIAL EQUATIONS 10.1 Introduction: A Model for Heat Flow 10.2. Method of Separation of Variables 10.3 Fourier Series 10.4 Fourier Cosine and Sine Series 10.5. The Heat Equation 10.6 ‘The Wave Equation 10.7. Laplace’s Equation Chapter Summary ‘Technical Writing Exercises Group Projects for Chapter 10 xxi 598 599 600, 600 603 613 631 636 649 662 675 677 678 A. Steady-State Temperature Distribution in a Circular Cylinder 678 xxii Contents B. A Laplace Transform Solution of the Wave Equation C. Green's Function D. Numerical Method for Au = fon a Rectangle APPENDICES A. Newton’s Method B. Simpson's Rule C. Cramer's Rule D. Method of Least Squares E. Runge-Kutta Procedure for n Equations ANSWERS TO ODD-NUMBERED PROBLEMS INDEX 679 680 682 Al A-3 AS A6 Ag B-1 HAPTER Introduction TT BackGROUND In the sciences and engineering, mathematical models are developed to aid in the understand- ing of physical phenomena. These models often yield an equation that contains some deriva- tives of an unknown function. Such an equation is called a differential equation, Two exam- ples of models developed in calculus are the free fall of a body and the decay of a radioactive substance, In the case of free fall, an object is released from a certain height above the ground and falls under the force of gravity.' Newton's second law, which states that an object’s mass times its acceleration equals the total force acting on it, can be applied to the falling object. This leads to the equation (see Figure 1.1) &h ar m —mg , where m is the mass of the object, h is the height above the ground, d7h/d?? is its accelera- tion, g is the (constant) gravitational acceleration, and ~mg is the force due to gravity. This is a differential equation containing the second derivative of the unknown height hk as a func- tion of time. Figure 1.1 Apple in free fall ‘We are assuming here that gravity is the onty force acting on the object and that this force is constant, More general models would take into account other forces, such as air resistance, 2 Chapter 1 Introduction Fortunately, the above equation is easy to solve for h. All we have to do is divide by m and integrate twice with respect to . That is, Ph ar , so dh G = e+e, and ‘We shall see that the constants of integration, c, and c, are determined if we know the initial height and the initial velocity of the object. We then have a formula for the height of the object at time 1. In the case of radioactive decay (Figure 1.2), we begin from the premise that the rate of de- cay is proportional to the amount of radioactive substance present. This leads to the equation dA a kA, k>0, where A(>0) is the unknown amount of radioactive substance present at time rand k is the pro- portionality constant. To solve this differential equation, we rewrite it in the form l a“ —kdt and integrate to obtain 1 =|- [ban [ta MA+C,=-kt+Q. Solving for A yields A= Ali) = 24 =e MeO-G = Conk, Figure 1.2 Radioactive decay Section 1.1 Background 3 where C is the combination of integration constants e©~¢!. The value of C, as we shall see later, is determined if the initial amount of radioactive substance is given. We then have a for- mula for the amount of radioactive substance at any future time Even though the above examples were easily solved by methods leaned in calculus, they do give us some insight into the study of differential equations in general. First, notice that the solu- tion of a differential equation is a fiunction, like h(t) or A(t), not merely a number. Second, inte- gration is an important tool in solving differential equations (not surprisingly!). Third, we cannot expect to get a unique solution to a differential equation, since there will be arbitrary “constants of integration.” The second derivative dh/dr’ in the free fall equation gave rise to two constants, c; and cp, and the first derivative in the decay equation gave rise, ultimately, to one constant, C. Whenever a mathematical model involves the rate of change of one variable with respect to another, a differential equation is apt to appear. Unfortunately, in contrast to the examples for free fall and radioactive decay, the differential equation may be very complicated and diffi- cult to analyze. Differential equations arise in a variety of subject areas, including not only the physical sciences but also such diverse fields as economics, medicine, psychology, and operations re- search. We now list a few specific examples. 