Appendix A Complex Variable Theory

TO ACCOMPANY AUTOMATIC CONTROL SYSTEMS EIGHTH EDITION

BY BENJAMIN C. KUO FARID GOLNARAGHI

JOHN WILEY & SONS, INC.

Copyright © 2003 John Wiley & Sons, Inc. 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, scanning or otherwise, except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923, (508)750-8400, fax (508)750-4470. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 605 Third Avenue, New York, NY 10158-0012, (212) 850-6011, fax (212) 850-6008, E-Mail: PERMREQ@WILEY.COM. To order books or for customer service please call 1-800-CALL WILEY (225-5945). ISBN 0-471-13476-7

APPENDIX

A
Complex-Variable Theory

A-1 COMPLEX-VARIABLE CONCEPT
A-1-1 Complex Variable
A complex variable s has two components: a real component and an imaginary component . Graphically, the real component of s is represented by a axis in the horizontal direction, and the imaginary component is measured along the vertical j axis, in the complex s-plane. Figure A-1 illustrates the complex s-plane, in which any arbitrary point s s1 is defined by the coordinates 1, and 1, or simply s1 j 1. 1

A-1-2

Functions of a Complex Variable
The function G(s) is said to be a function of the complex variable s if for every value of s, there is one or more corresponding values of G(s). Since s is defined to have real and imaginary parts, the function G(s) is also represented by its real and imaginary parts; that is, G1s2 Re G1s2 j Im G1s2 (A-1)

where Re G(s) denotes the real part of G(s), and Im G(s) represents the imaginary part of G(s). The function G(s) is also represented by the complex G(s)-plane, with Re G(s) as the real axis and Im G(s) as the imaginary axis. If for every value of s there is only one corresponding value of G(s) in the G(s)-plane, G(s) is said to be a single-valued function, and the mapping from points in the s-plane onto points in the G(s)-plane is described

jv s-plane v1 s1 s1 + jv1

0

s1

s

Figure A-1 The complex s-plane.

A-1

the mapping is called one-to-one. If the mapping from the G(s)-plane to the s-plane is also single-valued. A-1-4 Singularities and Poles of a Function The singularities of a function are the points in the s-plane at which the function or its derivatives does not exist. it is said to have a pole of order r at s si if the limit sSsi lim S 1s si 2 r G1s2 T has a finite. . for instance. there are many functions for which the mapping from the function plane to the complex-variable plane is not singlevalued. However. as single-valued (Fig. A-1-3 Analytic Function A function G(s) of the complex variable s is called an analytic function in a region of the s-plane if the function and all its derivatives exist in the region. the function G(s) s 2 is analytic at every point in the finite s-plane. In other words. given the function G1s2 1 s1s 12 (A-2) it is apparent that for each value of s.A-2 Appendix A Complex-Variable Theory jv s1 = s1 + jv 1 s-plane v1 G (s)-plane j Im G 0 s1 s 0 Re G G (s1) Figure A-2 Singlevalued mapping from the s-plane to the G(s)-plane. A pole is the most common type of singularity and plays a very important role in the studies of classical control theory. However. there is only one unique corresponding value for G(s). The definition of a pole can be stated as: If a function G(s) analytic and single-valued in the neighborhood of si. so when s si. If r 1. the function given in Eq. At these two points. the function G1s2 101s s1s 121s 22 32 2 (A-3) has a pole of order 2 at s 3 and simple poles at s 0 and s 1. the function becomes infinite. the pole at s si is called a simple pole. As an example. For instance. As another example. the denominator of G(s) must include the factor (s si)r. the inverse mapping is not true. It can also be said that the function G(s) is analytic in the s-plane except at these poles. the point G(s) is mapped onto two points. s 0 and s 1. nonzero value. the value of the function is infinite. For instance. A-2). (A-2) is analytic at every point in the s-plane except at the point s 0 and s 1. in the s-plane.

1965. since sSq lim G1s2 lim sSq s3 10 0 (A-4) Therefore. Prentice Hall. the function has a total of four poles and four zeros in the entire s-plane. Methods of Applied Mathematics. and taking into account of the poles and zeros at infinity. Englewood Cliffs. the function in Eq. 2nd ed. Hildebrand. G(s) has a zero of order r at s si if 1 G(s) has an rth-order pole at s si. B. (A-3) has four finite poles at s 0.. simply. 1. a quotient of two polynomials of s. . that is. but there are three zeros at infinity. there is one finite zero at s 2. counting the multiple-order poles and zeros. (2-3) has a simple zero at s 2. nonzero value. For example. the total number of poles equals the total number of zeros.A-1 Complex-Variable Concept A-3 A-1-5 Zeros of a Function The definition of a zero of a function can be stated as: If the function G(s) is analytic at s si. The function in Eq. including infinity. REFERENCES F. it is said to have a zero of order r at s si if the limit sSsi lim S 1s si 2 r G1s2 T • The total numbers of poles and zeros of a rational function is the same. Or. and 3. counting the ones at infinity has a finite. If the function under consideration is a rational function of s. NJ. 3.

Appendix B Differential and Difference Equations TO ACCOMPANY AUTOMATIC CONTROL SYSTEMS EIGHTH EDITION BY BENJAMIN C. KUO FARID GOLNARAGHI JOHN WILEY & SONS. INC. .

electronic. E-Mail: PERMREQ@WILEY. Danvers.COM. recording. stored in a retrieval system or transmitted in any form or by any means. mechanical. MA 01923. All rights reserved. fax (508)750-4470. without either the prior written permission of the Publisher. 222 Rosewood Drive. photocopying. Requests to the Publisher for permission should be addressed to the Permissions Department. fax (212) 850-6008.Copyright © 2003 John Wiley & Sons. except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act. Inc. To order books or for customer service please call 1-800-CALL WILEY (225-5945). (508)750-8400. (212) 850-6011. New York. scanning or otherwise. NY 10158-0012. No part of this publication may be reproduced. Inc. ISBN 0-471-13476-7 .. 605 Third Avenue. John Wiley & Sons. or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center.

an 1 are not functions of y(t). the differential equation of an nth-order system is written d n y1t2 dt n an d n 1y1t2 1 p dt n 1 a1 dy1t2 dt a 0 y1t2 f 1t2 (B-2) which is also known as a linear ordinary differential equation if the coefficients a0. the capacitance. For systems with distributed parameters. Equation (B-1) is referred to as a second-order differential equation. partial differential equations are used. In this case. In general. B-1-2 Nonlinear Differential Equations Many physical systems are nonlinear and must be described by nonlinear differential equations. the independent variable. the applied voltage. and we refer to the system as a second-order system. and i(t). e(t) is the forcing function. (B-1) should be referred to as an integrodifferential equation. B-1 is ML d 2u1t2 dt 2 Mg sin u1t2 0 (B-3) B-1 . Strictly speaking. a series electric RLC (resistance-inductance-capacitance) network can be represented by the differential equation Ri1t2 L di1t2 dt 1 i1t2dt C e1t2 (B-1) where R is the resistance. such as in heat-transfer systems. since we treat only systems that contain lumped parameters. These equations generally involve derivatives and integrals of the dependent variables with respect to the independent variable. the dependent variable or unknown that is to be determined by solving the differential equation. Eq. the current in the network. the differential equations encountered are all of the ordinary type. and e(t). For instance.APPENDIX B Differential and Difference Equations B-1 DIFFERENTIAL EQUATIONS B-1-1 Linear Ordinary Differential Equations A wide range of systems in engineering are modeled mathematically by differential equations. In this text. t. C. since an integral is involved. the differential equation that describes the motion of the pendulum shown in Fig. … . i(t). the inductance. For instance. a1. L.

let us define y1t2 dy1t2 dt d n 1y1t2 dt n 1 (B-8) then the nth-order differential equation is decomposed into n first-order differential equations: dxn 1t2 dt . first-order differential equations are used in the analytical studies of control systems. (B-3) is nonlinear. Since (t) appears as a sine function. for Eq. (B-1) is decomposed into the following two first-order differential equations: (B-6) (B-7) In a similar manner. in principle. Eq. an nth-order differential equation can be decomposed into n first-order differential equations. and the system is called a nonlinear system.B-2 Appendix B Differential and Difference Equations L u M Mg sin u Mg Figure B-1 Simple pendulum. Because. (B-1). (B-2). first-order differential equations are simpler to solve than higher-order ones. For the differential equation in Eq. if we let x1 1t2 and x2 1t2 dx1 1t2 dt dx2 1t2 dt x2 1t2 1 x 1t2 LC 1 x1 1t2 x2 1t2 xn 1t2 dx1 1t2 dt dx2 1t2 dt x2 1t2 x3 1t2 o a 0 x1 1t2 a1x2 1t2 p an 2 xn 1 1t2 an 1xn 1t2 f 1t2 (B-9) R x 1t2 L 2 1 e1t2 L i1t2dt dx1 1t2 dt (B-4) i1t2 (B-5) then Eq. B-1-3 First-Order Differential Equations: State Equations In general.

this requirement. In general. An output of a must always be measurable. such state variables as the winding current. present. since the former are more easily programmed on a digital computer and are easier to solve. (B-9) is called the state equations. and x1. an output variable can be expressed as an algebraic combination of the state variables. Definition of State Variables The state of a system refers to the past. … . x2(t). but a state variable does not always satisfy • The state of a system refers to the past. xn are called the state variables. rotor velocity. …. and future of the system. then the output equation is simply y(t) x1(t). present.B-1 Differential Equations B-3 Notice that the last equation is obtained by equating the highest-ordered derivative term in Eq. since it represents the past. x2. such that knowledge of these variables at any time t0 and information on the input excitation subsequently applied are sufficient to determine the state of the system at any time t t0. (B-8) are the state variables of the nth-order system described by Eq. xn(t) defined in Eq. there are some basic rules regarding the definition of a state variable and what constitutes a state equation. it is convenient to define a set of state variables and state equations to model dynamic systems. Once the inputs of the system for t t0 and the initial states just defined are specified. the variables x1(t). present. Difference equations are also used to approximate differential equations. The Output Equation • The output of a system One should not confuse the state variables with the outputs of a system. In general. For instance. xn(t0) define the initial states of the system. and these variables all qualify as output variables. (B-2) to the rest of the terms. system is a variable that can be measured. but it cannot be measured directly during operation and therefore does not ordinarily qualify as an output variable. k n denotes the discrete dependent variable y at the ith instant if the independent variable is time. In control systems theory. x1(t). magnetic flux can also be regarded as a state variable in an electric motor. x2(t). x2 (t0). and displacement can be measured physically. … . 2. the independent variable can be any real quantity. As it turns out. For the system described by Eq. … . In general. the set of first-order differential equations in Eq. 1. A linear nth-order difference equation with constant coefficients can be written as y1k n2 an 1y1k n 12 p a1y1k 12 a 0 y1k2 f 1k2 (B-10) where y(i). xn(t). The state variables of a system are defined as a minimal set of variables. in an electric motor. and the n first-order differential equations are the state equations. the state variables should completely define the future behavior of the system. (B-2). and future states of the motor. i k. … . if y(t) is designated as the output. On the other hand. difference equations are used for systems with discrete of digital data. it is necessary to establish equations that relate digital and discrete-time signals. and future conditions of the system. (B-2). k 1. the state variables x1(t0). • The state variables must always be a minimal set. Difference Equations Because digital controllers are frequently used in control systems. The state variables must satisfy the following conditions: At any initial time t t0. . Just as differential equations are used to represent systems with analog signals. From a mathematical perspective.

Advanced Engineering Mathematics. Linear Networks and Systems.B-4 Appendix B Differential and Difference Equations Similar to the case of the analog systems. and the last one is given by Eq. The n state equations are written in vector-matrix form: x1k where x1 1k2 x 1k2 D 2 T o xn 1k2 p p ∞ 0 p 0 0 o U 1 an 1 0 0 EoU 0 1 12 Ax1k2 Bu1k2 (B-13) x1k2 (B-14) is the n 1 state vector. REFERENCES 1. McGraw-Hill. 1967. R. or state equations. New York. C. McGraw-Hill.. if we let x1 1k2 x2 1k2 xn 1 1k2 xn 1k2 xn 1k 12 y1k2 x1 1k o xn 2 1k xn 1 1k 12 12 12 y1k y1k y1k 12 n n 22 12 f 1k2 (B-11) then by equating the highest-order term to the rest. 1960. . Jr. 2. (B-10). New York. C. For the difference equation in Eq. B. (B-11). Wylie. 2nd ed. the equation is written as a 0 x1 1k2 a1x2 1k2 p an 1xn 1k2 (B-12) The first n 1 state equations are taken directly from the last n 1 equations in Eq. (B-12). Kuo. to represent a high-order difference equation. and 0 0 E o 0 a0 1 0 o 0 a1 0 1 o 0 a2 A B (B-15) and u(k) f(k). it is convenient to use a set of first-order difference equations.

. KUO FARID GOLNARAGHI JOHN WILEY & SONS.Appendix C Elementary Matrix Theory and Algebra TO ACCOMPANY AUTOMATIC CONTROL SYSTEMS EIGHTH EDITION BY BENJAMIN C. INC.

COM. MA 01923. Danvers. mechanical. or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center. Inc. Inc. (508)750-8400. recording. No part of this publication may be reproduced. electronic. Requests to the Publisher for permission should be addressed to the Permissions Department.. stored in a retrieval system or transmitted in any form or by any means. All rights reserved. photocopying. New York. John Wiley & Sons. 605 Third Avenue. 222 Rosewood Drive.Copyright © 2003 John Wiley & Sons. fax (508)750-4470. except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act. (212) 850-6011. fax (212) 850-6008. ISBN 0-471-13476-7 . without either the prior written permission of the Publisher. To order books or for customer service please call 1-800-CALL WILEY (225-5945). E-Mail: PERMREQ@WILEY. NY 10158-0012. scanning or otherwise.

(C-1). As a motivation to the reason of using matrix notation. The matrix notation usually makes the equations much easier to handle and manipulate. The three matrices involved are defined as a11 a12 p a1n a21 a22 p a2n (C-3) A D T o o ∞ o an1 an2 p ann x x1 x2 D T o xn (C-4) y which are simply bracketed arrays of coefficients and variables. which will be discussed shortly.APPENDIX C Elementary Matrix Theory and Algebra C-1 ELEMENTARY MATRIX THEORY In the study of modern control theory. C-1 y1 y2 D T o yn (C-5) . These examples of matrices prompted the following definition of a matrix. and y are defined as matrices. The symbols A. (C-2) can be stated as the product of the matrices A and x is equal to the matrix y. In terms of matrix algebra. Eq. it is often desirable to use matrix notation to simplify complex mathematical expressions. x. let us consider the following set of n simultaneous algebraic equations: a11x1 a12x2 p a1n xn y1 a21x1 a22 x2 p a2n xn y2 #################################### (C-1) p ann xn yn an1x1 an2 x2 We may use the matrix equation Ax y (C-2) as a simplified representation of Eq. which contain the coefficients and variables of the original equations as their elements.

Column Matrix: A column matrix is one that has one column and more than one row. Determinant • An array of numbers or elements with n rows and n columns (always square). Square Matrix: A square matrix is one that has the same number of rows as columns. (C-3) through (C-5). where n 1. . A unity matrix is often designated by I or U. An example of a unity matrix is I 1 £0 0 0 1 0 0 0§ 1 (C-7) Null Matrix: A null matrix is one whose elements are all equal to zero. (C-4) is a typical n-vector. The basic characteristics of these are listed as follows: Matrix • An array of numbers or elements with n rows and m columns. the matrix in Eq. Matrix Elements: When a matrix is written A a11 a12 £ a21 a22 a31 a32 a13 a23 § a33 (C-6) where aij is defined as the element in the ith row and the jth column of the matrix. Diagonal Matrix: A diagonal matrix is a square matrix with aij Examples of a diagonal matrix are a11 £ 0 0 0 a22 0 0 0 § a33 c 5 0 0 d 3 0 for all i j. In this text. Row Matrix: A row matrix is one that has one row and more than one column. Quite often. Unity Matrix (Identity Matrix): A unity matrix is a diagonal matrix with all the elements on the main diagonal (i j) equal to 1. The matrix in Eq. although a square matrix (n m) has a determinant. Some important definitions of matrices are given in the following paragraphs. the square brackets. m 1. • It is important to distinguish between a matrix and a determinant. There are several ways of bracketing a matrix. • Has a value. a 1 n matrix. a column matrix is referred to as a column vector or simply an m-vector if there are m rows and one column. • Does not have a value. For example. are used to represent matrices. A matrix with n rows and m columns is termed n m. such as those in Eqs. It is important to distinguish between a matrix and a determinant.C-2 Appendix C Elementary Matrix Theory and Algebra C-1-1 Definition of a Matrix A matrix is a collection of elements arranged in a rectangular or square array. that is. As a rule. (C-6) has three rows and three columns and is called a 3 3 (three-by-three) matrix. Order of a Matrix: The order of a matrix refers to the total number of rows and columns of the matrix. that is. an m 1 matrix. or n by m. A row matrix can also be referred as a row-vector. we always refer to the row first and the column last.

. That is. or 2. Let A be an n n matrix. the same matrix is obtained. then the determinant of A can be written in terms of the cofactor of any row or the cofactor of any column. (C-10) is A11 1 12 1 1 ` a22 a32 a23 ` a33 a22a33 a23a32 (C-11) In general. p . A symmetric matrix has the property that if its rows are interchanged with its columns. n det A or a aij Aij j 1 1i 1. If a square matrix has a nonzero determinant. singularity of the matrix means that these equations are not independent of each other. or n2 (C-12) n det A a aij Aij i 1 1j 1. Two examples of the symmetric matrix are A 6 5 £5 0 1 10 1 10 § 1 B c 1 4 4 d 1 (C-8) Determinant of a Matrix: With each square matrix. When a matrix is singular. the cofactor of the element a11 of 0A 0 in Eq. the determinant of the matrix in Eq. it is called a nonsingular matrix. (C-6) is 0A 0 a11 † a21 a31 a12 a13 a22 a23 † a32 a33 (C-10) Cofactor of a Determinant Element: Given any nth-order determinant 0A 0 . or n2 (C-13) EXAMPLE C-1 The value of the determinant in Eq. the cofactor Aij of any element aij is the determinant obtained by eliminating all elements of the ith row and jth column and then multiplied by ( 1)i j. or 2. the value of a determinant can be written in terms of the cofactors. it usually means that not all the rows or not all the columns of the matrix are independent of each other.C-1 Elementary Matrix Theory C-3 Symmetric Matrix: A symmetric matrix is a square matrix that satisfies the condition aij aji for all i and j. For example. p . When the matrix is used to represent a set of algebraic equations. a determinant having the same elements and order as the matrix may be defined. (C-10) is det A 0A 0 a11 A11 a12 A12 a13 A13 a11 1a22a33 a23a32 2 a12 1a21a33 a13 1a21a32 a22a31 2 a23a31 2 (C-14) Singular Matrix: A square matrix is said to be singular if the value of its determinant is zero. The determinant of a square matrix A is designated by det A ¢A 0A 0 (C-9) For example.

In matrix form.C-4 Appendix C Elementary Matrix Theory and Algebra EXAMPLE C-2 Consider the following set of equations: 2x1 x1 x1 3x2 x2 2x2 x3 x3 2x3 0 0 0 (C-15) The third equation is equal to the sum of the first two. denoted by adj A. 3Aij of det A4 ¿n. A¿ 3 £2 1 0 1§ 5 (C-21) Some Properties of Matrix Transpose 1. is defined as adj A where Aij denotes the cofactor of aij. the matrix A is singular.n n. (AB) BA (C-22) (C-23) (C-24) (C-25) Adjoint of a Matrix: Let A be a square matrix of order n. In this case. Thus. The adjoint matrix of A. thus. (A B) A B 4. denoted by A . (kA) kA . and. these equations may be written as Ax where A 2 £ 1 1 3 1 2 1 1§ 2 x £ x2 § x3 x1 (C-17) 0 (C-16) and 0 is a 3 1 null matrix. the rows of A are dependent. these three equations are not completely independent. but the order of A is m The transpose of A is obtained by interchanging the rows and the columns. (A ) A 2. is given by A¿ Notice that the order of A is n EXAMPLE C-3 Consider the 2 3 matrix A c 3 0 2 1 d 1 5 (C-20) transpose of A (C-19) m.n (C-26) .m 3aij 4 m. Let A be an n m matrix that is represented by A 3aij 4 n. (C-18) The transpose of A. The determinant of A is 0. Transpose of a Matrix: The transpose of a matrix A is defined as the matrix that is obtained by interchanging the corresponding rows and columns in A. where k is a scalar 3.

multiplication. The order of the matrices is preserved after addition or subtraction.m aij bij 3bij 4 n. and division. a22 b22. C-2-1 Equality of Matrices Two matrices A and B are said to be equal to each other if they satisfy the following conditions: 1. denoted as tr(A). a12 b12. subtraction. the adjoint matrix of A is c a22 a21 a12 d a11 (C-28) The cofactors are A11 a22.m C B if they are of the same (C-34) (C-35) 3cij 4 n. a21 b21.C-2 Matrix Algebra C-5 EXAMPLE C-4 Consider the 2 2 matrix A c a11 a12 d a21 a22 a12. and A22 c a22 a12 a21 ¿ d a11 (C-27) a11. The corresponding elements are equal. C-2-2 Addition and Subtraction of Matrices Two matrices A and B can be added or subtracted to form A order. is defined as the sum of the elements on the main diagonal of A. the trace of A. That is A where cij B 3aij 4 n. . A12 c A11 A21 a21. They are of the same order.m for all i and j. 2. it is necessary to define matrix algebra in the form of addition. tr1A B2 tr1A2 tr1B2 (C-30) (C-31) C-2 MATRIX ALGEBRA When carrying out matrix operations. that is n tr1A2 a aii i 1 (C-29) The trace of a matrix has the following properties: 1. tr(A ) 2. For n tr(A) n square matrices A and B. that is. aij bij for every i and j c a11 a21 a12 d a22 c b11 b21 b12 d b22 (C-32) EXAMPLE C-5 A B (C-33) implies that a11 b11. Thus. A21 A12 ¿ d A22 adj A Trace of a Square Matrix: Given an n n matrix with elements aij.

usually AB BA.m (C-40) (C-41) Then A and B are conformable to form the product 3aij 4 n. The following references are made with respect to matrix manipulation whenever they exist: AB A postmultiplied by B or B premultiplied by A (C-42) C-2-6 Rules of Matrix Multiplication When the matrices A (n p) and B (p the ijth element of C. The matrix C will have the same number of rows as A and the same number of columns as B. Then the sum of A and B is C A B £ 3 1 0 0 1 1 2 4 1 3 2§ 0 3 £ 2 1 5 6§ 1 (C-37) C-2-3 Associative Law of Matrix (Addition and Subtraction) The associative law of scalar algebra still holds for matrix addition and subtraction. let A C AB 3aij 4 n. 1A B2 C A 1B C2 (C-38) C-2-4 Commutative Law of Matrix (Addition and Subtraction) The commutative law for matrix addition and subtraction states that the following matrix relationship is true: A B C B C A A C B (C-39) as well as other possible commutative combinations. C-2-5 Matrix Multiplication The matrices A and B may be multiplied together to form the product AB if they are conformable. It is important to note that A and B may be conformable to form AB. a aik bkj k 1 (C-43) . This points out an important fact that the commutative law is not generally valid for matrix multiplication.m 3cij 4 n. if and only if p q. cij. n also equals m. is given by cij for i 1. m) are conformable to form the matrix C p AB.p 3bij 4 q.C-6 Appendix C Elementary Matrix Theory and Algebra EXAMPLE C-6 Consider the matrices A 3 £ 1 0 2 4§ 1 B 0 £ 1 1 3 2§ 0 (C-36) which are of the same order. That is. unless in Eq. but they may not be conformable for the product BA. This means that the number of columns of A must equal the number of rows of B. 2. p . It is also noteworthy that even though A and B are conformable for both AB and BA. n. 2. p . m. p B 3bij 4 q. and j 1.m • Two matrices can be multiplied only if they are conformable. In other words. (C-41).