1, A classic application of differential equations is found in the study of an electric cir- cuit consisting of a resistor, an inductor, and a capacitor driven by an electromotive force (see Figure 1.3). Here an application of Kirchhoff’s laws! leads to the equation dq dg 1 1 LS +R + ou de dt CO =F), where L is the inductance, R is the resistance, C is the capacitance, E(r) is the electro- motive force, q(t) is the charge on the capacitor, and f is the time. RL AN "00° Figure 1.3 Schematic for a series RLC circuit 2. In the study of the gravitational equilibrium of a star, an application of Newton’s law of gravity and of the Stefan-Boltzmann law for gases leads to the equilibrium equation. ; @ 444 (: «) = -4npG, where P is the sum of the gas kinetic pressure and the radiation pressure, r is the dis- “We will discuss the applications of Kirchhoft’s laws in Section 5.5. Chapter 1 Introduction tance from the center of the star, p is the density of matter, and G is the gravitational constant. 3. In psychology, one model of the learning of a task involves the equation dyfat @ mH -»* yP(1 = yh Van Here the variable y represents the learner’s skill level as a function of time f. The constants p and n depend on the individual learner and the nature of the task. 4. In the study of vibrating strings and the propagation of waves, we find the partial differential equation Pu .u a 4 <$-2°3=0,1 ST or ax’ where f represents time, x the location along the string, c the wave speed, and u the displacement of the string, which is a function of time and location. To begin our study of differential equations we need some common terminology. If an equation involves the derivative of one variable with respect to another, then the former is called a dependent variable and the latter an independent variable. Thus in the equation ; 6 sa sm=o, dt dt tis the independent variable and x is the dependent variable. We refer to a and k as coefficients in equation (5). In the equation © ox ay 72 > xxand y are independent variables and w is a dependent variable. A differential equation involving only ordinary derivatives with respect to a single indepen- dent variable is called an ordinary differential equation. A differential equation involving partial derivatives with respect to more than one independent variable is a partial differential equation. Equation (5) is an ordinary differential equation, and equation (6) is a partial differential equation. The order of a differential equation is the order of the highest-order derivatives present in the equation, Equation (5) is a second order equation because dx/dt? is the highest-order derivative present. Equation (6) is a first order equation because only first order partial deriva- tives occur. It will be useful to classify ordinary differential equations as being either linear or nonlin- ear. Remember that lines (in two dimensions) and planes (in three dimensions) are especially easy to visualize, when compared to nonlinear objects such as cubic curves or quadric surfaces. For example, all the points on a line can be found if we know just two of them, Correspond- ingly, linear differential equations are more amenable to solution than nonlinear ones. Now the ‘Historical Footnote: This partial differential equation was first discovered by Jean le Rond d’ Alembert (1717-1783) in 1747, Section 1.1 Background 5 equations for lines ax + by = c and planes ax + by + cz = dhave the feature that the variables appear in additive combinations of their first powers only. By analogy, a linear differential equation is one in which the dependent variable y and its derivatives appear in additive combi- nations of their first powers. More precisely, a differential equation is lineal has the format a a + yl) 2 + ag(xly = Fle) , dx where a,(x), dy (x), .. . ,@o(x) and F(x) depend only on the independent variable x. The addi- tive