Thus. even though AB and BA both exist. . The conditions that A A is a square matrix. 1AB2C if the product is conformable. In matrix algebra. Although the commutative law does not hold in general for matrix multiplication. A1BC2 (C-50) C-2-7 Multiplication by a Scalar k Multiplying a matrix A by any scalar k is equivalent to multiplying each element of A by k.1 (C-44) The two matrices are conformable for the product AB but not for BA. For the associative law. C-2-8 Inverse of a Matrix (Matrix Division) y a is also (C-51) 1 • Only square. if Ax y. In fact. 1 A 1y exists are denotes the matrix inverse of A.3 B 3bij 4 3. the associative and distributive laws are valid. we state that A1B C2 AB AC (C-49) if the products are conformable. the it may be possible to write x where A 1. A must be nonsingular.C-2 Matrix Algebra C-7 EXAMPLE C-7 Given the matrices A 3aij 4 2. in this case the products are not of the same order. when we write y ax. a11 a12 c a21 a22 b11 a13 £ b21 § d a23 b31 c a11b11 a21b11 a12b21 a22b21 a13b31 d a23b31 AB (C-45) EXAMPLE C-8 Given the matrices A 3 £0 2 1 1§ 0 B c 1 2 0 1 1 d 0 (C-46) The two matrices are conformable for AB and BA. nonsingular In the algebra of scalar quantities. they are not equal. AB 3 £0 2 c 1 1 1§ c 2 0 1 0 2 1 0 1 3 1 d £0 0 2 1 d 0 1 1§ 0 1 £2 2 c 1 6 1 1 0 1 d 1 3 0§ 2 (C-47) BA (C-48) Therefore. For the distributive law. 2. it implies that x matrices have inverses. true.

where for A to be nonsingular. that is. However. AA 1 A 1A I 1 1 2. then 1AB2 1 B 1A 1 (C-61) C-2-9 Rank of a Matrix The rank of a matrix A is the maximum number of linearly independent columns of A. if A is a square. Equation (C-54) shows that adj A of a 2 2 matrix is obtained by interchanging the two elements on the main diagonal and changing the signs of the elements off the diagonal of A. 4. in general. In matrix algebra. AB AC does not necessarily imply that B C. nonsingular matrix. we can premultiply both sides of AB AC by A 1. If A and B are square matrices and are nonsingular. it is the order of the largest nonsingular matrix contained in A. The reader can easily construct an example to illustrate this property. (A ) A (C-59) 3. A 1AB A 1AC (C-60) which leads to B C. it is given by A 1 adj A 0A 0 c a11 a12 d a21 a22 (C-52) EXAMPLE C-9 Given the matrix A the inverse of A is given by • The adjoint of a 2 by 2 a12 a22 d c matrix is obtained by adj A a21 a11 A 1 (C-54) interchanging the two main a11a22 a12a21 0A 0 diagonal elements and changing the signs of the 0. . If A 1 exists.C-8 Appendix C Elementary Matrix Theory and Algebra 3. 0A 0 elements off the diagonal. the adjoint of A is adj A £ a22a33 1a21a33 a21a32 a23a32 a23a31 2 a22a31 1 a12a33 a11a33 1 a11a32 a13a32a21 a13a32 2 a13a31 a12a31 2 a13a 22a31 a12a23 1a11a23 a11a22 a13a22 a21a13 2 § a12a21 (C-56) a11 a12 £ a21 a22 a31 a32 a13 a23 § a33 (C-55) The determinant of A is 0A 0 a11a22a33 a12a23a31 a12a21a33 a11a23a32 (C-57) Some Properties of Matrix Inverse (C-58) 1. or a11a22 a12a21 0. Then. (C-53) EXAMPLE C-10 Given the matrix A To find the inverse of A.

Bellman. Theory and Problems of Matrices. Rank of AA Rank of A. 2. Rank of A Rank of A. McGraw-Hill. Shaum’s Outline Series. since A A and AA are always square. the rank condition can be checked by evaluating the determinant of these matrices. Rank of A A Rank of A. McGraw-Hill. . 3. F. New York. REFERENCES 1. C-3 COMPUTER-AIDED SOLUTIONS OF MATRICES Many commercial software packages such as MATLAB. Ayres. Given m matrix A. 1962.References C-9 EXAMPLE C-11 Several examples on the rank of a matrix are as follows: c 3 £1 2 0 1 d 0 0 9 2 3 0§ 6 1 rank 1 c 0 3 5 1 4 d 0 3 2 3 0 0 £1 2 0§ 0 0 1 rank 2 rank 2 rank 3 an n The following properties are useful in the determination of the rank of a matrix.. Properties 2 and 3 are useful in the determination of rank. New York. 1960. 1. Introduction to Matrix Analysis. Maple and MATHCAD contain routines for matrix manipulations. 2. Jr. R.

KUO FARID GOLNARAGHI JOHN WILEY & SONS. .Appendix D Laplace Transform Table TO ACCOMPANY AUTOMATIC CONTROL SYSTEMS EIGHTH EDITION BY BENJAMIN C. INC.

fax (508)750-4470. No part of this publication may be reproduced. (212) 850-6011. or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center. stored in a retrieval system or transmitted in any form or by any means..Copyright © 2003 John Wiley & Sons. Inc. Requests to the Publisher for permission should be addressed to the Permissions Department. 222 Rosewood Drive.COM. John Wiley & Sons. New York. All rights reserved. without either the prior written permission of the Publisher. E-Mail: PERMREQ@WILEY. (508)750-8400. 605 Third Avenue. electronic. ISBN 0-471-13476-7 . recording. scanning or otherwise. To order books or for customer service please call 1-800-CALL WILEY (225-5945). fax (212) 850-6008. NY 10158-0012. Danvers. Inc. MA 01923. photocopying. except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act. mechanical.

APPENDIX D Laplace Transform Table Laplace Transform F(s) 1 1 s 1 s2 n! sn 1 1 s 1s 1s 1s 1s 1 a2 2 n! a2 n 1 a2 1s s a2 1s 1 s1s 1 s1s s2 1s s2 1s 1s 1 a2 1 a2 2 s a2 2 a2 2 a2 1 11 a2 1 1at a2 1 ct a2 2 a 11 1 Time Function f 1t2 Unit-impulse function d1t2 Unit-step function us 1t2 Unit-ramp function t tn 1n positive integer2 e te tne 1 at t a t 1n 1e positive integer2 at bt b2 b 1 a e 2 1a at b2 b2 b a 1be bt ae 2 1a b2 1 11 a e at e at 2 at ate at 2 1 at at2e e 2 at 2 be a at d D-1 .

D-2 Appendix D Laplace Transform Table (continued) Laplace Transform F(s) vn s2 s s2 v2 n s1s2 v2 1s n s 1s 2 Time Function f(t) sin cos nt v2 n v2 n v2 2 n a2 v2 n v2 2 n vn 21 1 a2 v2 n 1 nt vn 2a2 v2 sin 1vnt u2 n where u tan 1 1vn a2 cos nt a2 1s2 v2 n vn e sin 1vnt 2a v2 n 1 where u tan 1vn a2 at zvn t vn s 2 2zvns v2 n v2 n v2 2 n s1s2 2zvns sv2 n s2 2zvns v2 1s n 2zvns v2 n a2 v2 n s2 vn B e 21 z2 where u v2 n a2 2azvn 1 z2 21 z2 where u z2 e 1 sin vn 21 2 1 u2 e zvn t cos zvnt sin 1vn 21 1 sin 1vn 21 z z2 t 1z 6 12 z2 t u2 1z 6 12 z2 t u2 v2 n 1 s 1s 2 v2 n 2zvns 2 v2 2 n t 2z 1 e zvnt sin 1vn 21 z2 t vn vn 21 z2 where u cos 1 12z2 12 1z 6 12 2 where u tan vn 21 z a zvn e zvn t cos 1 z sin 1vn 21 1z 6 12 z2 t 12 u2 1z u2 .

INC.Appendix E Operational Amplifiers TO ACCOMPANY AUTOMATIC CONTROL SYSTEMS EIGHTH EDITION BY BENJAMIN C. KUO FARID GOLNARAGHI JOHN WILEY & SONS. .

All rights reserved. To order books or for customer service please call 1-800-CALL WILEY (225-5945). No part of this publication may be reproduced. fax (212) 850-6008. fax (508)750-4470. scanning or otherwise. Requests to the Publisher for permission should be addressed to the Permissions Department. except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act. 222 Rosewood Drive. or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center. without either the prior written permission of the Publisher. stored in a retrieval system or transmitted in any form or by any means. NY 10158-0012. Inc. New York. recording.. MA 01923. Danvers. E-Mail: PERMREQ@WILEY.COM. Inc. 605 Third Avenue. mechanical. electronic. ISBN 0-471-13476-7 . (212) 850-6011. John Wiley & Sons.Copyright © 2003 John Wiley & Sons. photocopying. (508)750-8400.

E-1-1 The Ideal Op-Amp When good engineering practice is used. E-1 .APPENDIX E Operational Amplifiers E-1 OPERATIONAL AMPLIFIERS MATLAB Operational amplifiers. e e . 4. and has the following properties: The voltage between the and terminals is zero. while keeping in mind that higher-order transfer functions are also important. In control systems. In fact. or simply op-amps. Thus. The ideal op-amp circuit is shown in Fig. + e– + e+ – – – + + eo – Figure E-1 Schematic diagram of an op-amp. Our primary goal here is to show how to implement first-order transfer functions with op-amps. A(e e ) where the gain A approaches 1. the output is an ideal voltage source. An in-depth presentation of op-amps is beyond the scope of this text. so in this appendix we illustrate common op-amp configurations. Some of the practical issues associated with op-amps are demonstrated in simlab (see Chapter 11). This property is commonly called the virtual ground or virtual short. an op-amp circuit can be accurately analyzed by considering the op-amp to be ideal. the input impedance is infinite. op-amps are often used to implement the controllers or compensators that evolve from the controlsystem design process. many texts are available that are devoted to all aspects of op-amp circuit design and applications [References 1 and 2]. For those interested. E-1. The currents into the and input terminals are zero. The impedance seen looking into the output terminal is zero. The input-output relationship is eo infinity. Only a representative sample of the multitude of op-amp configurations will be discussed. or realize continuous-data or s-domain transfer functions. 3. simple high-order transfer functions can be implemented by connecting first-order op-amp configurations together. Thus. 2. that is. implement. offer a convenient way to build.

E-1. . E-1-2 Sums and Differences As illustrated in the last chapter. Rather. R + – + eb – R R ea – + + eo = –(ea + eb) – (a) R R – R ea + – + eb – – (b) R + – R + eb – R + eo = eb – ea R R eo = ea + eb + + ea – + – (c) Figure E-2 Op-amps used to add and subtract signals. linear operation requires the addition of feedback of the output signal to the input terminal. When these signals are voltages. An op-amp cannot be used as shown in Fig.E-2 Appendix E Operational Amplifiers The input-output relationship for many op-amp configurations can be determined by using these principles. E-2. op-amps provide a simple way to add or subtract signals as shown in Fig. one of the most fundamental elements in a block diagram or an SFG is the addition or subtraction of signals.

the input-output relationship. E-2(b) can be used. Using superposition and the ideal properties given in the preceding section. E-2(a) is vo (va vb). E-3 and using resistors and capacitors as elements to compose Z1(s). E-2(c). all the transfer functions have negative gains. The negative gain is usually not an issue.E-1 Operational Amplifiers E-3 where all the resistors have the same value. it is possible to implement poles and zeros along the negative real axis as well as at the origin in the s-plane. the output is the negative sum of the input voltages. E-2(b) slightly gives the differencing circuit shown in Fig. Because the inverting op-amp configuration was used. op-amps can be used to implement transfer functions of continuous-data systems. Using the inverting op-amp configuration shown in Fig. and Z2(s). (E-1) apply conveniently to the different compositions of the circuit impedances. Using ideal op-amp properties. The different transfer function forms given in Eq. such as Eo 1s2 Ei 1s2 Z2 1s2 Z1 1s2 1 Z1 1s2Y2 1s2 Y1 1s2 Y2 1s2 G1s2 (E-1) Z2 1s2Y1 1s2 where Y1(s) 1 Z1(s) and Y2(s) 1 Z(s) are the admittances associated with the circuit impedances. Z2(s) Z1(s) – + + Eo(s) – + Ei (s) – Figure E-3 Inverting op-amp configuration. . Thus. since it is simple to add a gain of 1 to the input and output signal to make the net gain positive. While many alternatives are available. Modifying Fig. In the figure. which has an input-output relationship of eo eb ea. Inductors are not commonly used because they tend to be bulkier and more expensive. E-3 can be written in a number of ways. the circuit shown in Fig. E-3. When a positive sum is desired. As shown in the Table E-1. the input-output relationship in Fig. or transfer function. Here the output is given by eo ea eb. Z1(s) and Z2(s) are impedances commonly composed of resistors and capacitors. Table E-1 illustrates a number of common transfer function implementations. E-1-3 First-Order Op-Amp Configurations In addition to adding and subtracting signals. of the circuit shown in Fig. we will explore only those that use the inverting op-amp configuration shown in Fig.

an integrator.E-4 Appendix E Operational Amplifiers TABLE E-1 Inverting Op-Amp Transfer Functions Input Element (a) R1 Z1 = R1 R1 Feedback Element R2 Z2 = R2 C2 Transfer Function R2 R1 a 1 1 b R1C2 s Comments Inverting gain. if R1 R2. In Chapter 10 this transfer function will be called the PID controller. the proportional gain can be implemented using line (a). the second an Integral term.g. i.. and derivative . R2 (g) C1 as C2 s C1 Y1 = 1 + sC1 R1 C2 Y2 = 1 + sC2 R2 1 b R1C1 1 R2C2 Pole at s 1 and a R2C2 1 zero at s . R1C1 a PD controller.e. i. a PI Controller... (c) Y1 = sC1 ( R2C1)s Zero at the origin.e. This sum can be implemented by adding an additional input resistance to the circuit shown in Fig.e. a differentiator. the negative gains of the proportional. (f ) C1 1 + sC Y1 = 1 R1 R1 R2 Z2 = R2 R2C1 as 1 b R1C1 Zero at s 1 . eo e1.e. the output of G(s) is the sum of the responses due to each term in G(s).. EXAMPLE E-1 As an example of op-amp realization of transfer functions. integral. a lead or lag controller. and KI are real constants. i. R1 (e) Z1 = R1 R1 Z2 = R2 + 1 sC2 R2 s a R1 1 R2C2 b s Pole at the origin and a zero at 1 R2C2. i. By making the sum negative. KD.. and the third a Derivative term. (d) R1 Z1 = R1 C2 Y2 = 1 + sC2 R2 R2 C2 1 R1C2 1 s R2C2 Pole at 1 with a dc R2C2 gain of R2 R1. since the first term is a Proportional gain. e. the integral term can be implemented using line (b). E-2(a). and the derivative term can be implemented using line (c). By superposition. R1C1 i.e. (b) Z1 = R1 Y2 = sC2 C1 R2 Z2 = R2 R2 Pole at the origin. consider the transfer function MATLAB G1s2 Kp KI s KDs (E-2) where KP.. Using Table E-1.

giving the desired result shown in Fig. E-4. which is often referred to as antiwindup protection. One advantage of the implementation shown in Fig. For example. (E-2) and (E-7). term implementations are canceled. (E-2). the transfer function of the PID op-amp circuit is G1s2 Rd Cds (E-7) By equating Eqs. . the design is completed by choosing the values of the resistors and the capacitors of the op-amp circuit so that the desired values of KP. E-4 is that each of the three constants KP. and KD are matched. E-4 is just one of many possible implementations of Eq. it is possible to implement the PID controller with just three op-amps. The design of the controller should be guided by the availability of standard capacitors and resistors. it is common to add components to limit the high-frequency gain of the differentiator and to limit the integrator output magnitude.E-1 Operational Amplifiers E-5 R2 R1 – KP + Ep(s) R Ci Ri E(s) R – KI + EI (s) – R + Rd Cd – KD + ED (s) R Eo(s) Figure E-4 Implementation of a PID controller. Also. and KD can be adjusted or tuned individually by varying resistor values in their respective op-amp circuits. KI. KI. It is important to note that Fig. The transfer functions of the components of the op-amp circuit in Fig. E-4 are Proportional: Integral: Derivative: The output voltage is Eo 1s2 3EP 1s2 Eo 1s2 E1s2 R2 R1 EI 1s2 1 RiCis ED 1s2 4 (E-6) EP 1s2 E1s2 EI 1s2 E1s2 ED 1s2 E1s2 R2 R1 1 RiCis Rd Cds (E-3) (E-4) (E-5) Thus.

Korn. V. and realization of nonlinear elements for system compensation. Operational Amplifier Circuits. Holt. 1988. P. and G.. Huelsman. 2. PROBLEM E-1. New York. McGraw-Hill. 1992. EP-1. REFERENCES 1. Wait. E. Fort Worth. TX. Kennedy. Rinehart and Winston. Second Edition. J. A. R2 R1 C1 – + + Eo(s) – R2 C1 + E(s) R1 – + – + Eo(s) – (b) C2 (a) + E(s) – RF + E(s) – C R1 – R2 + + Eo(s) – (c) C R1 – R2 + RF + E(s) – + Eo(s) – (d) Figure EP-1 . L.E-6 Appendix E Operational Amplifiers Op-amps are also used in control systems for A D and D A converters. Find the transfer functions Eo(s) E(s) for the circuits shown in Fig. sampling devices. J. Introduction to Operational Amplifier Theory and Applications.

Appendix F Properties and Construction of the Root Loci TO ACCOMPANY AUTOMATIC CONTROL SYSTEMS EIGHTH EDITION BY BENJAMIN C. KUO FARID GOLNARAGHI JOHN WILEY & SONS. INC. .

without either the prior written permission of the Publisher. electronic. photocopying. John Wiley & Sons. New York.COM. 222 Rosewood Drive. stored in a retrieval system or transmitted in any form or by any means. fax (212) 850-6008. Requests to the Publisher for permission should be addressed to the Permissions Department.Copyright © 2003 John Wiley & Sons. E-Mail: PERMREQ@WILEY. scanning or otherwise. Danvers. or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center. (212) 850-6011. All rights reserved. fax (508)750-4470.. To order books or for customer service please call 1-800-CALL WILEY (225-5945). mechanical. MA 01923. recording. 605 Third Avenue. ISBN 0-471-13476-7 . No part of this publication may be reproduced. NY 10158-0012. (508)750-8400. Inc. Inc. except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act.

2. if any. It is useful to consider that infinity in the s-plane is a point concept. EXAMPLE F-1 Consider the equation s1s 221s 32 K1s 12 0 (F-2) When K 0. the three roots of the equation are at s 1. When the magnitude of K is infinite. (8-12). as the magnitude of K approaches infinity. Then. The properties are developed based on the relationship between the poles and zeros of G(s)H(s) and the zeros of 1 G(s)H(s). the three roots of the equation are at s 0. The K points on the root loci are at the zeros of G(s)H(s). infinity in the s-plane is a point on the opposite side of the sphere that we face. we get 1 which gives G1s2H1s2 K1s s1s 221s 12 32 (F-4) G1s2H1s2 1 K1s s1s 221s 12 32 0 (F-3) F-1 . G1(s)H1(s) approaches infinity.APPENDIX F Properties and Construction of the Root Loci The following properties of the root loci are useful for constructing the root loci manually and for understanding the root loci. (F-2) by the terms that do not contain K. and 3. The reason for this is seen from the condition of the root loci given by Eq. and . so s must approach the poles of G1(s)H1(s) or G(s)H(s). s must approach the poles of G(s)H(s). which are the roots of the characteristic equation. Similarly. We can visualize that the finite s-plane is only a small portion of a sphere with an infinite radius. F-1 K 0 and K Points The K 0 points on the root loci are at the poles of G(s)H(s). G1 1s2H1 1s2 1 K (F-1) As the magnitude of K approaches zero. Dividing both sides of Eq. The poles and zeros referred to here include those at infinity.

we can regard this axis of symmetry as if it were the real axis of a new complex plane obtained through a linear transformation. In other words. In general. 0 3) Thus. it is up to the user to make the distinctions. The following property of the root loci results. (F-2) when K 0 are the same as the poles of the function G(s)H(s). ticularly true when the root locus plot is done by a computer. since unless each root locus branch is coded by a different color. The reason behind this property is because for a polynomial with real coefficients the roots must be real or in complex-conjugate pairs. (F-2) when K are at the three zeros of G(s)H(s).F-2 Appendix F Properties and Construction of the Root Loci jv s-plane K=0 –3 K=0 –2 K=± –1 0 K=0 s Figure F-1 Roots at which K on the root loci of s(s 2)(s K(s 1) 0. F-3 SYMMETRY OF THE ROOT LOCI • It is important to pay attention to the symmetry of the root loci. The three roots of Eq. In general. the root loci are symmetrical with respect to the axes of symmetry of the polezero configuration of G(s)H(s). The number of branches of the root loci of F(s) P(s) KQ(s) 0 is equal to the order of the polynomial. . and thus there should be three root loci. EXAMPLE F-2 The number of branches of the root loci of s1s 221s 32 K1s 12 0 (F-5) is three. the equation has three roots. F-2 NUMBER OF BRANCHES ON THE ROOT LOCI A branch of the root loci is the locus of one root when K varies between and . if the poles and zeros of G(s)H(s) is symmetrical to an axis in addition to the real axis in the s-plane. since the number of branches of the root loci must equal the number of roots of the equation. the three roots of Eq. The three points on the root loci at which K 0 and those at which K are shown in Fig. • It is important to keep Keeping track of the individual branches and the total number of branches of the root track of the total number of locus diagram is important in making certain that the plot is done correctly. since the equation is of the third order. The root loci are symmetrical with respect to the real axis of the s-plane. F-1. This is parbranches of the root loci. including those at infinity.

and 2. 1. s 0. the root locus plot is symmetrical to the two axes. The points at which K 0 are at the poles of G(s)H(s).F-3 Symmetry of the Root Loci jv K F-3 – K s-plane K<0 K>0 K K>0 K=0 –2 K=0 –1 K=0 0 K<0 K – s Axis of symmetry K<0 K>0 K – Axis of symmetry K K>0 K<0 Figure F-2 Root loci of s(s 2)(s 3) showing the properties of symmetry. (F-6) are shown in Fig. through the K 0 point on the same branch. F-2. . we conduct the following exercise with regard to the root loci in Fig. F-2 for K to K . The function G(s)H(s) has three zeros at s at which K . we get G1s2H1s2 K 121s (F-7) s1s 22 The root loci of Eq. and ending at K at s . The reader should try to trace out the three separate branches of the root loci by starting from one of the K points. EXAMPLE F-3 Consider the equation s1s 121s 22 K 0 (F-6) Dividing both sides of the equation by the terms that do not contain K. As a review of all the properties of the root loci presented thus far. Since the pole-zero configuration of G(s)H(s) is symmetrical with respect to the real axis as well as the s 1 axis. K(s 1) 0.

the order of Q(s). the root loci for K 0 (RL) are asymptotic to asymptotes with angles given by ui 12i 0n 12 m0 180° n m (F-9) where i 0. and n and m are the number of finite poles and zeros of G(s)H(s). 1. The angles of the asymptotes and their intersect with the real axis of the s-plane are described as follows. F-3. For large values of s. 2 0n m 0 of the loci will approach infinity in the s-plane. n m 1. 2. the order of P(s) is not equal to m. 2s 2) K 0. The properties of the root loci near infinity in the s-plane are described by the asymptotes of the loci when 0s 0 S .F-4 Appendix F Properties and Construction of the Root Loci EXAMPLE F-4 When the pole-zero configuration of G(s)H(s) is symmetrical with respect to a point in the s-plane. … . when n. respectively. . F-4 ANGLES OF ASYMPTOTES OF THE ROOT LOCI AND BEHAVIOR OF THE ROOT LOCI AT |s| • Asymptotes of root loci refers to behavior of root loci at s S . As shown by the root loci in Figs. F-2 and F-3. This is illustrated by the root locus plot of s1s shown in Fig. 121s 1 j21s 1 j2 K 0 (F-8) – jv K K K=0 K s-plane j1 – K K=0 –2 –1 K=0 0 K – –j1 K=0 K K K K>0 – K<0 Figure F-3 Root loci of s(s 2)(s2 showing the properties of symmetry. the root loci will also be symmetrical to that point.

respectively. K 0: The points at which K 1 j. K 4. (F-11). (8-35) always cancel each other out. n F-5 INTERSECT OF THE ASYMPTOTES (CENTROID) The intersection of the 2|n the s-plane. the imaginary parts in the numerator of Eq. There are four root loci branches. the following information concerning the root loci of Eq. 2. The intersection of the asymptotes 1 represents the center of gravity of the root loci. That is. … . 4. and . 1. (F-13) and (F-14) are of the fourth order. respectively. . F-4. Since the poles and zeros of G(s)H(s) are either real or in complex-conjugate pairs. since Eqs. at s1 m| asymptotes of the root loci lies on the real axis of © finite zeros of G1s2H1s2 m © finite poles of G1s2H1s2 n (F-11) where n is the number of finite poles and m is the number of finite zeros of G(s)H(s). and is always a real number. and 1 j. 180° n m (F-10) where i 0. 2. (F-14) when K varies from to is obtained: 1. in Eq. 3. s1 © real parts of poles of G1s2H1s2 n © real parts of zeros of G1s2H1s2 m (F-12) EXAMPLE F-5 Consider the transfer function G1s2H1s2 K1s s1s 421s 2 12 2s 22 (F-13) which corresponds to the characteristic equation s1s 421s2 2s 22 K1s 12 0 (F-14) The pole-zero configuration of G(s)H(s) is shown in Fig. From the six properties of the root loci discussed so far. . . 0 on the root loci are at the poles of G(s)H(s): s on the root loci are at the zeros of G(s)H(s): s 0. : The points at which K 1.F-5 Intersect of the Asymptotes (Centroid) F-5 For K 0 (RL). Thus. the angles of the asymptotes are ui 0n m 2i m0 1. the terms in the summations may be replaced by the real parts of the poles and zeros of G(s)H(s). The root loci are symmetrical to the real axis.

three root loci approach s . 5. and are The angles of the asymptotes of the root loci for K calculated to be 0 . (F-9): 60° 180° 300° 0 are given by Eq. K< 0 K=0 0 K<0 s K=0 –j K> 0 . Since the number of finite poles of G(s)H(s) exceeds the number of finite zeros of G(s)H(s) by three (n m 4 1 3). (F-10). The intersection of the asymptotes is given by Eq. 6. The angles of the asymptotes of the RL (K j j j 0: 1: 2: u0 u1 u2 180° 3 540° 3 900° 3 0) are given by Eq. 120 . F-4.F-6 Appendix F Properties and Construction of the Root Loci jv s-plane K< K=0 60° K>0K=0 –4 60° K<0 s1 = –5 3 K=± –1 K > 0 60° 60° K >0 j 0 Figure F-4 Asymptotes of the root loci of s(s 4)(s2 2s 2) K(s 1) 0. and 240 . (F-12): s1 1 4 1 4 12 1 1–12 5 3 (F-15) The asymptotes of the root loci are shown in Fig. when K .