combinations are permitted to have multipliers (coefficients) that depend on x; no restric- tions are made on the nature of this x-dependence. If an ordinary differential equation is not linear, then we call it nonlinear. For example, is a nonlinear second order ordinary differential equation because of the y’ term, whereas dy we is linear (despite the x° term). The equation is nonlinear because of the y dy/dx term. Although the majority of equations one is likely to encounter in practice fall into the nontin- ear category, knowing how to deal with the simpler linear equations is an important first step (just as tangent lines help our understanding of complicated curves by providing local approximations). In Problems I through 12, a differential equation is given along with the field or problem area in which it arises. Classify each as an ordinary differential equation (ODE) or a partial differential equation (PDE), give the order, and indicate the independent and dependent variables. If the equation is an ordinary differential equation, indi- cate whether the equation is linear or nonlinear. 2 1.32% 44% + or = 2.00837 dv’ dt (mechanical vibrations, electrical circuits, seismology). 2, S xe +2y=0 (Hermite’s equation, quantum mechanical harmonic oscillator). dy _ y(2- 3x) dx x(1 — 3y) n between two species, ecology), (Laplace’s equation, potential theory, electricity, heat, aerodynamics) 6 Chapter 1 Introduction dp Sop kp(P — p), where k and P are constants (logistic curve, epidemiology, economics). dx _ - - 6 (4 = x)(1 = x) (chemical reaction rates). nylrs (2) here C 9) + (2) | = C, where Cis a constant tochrone problem," calculus of variations), —d? dy 8 Vi-y +2 =0 de ax (Kidder’s equation, flow of gases through a porous medium). dy dy 9. x2 +24 xy=0 “ae ax (aerodynamics, stress analysis). dty = x1 =x) (deflection of beams). ° 11, WBN 19 say, sere kia constant ot or’ r or (nuclear fission). 10. 8: (van der Pol’s equation, triode vacuum tube). In Problems 13 through 16, write a differential equation that fits the physical description. 13. The rate of change of the population p of bacteria at time tis proportional to the population at time r. 14. The velocity at time ¢ of a particle moving along a straight line is proportional to the fourth power of its, position x. 15. The rate of change in the temperature T of coffee at time 1 is proportional to the difference between the temperature M of the air at time ¢ and the tempera- ture of the coffee at time 16, The rate of change of the mass A of salt at time 1 is proportional to the square of the mass of salt present at time t 17. Drag Race. Two drivers, Alison and Kevin, are par- ticipating in a drag race. Beginning from a standing start, they each proceed with a constant acceleration Kevin covers the last 1/4 of the distance in 3 seconds, ‘whereas Alison covers the last 1/3 of the distance in 4 seconds, Who wins and by how much time? TZ" SOLUTIONS AND INITIAL VALUE PROBLEMS An nth order ordinary differential equation is an equality involving the independent variable x, the dependent variable y, and the first n derivatives of y. Examples are 3 (second order) 0 (second order) (fourth order). “Historical Footnote: In 1630 Galileo formulated the brachistochrone problem (Bp xieros = shortest, xpovos = time), that is, 0 determine a path down which a particle will fall from one given point (o another in the shortest time. It ‘was reproposed by John Bemoulli in 1696 and solved by him the following year. EXAMPLE 1 SOLUTION EXAMPLE 2 SOLUTION Section 1.2 Solutions and Initial Value Problems 7 Thus a general form for an nth order equation would be dy aty dy ron% 8) =0 where F is a function of the independent variable x, the dependent variable y, and the deriva- tives of y up to order n; that is, x, y,..., d"y/dx". We assume that the equation holds for all x in an interval J (which may or may not include its endpoints: a = x = b,a are defined for all x # 0. Substitution of (x) for y in equation (3) gives 2 2 ) Bt — a4 (2 = 273) - (2 - 2x73) -— (2-204) =0 x Since this is valid for any.x # 0, the function (x) = is an explicit solution to (3) on (-co, 0) and also on (0, 00). mend