. F-5. jv s-plane K>0 s-plane K<0 jv K K<0 0 1 K<0 s K<0 s1 = –(p1 + p2 + p3) 4 > 0 K > 45° K<0 45° 0 s = –p1/ 2 K>0 0 K > 0 G(s)H(s) = K>0 K s(s + p1) K<0 G(s)H(s) = K s(s + p1)(s + p2)(s + p3) K<0 K>0 s-plane K<0 jv K<0 K<0 K < K >0 0 45° K<0 s s1= K > K>0 –(p1 + p2 + p3) 5 36° 36° 0 K > K<0 –(p1 + p2 + p3) – (–z1) 4 0 45° 0 K > 0 K jv K> 0 > 0 K<0 s K <0 0 0 > 0 K< K K> K > K<0 0 G(s)H(s) = K s2(s + p1)(s + p2)(s + p3) K<0 0 G(s)H(s) = K(s + z1) s2(s + p1)(s + p2)(s + p3) K<0 K>0 K>0 Figure F-5 Examples of the asymtotes of the root loci.F-5 Intersect of the Asymptotes (Centroid) F-7 EXAMPLE F-6 The asymptotes of the root loci of several equations are shown in Fig.

jv s-plane K>0 K<0 K>0 K<0 K>0 0 K<0 s jv s-plane K<0 K>0 K<0 K>0 K<0 0 K<0 s K>0 K<0 Figure F-6 Properties of root loci on the real axis. 2. there must be an odd number of poles and zeros of G(s)H(s) to the right of the point. The last observation shows that for s1 to be a point on the root locus. the total number of poles and zeros of G(s)H(s) to the right of the point must be even.F-8 Appendix F Properties and Construction of the Root Loci F-6 ROOT LOCI ON THE REAL AXIS • The entire real axis of the s-plane is occupied by root loci. Complex poles and zeros of G(s)H(s) do not affect the type of root loci found on the real axis. F-6. because real poles and zeros that lie to the left of the point contribute nothing. The entire real axis of the s-plane is occupied by the root loci for 1. K 0: On a given section of the real axis. Each real pole of G(s)H(s) to the right of s1 contributes 180 degrees. At any point s1 on the real axis. only the real zeros and poles of G(s)H(s) contribute to the angular relations in Eqs. . K 0: On a given section of the real axis. These properties are arrived at based on the following observations: 1. the angles of the vectors drawn from the complexconjugate poles and zeros of G(s)H(s) add up to zero. K . root loci are found in the section only if the total number of poles and zeros of G(s)H(s) to the right of the section is odd. (8-18) and (8-19). (8-18) and (8-19). Notice that the entire real axis is occupied by the root loci for all values of K. For s1 to be a point on the branch of the root loci for K 0. and each real zero of G(s)H(s) to the right of s1 contributes 180 degrees to Eqs. Only the real poles and zeros of G(s)H(s) that lie to the right of the point s1 contribute to Eqs. 3. root loci are found in the section only if the total number of real poles and zeros of G(s)H(s) to the right of the section is even. EXAMPLE F-7 The root loci on the real axis for two pole-zero configurations of G(s)H(s) are shown in Fig. Thus. The following example illustrates the determination of the properties of the root loci on the real axis of the s-plane. 2. (8-18) and (8-19).

F-7. s1 must satisfy Eq. The details are illustrated by the following example. the angles of the vectors drawn from the other three poles are approximated by considering that s1 is at 1 j.6 K – u2 K=0 s1 j1. (8-18). F-7.16 K=0 K – s – K K=0 –3 K K K>0 – K K<0 Figure F-7 Root loci of s(s departure or arrival. respectively. EXAMPLE F-8 For the root-locus diagram shown in Fig.F-7 Angles of Departure and Angles of Arrival of the Root Loci F-9 F-7 ANGLES OF DEPARTURE AND ANGLES OF ARRIVAL OF THE ROOT LOCI The angle of departure or arrival of a root locus at a pole or zero. of G(s)H(s) denotes the angle of the tangent to the locus near the point. As shown in Fig. Since s1 is assumed to be very close to the pole at 1 j. the root locus near the pole s 1 j may be more accurately sketched by knowing the angle at which the root locus leaves the pole. (8-19) for root loci for negative K.095 Kc = 8. measured with respect to the real axis. Then. From j s-plane K K ~ u4 – 26.095 Kc = 8. (8-18) for root loci for positive K and Eq. Let us assign s1 to be a point on the RL leaving the pole at 1 j and is very close to the pole. The angles of departure and arrival are determined using Eq. 3)(s 2 2s 2) K 0 to illustrate the angles of . Thus. G1s1 2H1s1 2 1u1 u2 u3 u4 2 12i 12180° (F-16) where i is any integer.16 – u1 ~ 135 K=0 –5 4 ~ u 3 – 90 0 – j1. the angle of departure of the root locus at s 1 j is represented by 2.

Similarly.F-10 Appendix F Properties and Construction of the Root Loci jv K>0 K=± –3 u3 = 0° K>0 K<0 –2 K<0 s-plane u1 = 180° K=0 0 s s1 60° K > 0 Three vectors. The result is u1 3u2 u3 12i 12180° (F-18) where 1 and 3 denote the angles of the vectors drawn from the pole at 0 and the zero at 3. Eq. we set i 1 and i 2 successively in Eq. EXAMPLE F-9 In this example we examine the angles of departure and arrival of the root loci at a multiple-order pole or zero of G(s)H(s). Similarly. since there are three poles at s 2. we can show that the root locus for negative K arrives at the pole s 3 with an angle of 180 . there are three positive-K loci leaving and three negativeK loci arriving at the point. and Eq.6 . which is the angle of departure of the positive-K root loci that lies between 3 s 0 and s 2. which is 180 71. and apply Eq. and the result for 2 is 71. one from each pole K<0 at s = –2. and 60 . F-8. F-8. 180 . F-7. to s1.4 . we can set i to be 1. as shown in Fig. . the angle of arrival of the negative-K root locus is 180 . When the angle of departure or arrival of a root locus for positive K at a simple pole or zero of G(s)H(s) is determined. Figure F-7 shows that the angle of arrival of the root locus for negative K at 1 j is 108. whereas the angle of departure of the positive-K root locus is 180 .6 . the angles of arrival and departure of root loci on the real axis are not affected by the complex poles and zeros of G(s)H(s). Since the total angles of the vectors drawn from complex-conjugate poles and zeros to any point on the real axis add up to be zero. for the three negative-K root loci that arrive at s 2. Eq. Angle = u2 Figure F-8 Angles of departure and arrival at a third-order pole. (F-18). for the root-locus diagram in Fig. (F-16) is written 1135° u2 90° 26. For the angles of departure of the other two positive-K loci. and the angles of arrivals are found to be 60 . For the pole at s 0. 0 we have 2 0 . Fig. and we have 2 120 and 120 . we assign a point s1 on one of the loci near s 2. These angles are also determined from the knowledge of the type of root loci on sections of the real axis separated by the poles and zeros of G(s)H(s). and since 1 180 . so that there are three vectors drawn from 2 to s1. (8-19) is now used. respectively. Setting i to zero in Eq. The angle 2 is multiplied by 3. For the third-order pole at s 2. the angle of arrival or departure of the root locus for negative K at the same point differs from this angle by 180 . and the root locus for positive K leaves the same pole at 0 . In this case.6°2 12i 12180° (F-17) where 2 is the only unknown angle. Only the real poles and zeros of G(s)H(s) are shown. (F-18). since the complex ones do not affect the type or the angles of arrival and departure of the root loci on the real axis. (8-19) is used. (8-18). To find the angles of departure of the positive-K root loci. Consider that a G(s)H(s) has a multiple-order (third-order) pole on the real axis.

For complex situations. EXAMPLE F-10 The root loci shown in Fig. (F-19).095. Figure F-9(a) illustrates a case in which two branches of the root loci meet at the breakaway point on the real axis and then depart from the axis in opposite directions. we have the critical value of K for stability at K 8. F-7 is for the equation s1s 321s2 2s 22 K 0 (F-19) Figure F-7 shows that the root loci intersect the j axis at two points. The points where the root loci intersect the imaginary axis of the s-plane. can also be used for this purpose. .F-9 Breakaway Points F-11 F-8 INTERSECTION OF THE ROOT LOCI WITH THE IMAGINARY AXIS • Routh-Hurwitz criterion may be used to find the intersection of the root loci on the imaginary axis. the intersects and the critical values of K can be determined with the help of the root-locus computer program. and by solving the auxiliary equation. and the corresponding crossover points on the j -axis are at j1. when the root loci have multiple number of intersections on the imaginary axis. F-9 BREAKAWAY POINTS F-9-1 (Saddle Points) on the Root Loci Breakaway points on the root loci of an equation correspond to multiple-order roots of the equation. associated with the frequency response. may be determined by means of the Routh-Hurwitz criterion. and the corresponding values of K. The Bode diagram method in Chapter 9. In s-plane K=0 K=0 Breakaway point s-plane K= K= Breakaway point (a) jv K K (b) K=0 j s-plane K=0 –2 –1 K=0 0 Breakaway point –j K s K=0 K (c) Figure F-9 Examples of breakaway points on the real axis in the s-plane. Applying the Routh-Hurwitz criterion to Eq.16.

of course. the four root loci arrive and depart with angles 90 apart. (F-20) are breakaway points. that is. this case. the solution of Eq. The complex-conjugate solutions of Eq. all breakaway points on the root loci for all values of K must satisfy Eq. but not all solutions of Eq. In general. F-12 for an example of root loci with complex breakaway points. whereas in Fig. the root loci shown in Figs. F-9(c). All real solutions of Eq. In general. Figure F-9(b) shows another common situation when two complex-conjugate root loci approach the real axis. whereas in Fig. 2. Moreover. we have dK ds (F-22) Thus. A root-locus diagram can have. (F-20): 1. the four root loci arrive and depart with angles 90 apart. since the entire real axis of the s-plane is occupied by the root loci. meet at the breakaway point. For example. Since the condition of the root loci is K 1 G1 1s2H1 1s2 dG1 1s2H1 1s2 ds 3G1 1s2H1 1s2 4 2 0 (F-21) taking the derivative on both sides of the equation with respect to s. (F-23) F-9-2 The Angle of Arrival and Departure of Root Loci at the Breakaway Point The angles at which the root loci arrive or depart from a breakaway point depend on the number of loci that are involved at the point. the following conclusions may be made with regard to the solutions of Eq. The properties of the breakaway points of root loci are given as follows: The breakaway points on the root loci of 1 KG1(s)H1(s) 0 must satisfy dG1 1s2H1 1s2 ds 0 (F-20) It is important to point out that the condition for the breakaway point given in Eq. 3. (F-21). (F-20) are breakaway points on the root loci for all values of K. F-9(a) and F-9(b) all arrive and break away at 180 apart. Because of the conjugate symmetry of the root loci. (F-20) is necessary but not sufficient. (F-20) must also satisfy the equation 1 KG1(s)H1(s) 0. the breakaway point represents a double root of the equation when the value of K is assigned the value corresponding to the point.F-12 Appendix F Properties and Construction of the Root Loci • A root-locus plot may have more than one breakaway points. more than one breakaway point. • Breakaway points may be complex conjugates in the s-plane. the breakaway point condition can also be written as dK ds where K is expressed as in Eq. and then depart in opposite directions along the real axis. n root loci ( K ) arrive or depart a breakaway point at 180 n degrees apart. the breakaway points need not always be on the real axis. a breakaway point may involve more than two root loci. To be a breakaway point. In other words. must also be a point on the root loci for some real K. In general. Figure F-9(c) illustrates a situation when the breakaway point represents a fourth-order root. F-9(c). (F-20). . (F-20) are breakaway points only if they satisfy the characteristic equation or are points on the root loci. the breakaway points not on the real axis must be in complexconjugate pairs. Refer to Fig.

EXAMPLE F-12 Consider the equation s2 2s 2 K1s 22 0 (F-28) The equivalent G(s)H(s) is obtained by dividing both sides of Eq. one between 0 and 2 and the other between 4 and . (F-28) by the terms that do not contain K. we find the two breakaway points of the root loci at s 1. we have G1 1s2H1 1s2 dG1 1s2H1 1s2 ds s s1s 4 22 (F-25) Applying Eq.828 K=± –4 K<0 K=0 –2 –1. the root loci of Eq. the breakaway points on the root loci must satisfy s1s 22 s2 1s 8 21s 121s 22 2 42 0 (F-26) or s2 8s 0 (F-27) 6. From Eq. which a rather tedious task to do manually.172 K=0 K<0 K 0 – K>0 K>0 K<0 Figure F-10 Root loci of s(s 2) K(s 4) 0. F-10 for K . (F-20). It can be proven that the complex portion of the root loci is a circle. We have G1s2H1s2 K1s s2 2s 22 2 (F-29) .172 and Figure F-10 shows that the two breakaway points are all on the root loci for positive K. (F-24) are sketched as shown in Fig.828. Solving Eq. (F-27). K>0 jv s-plane K –6. (F-24).F-9 Breakaway Points F-13 Many root-locus computer programs have features that will obtain the breakaway points. The two breakaway points are on the real axis. EXAMPLE F-11 Consider the second-order equation s1s 22 K1s 42 0 (F-24) Based on some of the properties of the root loci described thus far.

one for K 0 and one for K 0. the Since the poles of G1(s)H1(s) are symmetrical about the axes root loci of the equation are also symmetrical with respect to these two axes.414. Fig. we get dG1 1s2H1 1s2 ds 4s3 3s1s 6s2 24s2 421s2 72s 80 4s 202 4 2 20 0 0 (F-34) or s3 18s (F-35) The solutions of the last equation are s 2. F-12 shows that all the solutions of Eq. The plot shows that there are two breakaway points. These breakaway points are determined from dG1 1s2H1 1s2 ds or s2 4s 2 0 0. (F-29) are plotted as shown in Fig.586 K=0 –1 – j K – K>0 K>0 K<0 Figure F-11 Root loci of s2 2s 2 K(s 2) 0. and two of these points are complex.586 and s 3.414 K=± –2 K<0 0 – 0. F-11. 2 j2. (F-35) are breakaway points on the root loci. In this case.45.F-14 Appendix F Properties and Construction of the Root Loci jv K>0 K=0 –1 + j s-plane K –3. Based on the poles and zeros of G(s)H(s).45. and 2 j2. we have 1 KG1 1s2H1 1s2 1 K 421s2 4s 0 (F-33) s1s 202 2 and 0 in the s-plane. Taking the derivative of G1(s)H1(s) with respect to s. the root loci of Eq. (F-31) s 2 d a b ds s2 2s 2 s2 2s 1s2 2 21s 2s 121s 22 2 22 0 (F-30) The solution of this equation gives the breakaway point as s EXAMPLE F-13 Figure F-12 shows the root loci of the equation s1s 421s2 4s 202 K 0 (F-32) Dividing both sides of the last equation by the terms that do not contain K. .

667 j0.471. The root loci show that neither the K breakaway point in this case.45 Breakaway point –2 – j4 K=0 K K – K>0 K K<0 Figure F-12 Root loci of s(s 4)(s2 4s 20) K 0. F-13. EXAMPLE F-14 In this example.F-9 Breakaway Points jv K – s-plane K K F-15 – 2 + j4 Breakaway point K=0 –2 + j2.471 and 0. we have the equation for the breakaway points: 3s2 4s 2 0 (F-38) The roots of Eq. writing Eq.667 j0. since they do not satisfy Eq. (F-38) are s 0. (F-36) as 1 KG1 1s2H1 1s2 1 K 2s s1s2 22 0 (F-37) and applying Eq. The root loci of the equation s1s2 2s 22 K 0 0 loci nor the K (F-36) 0 loci has any are shown in Fig. . we shall show that not all the solutions of Eq. (F-20) are breakaway points on the root loci.45 458 K=0 –4 Breakaway point –2 K=0 0 – K K – s –2 – j2. These two roots are not breakaway points on the root loci. (F-20). However. (F-36) for any real value of K.

In general. and C is the length of the vector drawn from the zero of G(s)H(s) to s1. The value of K at the point s1 is given by K A C B (F-41) where A and B are the lengths of the vectors drawn from the poles of G(s)H(s) K(s 2) (s2 2s 2) to the point s1. F-10 CALCULATION OF K ON THE ROOT LOCI Once the root loci are constructed. s1 is on the locus where K is positive. . Graphically. Figure F-14 shows that the value of K at s 0 is 1. the value of K at the point where the root loci intersect the imaginary axis can also be found by the method just described. (8-20). In this case. The computer method and the Routh-Hurwitz criterion are other convenient alternatives of finding the critical value of K for stability.F-16 Appendix F Properties and Construction of the Root Loci – jv s-plane K K K=0 –1 + j K K=0 0 K – s K=0 –1 – j K K – K>0 K<0 Figure F-13 Root loci of s(s2 2s 2) K 0. F-14. the values of K at any point s1 on the loci can be determined by use of the defining equation of Eq. the root loci of the equation s2 2s 2 K1s 22 0 (F-40) are shown in Fig. the magnitude of K can be written as 0K 0 ß lengths of vectors drawn from the poles of G1 1s2H1 1s2 to s1 ß lengths of vectors drawn from the zeros of G1 1s2H1 1s2 to s1 (F-39) EXAMPLE F-15 As an illustration on the determination of the value of K on the root loci.

the properties on the root loci presented here should be adequate for making a reasonably accurate sketch of the root-locus diagram short of plotting it point by point. TABLE F-1 1. However. including the plotting of the final loci. The K points are at the zeros of G(s)H(s). 1. except for extremely complex cases. since the user still has to decide on the range and resolution of K so that the root-locus plot has a reasonable appearance. 4. . one cannot rely on the computer solution completely. The computer program can be used to solve for the exact root locations. 5. 2. the root loci are asymptotic to ui 180° where i n m 0. 3. The root loci are symmetrical about the axes of symmetry of the pole-zero configuration of G(s)H(s). including those at s .F-10 Calculation of K on the Root Loci jv K=0 –1 + j F-17 s-plane s1 C K A K= –2 B K 0 K = –1 – s K=0 –1 – j K>0 K<0 Figure F-14 Graphical method of finding the values of K on the real axis. K K Properties of the Root Loci of F (s) 0 points points 1 KG1(s)H1(s) 0 Number of separate root loci Symmetry of root loci Asymptotes of root loci as s S q The K 0 points are at the poles of G(s)H(s). and some of the other specific details of the root loci. For quick reference. the breakaway points. …. and number of finite zeros of G(s)H(s). including those at s . 0n m 0 1. The total number of root loci is equal to the order of the equation F(s) 0. For large values of s. 2. number of finite poles of G(s)H(s). In summary. the important properties described are summarized in Table F-1. the root loci for K 0 are asymptotic to asymptotes with angles given by ui For K 2i 0n 0n 2i 1 m0 m0 180° 0.

12180° K 7 0 K 6 0 2i 180° 1. Angles of departure Root loci for K 0 are found in a section of the real axis only if the total number of real poles and zeros of G(s)H(s) to the right of the section is odd. and applying the equation G1s1 2H1s1 2 m a k 1 1s1 zk 2 n a j 1 1s1 pj 2 21i where i 9. The angle of departure or arrival of the root loci from a pole or a zero of G(s)H(s) can be determined by assuming a point s1 that is very close to the pole. EXAMPLE F-16 Consider the equation s1s 521s 621s2 2s 22 K1s 32 0 (F-42) Dividing both sides of the last equation by the terms that do not contain K. There are five separate branches on the root loci. The breakaway points on the root loci are determined by finding the roots of dK ds 0.F-18 Appendix F Properties and Construction of the Root Loci TABLE F-1 6. 2. If the total number of real poles and zeros to the right of a given section is even. 10. These are necessary conditions only. real parts of poles of G1s2H1s2 Root loci on the real axis 8. 6. 3. or zero. . and 1 j. 1 j. (b) The point of intersection of the asymptotes is given by real parts of zeros of G1s2H1s2 m n Intersection of the asymptotes s1 7. or dG(s)H(s) ds 0. . …. step by step. Intersection of the root loci 0. 3. The root loci are symmetrical with respect to the real axis of the s-plane. The absolute value of K at any point s1 on the root loci is on the root loci determined from the equation 0K 0 1 0G1 1s1 2H1 1s1 2 0 The following example illustrates the construction of a root locus diagram manually. . 2. root loci for K 0 are found. we have G1s2H1s2 K1s s1s 521s 32 2s 22 621s2 (F-43) The following properties of the root loci are determined: 1. The K The K 0 points are at the poles of G(s)H(s): s points are at the zeros of G(s)H(s): s 5. 4. . Calculation of the values of K The crossing points of the root loci on the imaginary axis and with the imaginary axis the corresponding values of K may be found by use of the RouthHurwitz criterion. (continued) (a) The intersection of the asymptotes lies only on the real axis in the s-plane. Breakaway points 11. using the root locus properties given in Table F-1.

and 270 . the four root loci that approach infinity as K approaches infinity should approach asymptotes with angles of 45 . (F-10): ui 0n 2i m0 180° 05 2i 10 180° 6 K 0 (F-45) for i 0. 1. and 135 . 180 . 1. as K approaches . The angles of the asymptotes of the RL are given by [Eq. (F-9)] ui 2i 0n 1 180° m0 2i 05 1 180° 10 0 K 6 (F-44) for i 0. 2. four RL and CRL should approach infinity along the asymptotes.5 –1 0 K=0 –j K=0 K – s K K K K>0 – K<0 Figure F-15 Preliminary calculation of the root loci of s(s 5)(s 6)(s2 2s 2) K(s 3) 0. 3. Thus. Thus. It should be pointed out that in general the properties of the asymptotes do not indicate on which side of the asymptotes jv K s-plane K K – j K=0 – K K=0 –6 K=0 –5 K= –4 –3 45 –2 –2. . 90 . F-19 Since G(s)H(s) has five poles and one finite zero. 135 . respectively. F-15.5 0 should approach (F-46) The results from these six steps are illustrated in Fig.F-10 Calculation of K on the Root Loci 5. 6. 45 . four root loci for K infinity along asymptotes with angles of 0 . The angles of the asymptotes of the CRL at infinity are given by Eq. The intersection of the asymptotes is given by [Eq. 3. (F-12)] s1 1 5 6 1 4 12 1 32 2. 2.

5)(s 6)(s 2 2s 2) K(s 3) 0 the root loci lie.4° 2. There are K 0 root loci on the remaining portions of the real axis.8° s-plane jv u K=0 s1 j K=0 0 j s1 s K=0 –6 –5 K=0 (s1 + 5) –3 K= (s1 + 3) –1 K=0 (s1 + 6) (s1 + 1 + j) Figure F-17 Computation of angle of departure of the root loci of s(s 5)(s 6)(s2 2s 2) K(s 3) 0. 1. 7. that is. In fact. . Root loci on the real axis: There are K 0 root loci on the real axis between s 0 and 3. The segments of the root loci shown in Fig. (8-18). or 26. the root locus can even cross an asymptote in the finite s domain. F-16. and s 5 and 6. … Therefore.6° for i 0.F-20 Appendix F Properties and Construction of the Root Loci jv K=0 j – K K=0 –6 K=0 –5 K= –3 –1 K=0 K=0 0 s-plane K – s –j K>0 K<0 Figure F-16 Root loci of s(s on the real axis. and s 6 and . Angles of departure: The angle of departure of the root loci leaving the pole at 1 j is determined using Eq. 135° 90° u 12i 12180° (F-48) 2. selecting i u 43. If s1 is a point on the root loci leaving the pole at 1 j. Eq. and s1 is very close to the pole. between s 3 and 5. (F-49) 1s1 52 1s1 1 j2 (F-47) 8. F-15 can be accurately plotted only if additional information is obtained. as shown in Fig. The asymptotes indicate nothing more than the behavior of the root loci as s S . (8-18) gives 1s1 32 s1 1s1 12i 1 j2 12180° 14° 11. as shown in Fig F-17.