Show that for any choice of the constants ¢, and c>, the function (x) = cye™* + cye* is an explicit solution to (4) We compute $"(x) = ~cje™* + 2cye** and f"(x) = ce + 4epe**. Substitution of 6,4", and 6" for y. y", and y” in equation (4) yields (cye* + 4eqe*) — (—ce~* + 2cye*) — Acye~* + ce”) = (c, + 1 — 2eye® + (dey — 2ey — 2ephe* = 0 Since equality holds for all x in (~oo, 00), then (x) = cje~* + eye” is an explicit solution to (4) on the interval (—o0, 0) for any choice of the constants c; and c>. mend Chapter 1 EXAMPLE 3 SOLUTION EXAMPLE 4 SOLUTION Introduction As we shall see in Chapter 2, the methods for solving differential equations do not always yield an explicit solution for the equation. We may have to settle for a solution that is defined implicitly. Consider the following example. Show that the relation (5) y-xs+8=0 implicitly defines a solution to dy 3x7 (6) de> dy on the interval (2, 00). When we solve (5) for y, we obtain y = + Vx? — 8. Let's ry d(x) = V8 — 8 to see if it is it solution, Since d/dx = 3x2/(2VF = 8) , both ¢ and déb/dx are defined on (2, 00) an explic Substituting them into (6) yields -—3e 2(v = 8) which is valid for all x in (2, 00). Hence relation (5) is an implicit solution to (6) on (2, 00). (The reader can check that w(x) = — Vx? — 8 is also an explicit solution to (6).) mend ta cu NULL Definition 2. A relation G(x, y) = 0 is said to be an implicit solution to equation (1) on the interval / if it defines one or more explicit solutions on /. Show that M xtyte%=0 is an implicit solution to dy 2) eX? (8) (1 + xe?)F + 1 + ye First, we observe that we are unable to solve (7) directly for y in terms of x alone. However, for (7) to hold, we realize that any change in.x requires a change in y, so we expect the relation (7) to define implicitly at least one function y(x). This is difficult to show directly but can be rigor- ously verified using the implicit function theorem’ of advanced calculus, which guarantees that such a function y(x) exists that is also differentiable (see Problem 30). 'See Vector Calculus, Fourth Edition, by J. E. Marsden and A. J. Tromba (W. H. Freeman and Company, San Francisco, 1996) EXAMPLE 5 SOLUTION Section 1.2 Solutions and Initial Value Problems 9 Once we know that y is a differentiable function of x, we can use the technique of implicit differentiation. Indeed, from (7) we obtain on differentiating with respect to x, 1424 ey 45%) = a Xa) © or (1+ xe) +1 tye =0, 7 which is identical to the differential equation (8). Thus relation (7) is an implicit solution on some interval guaranteed by the implicit function theorem, ménd Verify that 4x? — y? implicit solutions to d o yo C, where C is an arbitrary constant, gives a one-parameter family of 4x=0 and graph several of these solution curves. When we implicitly differentiate the equation 4x? — y? = C with respect to.x, we find which is equivalent to (9). In Figure 1.4 we have sketched the implicit solutions for C = 0, +1, +4, The curves are hyperbolas with common asymptotes y = +2x. Notice that the implicit solution curves (with C arbitrary) fill the entire plane and are nonintersecting for C # 0. For C = 0, the implicit solution gives rise to the two explicit solutions y = 2x and y = —2x, both of, which pass through the origin, meng 10 Chapter 1 Introduction For brevity, we hereafter use the term solution to mean either an explicit or an implicit solution, In the beginning of Section 1.1, we saw that the solu tion invoked two arbitrary constants of integration c), n of the second order free fall equa- et A) =Stcrter, while the solution of the first order radioactive decay equation contained a single constant C: A(t) = Ce“ It is clear that integration of the simple fourth order equation as 2=0 dv brings in four undetermined constants: yx) = cyx? + cnx? + egy + cy It will be shown later in the text that, in general, the methods for solving nth order differential equations evoke n arbitrary constants. In most cases, we will be able to evaluate these constants if we know n initial values y(xo), y’(x),--- y"~ Y(ao). RTCA Definition 3. By an initial value problem for an nth order differential equation