(F-51) to have no roots on the imaginary axis or in the right-half of the s-plane.6 0. which are the points at which Breakaway points: Based on the information gathered from the preceding nine steps.6 105K 0.33 j1.163K2 65. only one root of the last equation is the correct solution of the breakaway point. The five roots of Eq.6 0. thus.2° (F-50) The intersection of the root loci on the imaginary axis is determined using Routh’s tabulation. Equation (F-42) is written s5 Routh’s tabulation is s5 s4 s3 s2 s1 s0 1 13 47. (F-43) with respect to s and set it to zero.5s4 66s3 142s2 123s 45 0 (F-57) Since there is only one breakaway expected.656 j0.8° 136.33 j1.769K 3K 0 0 60 K 3K 0 0 0 0 13s4 54s3 82s2 160 K2s 3K 0 (F-51) For Eq. all the roots of Eq.163K2 7 0 K 7 0 or or K 6 309 K 6 35 (F-52) (F-53) (F-54) Thus.468 . (F 55). 10. u¿ 9.7 65. we take the derivative on both sides of Eq. a trial sketch of the root loci indicates that there can be only one breakaway point on the entire root loci. (F-56) are s the root loci cross the j -axis.F-10 Calculation of K on the Root Loci F-21 Similarly. (F-51) cross the imaginary axis when K 35 and K 0.34 and j1. Thus.468 5. the following inequalities must be satisfied: 3940 65.656 j0. (F-51) will stay in the left-half s-plane if K lies between 0 and 35.6 0.204 0. the elements in the first column of Routh’s tabulation must all be of the same sign.2s2 The roots of Eq. Eq. It is easy to see that differs from by 180 .34.204 0. and the point should lie between the two poles of G(s)H(s) at s 5 and 6.53 s s 3.212K 3940 105K 0. (8-19) is used to determine the angle of arrival of the K 0 root loci arriving at the pole 1 j.212K 3K 54 82 0. the resulting equation is s5 13. To find the breakaway point. 105 0 (F-56) j1. we get 58. Substituting K 35 in Eq. (F-57) are: s s s 3.212K 7 0 0. The coordinates at the crossover points on the imaginary axis that correspond to K 35 are determined from the auxiliary equation: A1s2 165. which means that the root loci of Eq.212K2s2 3K 0 (F-55) which is obtained by using the coefficients from the row just above the row of zeros in the s1 row that would have happened when K is set to 35. 180° 43.

F-18 K K K K – .53.5 K>0 –1 K=0 –j –j1.F-22 Appendix F Properties and Construction of the Root Loci jv s-plane j1. The other four solutions do not satisfy Eq. (F-51) and are not breakaway points. the root loci of Eq.34(K = 35) K=0 0 K – s K K>0 – K K<0 Figure F-18 Root loci of s(s 5)(s 6)(s2 2s 2) K(s 3) 0. Based on the information obtained in the last 10 steps. Clearly.53 K<0 K=± –4 –3 –2.34(K = 35) j K=0 _ K K=0 –6 K=0 –5 –5. the breakaway point is at 5. (F-51) are sketched as shown in Fig.

.Appendix G Frequency-Domain Plots TO ACCOMPANY AUTOMATIC CONTROL SYSTEMS EIGHTH EDITION BY BENJAMIN C. INC. KUO FARID GOLNARAGHI JOHN WILEY & SONS.

photocopying. E-Mail: PERMREQ@WILEY. Requests to the Publisher for permission should be addressed to the Permissions Department. scanning or otherwise. Danvers. Inc. recording.COM. fax (508)750-4470. All rights reserved. except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act. ISBN 0-471-13476-7 . MA 01923. mechanical.Copyright © 2003 John Wiley & Sons. New York. Inc. No part of this publication may be reproduced. NY 10158-0012. (508)750-8400. fax (212) 850-6008. 222 Rosewood Drive. (212) 850-6011. without either the prior written permission of the Publisher. stored in a retrieval system or transmitted in any form or by any means. 605 Third Avenue. electronic.. or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center. To order books or for customer service please call 1-800-CALL WILEY (225-5945). John Wiley & Sons.

A simple example G-1 . 2. and can be written as G1 jv2 where 0G( j )0 denotes the magnitude of G( j ). Bode plot. with as a variable parameter on the curve. The frequency-domain analysis of the closed-loop system can be conducted from the frequency-domain plots of G(s) with s replaced by j . and G( j ) is the phase of G( j ). is a plot of the magnitude of G( j ) versus the phase of G( j ) on polar coordinates as is varied from zero to infinity. The following frequency-domain plots of G( j ) versus are often used in the analysis and design of linear control systems in the frequency domain. In practice. From an analytical standpoint.APPENDIX G Frequency-Domain Plots Let G(s) be the forward-path transfer function of a linear control system with unity feedback. A plot of the magnitude versus phase in the polar coordinates as is varied from zero to infinity. A plot of the magnitude (in decibels) versus the phase on rectangular coordinates. 3. G(s). especially if the function is of high order. 0G1 jv2 0 G1 jv2 (G-1) G-1 COMPUTER-AIDED CONSTRUCTION OF THE FREQUENCY-DOMAIN PLOTS The data for plotting of the frequency-domain plots are usually quite time consuming to generate if the computation is to be carried out manually. The ACSYS computer program can be used for this purpose. Polar plot. Many software packages that are available commercially can be used for the construction of the frequency-domain plots. the analyst and designer should be familiar with the properties of the frequency-domain plots so that proper interpretations can be made on these plots made by the computer. From a mathematical viewpoint. the process can be regarded as the mapping of the positive half of the imaginary axis of the s-plane onto the G( j )-plane. G-2 POLAR PLOTS The polar plot of a function of the complex variable s. Magnitude-phase plot. a digital computer should be used to do the computation as well as plotting of the graph. 1. A plot of the magnitude in decibels versus (or log10 ) in semilog (or rectangular) coordinates. The function G( j ) is generally a complex function of the frequency .

(G-3) is written G1 jv2 1 v2T 2 tan 1 0 v vT (G-4) When is zero. at 0. of this mapping is shown in Fig. counterclockwise is referred to as positive. For any frequency 1. As increases. G-1. Eq. we have G1 jv2 21 1 1 jvT (G-3) In terms of magnitude and phase. consider the function G1s2 where T is a positive constant. G( j ) is represented by a vector of unit length directed in the 0 direction. and clockwise is negative. In measuring the phase. the magnitude and phase of G( j 1) are represented by a vector in the G( j )-plane. j Im G G ( jv)-plane v 0 tan–1vT Phasor of G( jv) v=0 1 Re G 1 1 + v2T 2 Figure G-2 Polar plot of 1 G1 jv2 . the magnitude of G( j ) is unity.G-2 Appendix G Frequency-Domain Plots jv s-plane jv3 jv2 jv1 0 s v1 v2 |G( jv1)| v3 0 G( jv1) Re G G( jv)-plane j Im G Figure G-1 Polar plot shown as a mapping of the positive half of the j -axis in the s-plane onto the G( j )-plane. and the phase of G( j ) is at 0 . 11 jvT 2 . Thus. Setting s 1 1 Ts (G-2) j . EXAMPLE G-1 To illustrate the construction of the polar plot of a function G(s).

If T2 is greater than T1. such as the Nyquist stability criterion. j Im G G( jv)-plane v (T2 > T1) 0 T2/T1 v= v (T2 < T1) 1 v=0 v= T2/T1 Re G Figure G-3 Polar plots of G1 jv2 11 11 jvT1 2 jvT2 2 . an exact plot of the frequency response is not essential. EXAMPLE G-2 As a second illustrative example. and the vector rotates in the clockwise (negative) direction. Consider the following transfer function: G1s2 10 s1s 12 (G-7) . When approaches infinity. as shown in Fig. This is presented by a vector with an infinitesimally small length directed along the 90 -axis in the G( j )-plane. In many control-system applications. Equation (G-5) is written G1 jv2 1 v2T22 v2T12 1tan 1 vT2 tan 1 vT1 2 (G-6) The polar plot of G( j ). 2. and the phase is always negative. often we have to determine the basic properties of a polar plot.G-2 Polar Plots G-3 the magnitude of G( j ) decreases. the exact plot of G( j ) turns out to be a semicircle. (G-6) that correspond to these two conditions are shown in Fig. and the phase becomes more negative. the magnitude of G( j ) becomes zero. the length of the vector in the polar coordinates decreases. and the phase of G( j ) is always positive. The general shape of the polar plot of a function G( j ) can be determined from the following information. Often. a rough sketch of the polar plot of the transfer function G(j )H( j ) is adequate for stability analysis in the frequency domain. The polar plots of G( j ) of Eq. the magnitude of G( j ) is always less than unity. and the phase reaches 90 . and the values of at these intersections EXAMPLE G-3 In frequency-domain analyses of control systems. G-2. the magnitude of G( j ) is always greater than unity as is varied from zero to infinity. consider the function B1 G1 jv2 1 1 jvT2 jvT1 (G-5) where T1 and T2 are positive real constants. 1. As increases. in this case. If T2 is less than T1. G-3. depends on the relative magnitudes of T1 and T2. (G-4). By substituting other finite values of into Eq. The behavior of the magnitude and phase of G( j ) at 0 and The intersections of the polar plot with the real and imaginary axes.

the imaginary part of G( j ) is zero. From Eq. meaning that the G( j ) plot intersects only with the real axis of the G( j )-plane at the origin. Therefore. we are interested in the properties of the G( j ) at infinity. If the polar plot of G( j ) intersects the real axis. Under certain conditions. G-4. the properties of the polar plot of G( j ) at 0 and are ascertained. Im3G1 jv2 4 0 (G-12) To express G( j ) as the sum of its real and imaginary parts. at the point of intersection. the magnitude and phase of G( j ) at computed as follows: vS0 vS0 vS vS 0 and are lim 0 G1 jv2 0 lim G1 jv2 lim 0 G1 jv2 0 lim G1 jv2 10 v lim 10 jv vS0 10 lim 0 vS v2 lim 10 1 jv2 2 vS0 lim (G-8) 90° (G-9) (G-10) 180° (G-11) vS Thus. G( jv)-plane j Im G v= –10 0 Re G v Figure G-4 Polar plot of G1s2 0 10 s1s 12 . we determine the intersections. the polar plot of G( j ) is easily sketched without actual plotting. (G-13) to zero. Similarly. that is. we must rationalize G( j ) by multiplying its numerator and denominator by the complex conjugate of its denominator. which corresponds to 0 in this case. of the polar plot with the two axes of the G( j )-plane. if any. Based on this information. G( j ) is written G1 jv2 101 jv21 jv 12 jv1 jv 121 jv21 jv 12 Re3 G1 jv2 4 j Im3G1 jv2 4 10v2 v v2 4 j 10v v4 v2 (G-13) When we set Im[G( j )] to zero. The only real solution for is also . which corresponds to the origin of the G( j )-plane. (G-13). we get . as well as knowledge of the angles of G( j ) at 0 and . The conclusion is that the polar plot of G( j ) does not intersect any one of the axes at any finite nonzero frequency. (G-7). the intersection of G( j ) with the imaginary axis is found by setting Re[G( j )] of Eq. . Next. as shown in Fig.G-4 Appendix G Frequency-Domain Plots By substituting s j in Eq. we see that Im[G( j )] and Re[G( j )] 10 at 0.

we have v 22 rad sec. the last equation is written G1 jv2 Re3G1 jv2 4 j Im3G1 jv2 4 9v 2 22 rad sec. G1s2 s1s 121s 22 0 . Setting Im[G( j )] to zero. if G(s) is a rational function of s (a quotient of two polynomials of s). In general. which means that the G( j ) plot intersects the imaginary axis only at the origin. as shown in Fig. From Eq. This gives the point of intersection on the real axis at 30 12 v2 2 2 j1012 9v 3 v12 v2 2 2 (G-20) j Im G G-plane v= 0 –5 3 v = 2 rad/sec Re G v Figure G-5 Polar plot of 10 . the polar plot of G( j ) for negative values of is the mirror image of that for positive . it is now possible to make a sketch of the polar plot for the transfer function in Eq. (G-14). it The result. has no physical meaning. G1 j222 5 3 (G-21) Setting Re[G( j )] to zero. we have . and G( j ) 0. we also see that Re[G( j0)] and Im[G( j0)] . (G-20). G-5.G-2 Polar Plots G-5 EXAMPLE G-4 Given the transfer function G1s2 s1s 10 121s 22 (G-14) we want to make a rough sketch of the polar plot of G( j ). with the mirror placed on the real axis of the G( j )-plane. v simply represents a mapping point on the negative j -axis of the s-plane. because the frequency is negative. With this information. The following calculations are made for the properties of the magnitude and phase of G( j ) at 0 and : vS0 vS0 lim 0 G1 jv2 0 lim 0 G1 jv2 0 lim G1 jv2 vS0 vS0 lim 5 v q (G-15) (G-16) (G-17) (G-18) lim G1 jv2 lim vSq vSq 5 jv 90° 10 lim 0 vSq v3 lim 10 1 jv2 3 270° vSq To find the intersections of the G( j ) plot on the real and imaginary axes of the G( j )-plane. we rationalize G( j ) to give G1 jv2 101 jv21 jv jv1 jv 121 jv 121 jv 22 121 jv 22 v2 2 221 jv21 jv (G-19) After simplification.