fy ay) _ run. =0, we mean: Find a solution to the differential equation on an interval / that satisfies at xp the n initial conditions ay aw! (0) = where xo & / and yo, Yi, +++ Yn—1 are given constants, In the case of a first order equation, the initial conditions reduce to the single requirement Y(%0) = Yo» EXAMPLE 6 SOLUTION EXAMPLE 7 SOLUTION Section 1.2 Solutions and Initial Value Problems W and in the case of a second order equation, the initial conditions have the form dy yo) = Yo» Go) =H» The terminology initial conditions comes from mechanics, where the independent variable x represents time and is customarily symbolized as t. Then if fp is the starting time, y(to) = yo represents the initial location of an object and dy(to)/dt gives its initial velocity. Show that (x) = sin x — cos x is a solution to the initial value problem dy 10) —S+ty=0; (0) Gate Observe that (x) = sin. x — cos x, dp/dx = cos.x + sin x, and d*/dx? = —sin x + cos x are all defined on (—oo, 00). Substituting into the differential equation gives (-sin x + cos.x) + (sin.x — cosx) =0, which holds for all x € (—o0, co). Hence (.x) is a solution to the differential equation in (10) ‘on (—oo, co). When we check the initial conditions, we find (0) = sin 0 ~ cos 0 = ~ Hy de )) =cos0+sinO=1, which meets the requirements of (10). Therefore (1x) is a solution to the given initial value problem. mend As shown in Example 2, the function (x) = ce" + ce” is a solution to are satisfied, To determine the constants c; and cy, we first compute db/dx to get db/dx = —cye™* + 2c,e*. Substituting in our initial conditions gives the following system of equations: (0) = cye® + ce® =2, ate, =2, di or Bo) = -ce? + 20° = -3, me, +26 = -3 Adding the last two equations yields 3c) = —1, so c; = ~1/3. Since c, + cy = 2, we find ¢, = 7/3. Hence the solution to the initial value problem is $(x) = (7/3)e* — (1/3)e”. mend 12 Chapter 1. Introduction We now state an existence and uniqueness theorem for first order ini We presume the differential equation has been cast into the format value problems éy #9) Of course the right-hand side, f(x, y), must be well defined at the starting value xg for x and at the stipulated initial value yp = y(x) for y. The hypotheses of the theorem, moreover, require continuity of both f and af/ay for x in some interval a < x cos x is a two-parameter family of solutions to the differential equation. 15. Show that (x)= Ce +1 is a solution to dy/dx — 3y = —3 for any choice of the constant C. Thus Ce** + 1 is a one-parameter family of solu- 4.x=3cost—Ssint, x"+x=0. x=cos2r, 6.6 =2e%—e7, 16. 17. 18. 19. 2. Section 1.2 tions to the differential equation. Graph several of the solution curves using the same coordinate axes. Verify that x2 + cy? = 1, where c is an arbitrary nonzero constant, is a one-parameter family of im- plicit solutions to and graph several of the solution curves using the same coordinate axes Verify that (x) = 2/(1 — ce"), where c is an arbi- trary constant, is a one-parameter family of solutions to 4 _xy~2) dx 2 . Graph the solution curves corresponding to +1, £2 using the same coordinate axes. Let c > 0. Show that the function (x) = (2 — °)"! isa solution to the initial value problem dy/dx = 2xy*, y(0) = 1/c, on the interval “c¢ +6 + Sy=0. ae Oa dy dy od) ) 432 422% 0. de de | “de Determine for which values of m the function (x) = x” is a solution to the given equation. aay dy 3S + Ix - 3y=0. @) 3x dx? “« Solutions and Initial Value Problems 15 22. Verify that the function (x) = cye* + cze~** is a solution to for any choice of the constants cy and c), Determine cc; and c2 so that each of the following initial condi- tions is satisfied. (a) y0)=2, y'(0)=1 (b) x)= 1, y'(I=0 In Problems 23 through 28, determine whether Theorem 1 implies that the given initial value problem has a unique solution. 28, + cosx=sinr, )=0. 29. For the initial value problem (12) of Example 9, show that ;(x) = 0 and ,(x) = (x — 2)° are so- lutions. Hence the initial value problem does not have a unique solution. 30. Implicit Function Theorem. Ler G(x, y) have con- tinuous first partial derivatives in the rectangle R={(xyhka

    You might also like