The phase relations are also added and subtracted from each other algebraically. We shall show in the next section that approximate information on the polar plot can always be obtained from the Bode plot. 2. which can be sketched without any calculations. since the poles and zeros of G(s) are easily identified. one with the amplitude of G( j ) in decibels (dB) versus log10 or . Since practically all the terms in Eq. Consider the function: K1s z1 21s z2 2 p 1s zm 2 T s (G-22) G1s2 e d s j 1s p1 21s p2 2 p 1s pn 2 where K and Td are real constants. In simple terms. . G(s) is preferably written in the following form: K1 11 T1s211 T2s2 p 11 Tms2 T s G1s2 e d (G-23) s j 11 Tas211 Tb s2 p 11 Tns2 where K1 is a real constant. respectively. and n are real constants. (G-23) are of the same form. These names stem from the fact that the Bode plot can be constructed by using straight-line approximations that are asymptotic to the actual plot. (G-22) or Eq. sketches of the polar plots can be obtained with the help of the Bode plots. (G-22) is the preferred form for root-locus construction. such as the polar plot and the magnitude-versus-phase plot. can be easily generated from the Bode plot. G1s2 K11 s11 Tas211 T1s211 T2s2 2zs vn s2 v2 2 n e Td s 1. in general. the Bode plot has the following features: Since the magnitude of G( j ) in the Bode plot is expressed in dB. G-3 BODE PLOT (CORNER PLOT OR ASYMPTOTIC PLOT) • A Bode plot is also known as a corner plot or an asymptotic plot. and Td is the real time delay. Since the straight-line approximation of the Bode plot is relatively easy to construct. unless a digital computer is used. then either forms of Eq. which allow the simple sketching of the plot without detailed computation. (G-24) where K. the Ts may be real or complex (in conjugate pairs) numbers. A Bode plot is also known as a corner plot or an asymptotic plot of G( j ). The Bode plot of the function G( j ) is composed of two plots. Eq. product and division factors in G( j ) became additions and subtractions. T1. for more complicated transfer functions. Thus. It is assumed that the second-order polynomial in the denominator has complex-conjugate zeros. the polar plot is basically a tool for analysis. Ta. If the Bode plot is to be constructed with a computer program. . Furthermore. then without loss of generality.G-6 Appendix G Frequency-Domain Plots Although the method of obtaining the rough sketch of the polar plot of a transfer function as described is quite straightforward. In Chapter 8. and the zs and the ps may be real or complex (in conjugate pairs) numbers. and the other with the phase of G( j ) in degrees as a function of log10 or . For constructing the Bode plot manually. we can use the following transfer function to illustrate the construction of the Bode diagram. T2. The magnitude plot of the Bode plot of G( j ) can be approximated by straightline segments. (G-23) can be used. it is somewhat awkward for design purposes. the algebraic manipulation may again be quite involved. the data necessary for the other frequency-domain plots. for complicated transfer functions that may have multiple crossings on the real and imaginary axes of the transfer-function plane. Td.

that is. and r are positive integers. G-3-2 Poles and Zeros at the Origin. where Td. (G-24) and have many more factored terms. 5. the function G( j ) may be of higher order than that of Eq. the individual plots are then added or subtracted accordingly to yield the total magnitude in dB and the phase plot of G( j ). G-6 in semilog coordinates. Constant factor: K Poles or zeros at the origin of order p: ( j ) p Poles or zeros at s 1 T of order q: (1 j T ) q 2 r 2 Complex poles and zeros of order r: (1 j2 n n) j Td Pure time delay e . depending on whether or log10 is used as the abscissa. 3. The curves can be plotted on semilog graph paper or linear rectangular-coordinate graph paper. However. We shall now investigate sketching the Bode plot of different types of factors. d 1 20p log10 v2 d log10 v 20p dB/decade (G-30) . (G-25) and (G-26) indicate that additional terms in G( j ) would simply produce more similar terms in the magnitude and phase expressions. 2. q. G-3-1 Real Constant K Since KdB and K e 0° 180° K 7 0 K 6 0 (G-28) 20 log10 K constant (G-27) the Bode plot of the real constant K is shown in Fig.G-3 Bode Plot (Corner Plot or Asymptotic Plot) G-7 The magnitude of G( j ) in dB is obtained by multiplying the logarithm (base 10) of 0G( j )0 by 20. 4. G( j ) can contain just five simple types of factors: 1. (j ) p p The magnitude of ( j ) in dB is given by 20 log10 0 1 jv2 p 0 20p log10 v dB (G-29) for 0. p. (G-29) with respect to log10 . in general. so the basic method of construction of the Bode plot would be the same. The last expression for a given p represents a straight line in either semilog or rectangular coordinates. The slopes of these lines are determined by taking the derivative of Eq. We have also indicated that. we have 0G1 jv2 0 dB 20 log10 0G1 jv2 0 20 log10 0 jv 0 The phase of G( j ) is G1 jv2 K 11 11 jvT1 2 11 2 2zv vn v v2 2 n jvT2 2 jv vTd rad 11 jvTa 2 20 log10 0K 0 20 log10 01 jvT1 0 20 log10 01 20 log10 01 jvTa 0 20 log10 01 v2 v2 0 n jvT2 0 (G-25) j2zv (G-26) In general. Equations (G-25) and (G-26) verify one of the unique characteristics of the Bode plot in that each of the five types of factors listed can be considered as a separate plot. Eqs.

Thus. Instead of decades.1 1 10 100 20 log10 K dB (K = 10) v (rad/sec) 90 60 30 G( jv) (deg) 0 –30 –60 –90 –120 K (K < 0) –150 –180 0. the number of octaves between number of octaves log10 1v2 v1 2 log10 10 2 log10 a v2 b v1 (G-31) log10 1v2 v1 2 log10 2 and 1 is v2 1 log10 a b v1 0. The number of decades between any two frequencies 1 and 2 is given by number of decades Similarly. from 1 to 10.G-8 Appendix G Frequency-Domain Plots 40 30 20 | G( jv) | (dB) 10 0 –10 –20 –30 – 40 0. that is. (G-29) are said to be 20p dB decade of frequency.301 (G-32) . Furthermore. a unit change in log10 in the rectangular coordinates is equivalent to one decade of variation in . and so on. Thus. the slopes of the straight lines described by Eq. 10 to 100.1 1 10 100 K (K > 0) v (rad/sec) Figure G-6 Bode plot of constant K. a unit change in log10 corresponds to a change of 20p dB in magnitude. These lines pass through the 0-dB axis at 1. in the semilog coordinates. The frequencies 1 and 2 are separated by one octave if 2 1 2. sometimes octaves are used to represent the separation of two frequencies.

which has a simple pole at s 0.301 6p dB octave (G-34) For the function G(s) 1 s. and passes through the 0-dB axis at 1 rad sec. we have 20p dB decade 20p 0. G-7 for The magnitude and phase curves of the function ( j ) several values of P. 60 40 20 |G( jv) | (dB) 0 –20 d +4 B 0d /de e( cad jv) 2 B/d +20 d ecade ( jv) –20 – 40 –60 0.32 decades (G-33) Substituting Eq. (G-33) into Eq. the magnitude of G( j ) is a straight line with a slope of 20 dB decade. the relation between octaves and decades is number of octaves 1 0. .G-3 Bode Plot (Corner Plot or Asymptotic Plot) G-9 Thus.301 decades 3. The phase of ( j ) p is written 1 jv2 p p 90° p (G-35) are shown in Fig.1 1 v (rad/sec) 10 100 Figure G-7 Bode plots of ( j ) p.1 –6 B/dec 0d B/ de – 40 ca de (1/ jv ade(1 /jv) dB /de cad e(1 ) 3 /jv 2 ) 100 1 v (rad/sec) 10 180 ( jv)2 90 G( jv) (deg) ( jv) 0 (1/jv) –90 (1/jv)2 –180 (1/jv)3 –270 0. (G-30).

T 1. 3. It is useful to remember that the error is 3 dB at the corner frequency. we can approximate 1 Eq. a smooth curve can be sketched simply by locating the 3-dB point at the corner frequency and the 1-dB points at 1 octave above and below the corner frequency. then (G-39) Equation (G-38) represents a straight line with a slope of 20 dB/decade of frequency. As shown in Fig. Since the phase of G( j ) varies from 0 to 90 . Eq. the maximum deviation between the straight-line approximation and the actual curve is less than 6 .3 dB. which gives v 1 T (G-40) This frequency is also the intersect of high-frequency approximate plot and the lowfrequency approximate plot. and deviates only slightly from the straightline approximation. and 1 dB at 1 octave above ( 2 T) and 1 octave below ( 1 2T) the corner frequency. Draw the 20-dB decade (or 6-dB/octave) line and the horizontal line at 0 dB. the procedure of making of sketch of 01 jvT 0 dB is as follows: 1.G-10 Appendix G Frequency-Domain Plots G-3-3 Simple Zero. The actual 0G1 jv2 0 dB plot of Eq. as shown in Fig. The frequency given in Eq. Table G-1 gives the values of 11 jvT2 versus T. (G-40) is also known as the corner frequency of the Bode plot of Eq. 2. At very high frequencies. we can draw a line from 0 at 1 decade below the corner frequency to 90 at 1 decade above the corner frequency. Locate the corner frequency 1 T on the frequency axis. (G-37) becomes 0G1 jv2 0 dB 20 log10 vT T 2. a straight-line approximation can be made for the phase curve. with the two lines intersecting at 1 T. which is the 0-dB axis. (G-36). If necessary. the actual magnitude curve is obtained by adding the errors to the asymptotic plot at the strategic frequencies. (G-38) to Eq. (G-36) is a smooth curve. G-8. . T 1. The actual values and the straight-line approximation of 01 jvT 0 dB as functions of T are tabulated in Table G-1. At very low frequencies. The magnitude of G( j ) in dB is To obtain asymptotic approximations of 0G1 jv2 0 dB. the error is dropped to approximately 0. we consider both very large and very small values of . since the asymptotic plot forms the shape of a corner at this frequency. The intersect of these two lines is found by equating Eq. 1 j T is G1 jv2 tan 1 The phase of G( j ) vT (G-41) Similar to the magnitude curve. 1 j T Consider the function G1 jv2 0G1 jv2 0 dB 1 jvT 20 log10 21 (G-36) where T is a positive real constant. (G-39). At 1 decade above and below the corner frequency. Usually. Based on these facts. (G-37) is approximated by 0G1 jv2 0 dB 20 log10 2v2T 2 20 log10 1 0 dB 2 20 log10 0G1 jv2 0 v2T 2 (G-37) (G-38) T 2 by 2 since 2T 2 is neglected when compared with 1. The error between the actual magnitude curve and the straight-line asymptotes is symmetrical with respect to the corner frequency 1 T. G-8.

12 0 0.000043 0.3 1.00 10.4 T 0.3 0.76 1.6 37.12 1.00043 0.23 10.043 1 2 3 2 1 0.0 52.00 100.26 1.01 0.31 2.043 0.10 0.1 1 vT 10 100 TABLE G-1 Straight-Line Approximation 1 jVT dB 0 0 0 0 0 2.3 7.7 63.04 1.01 G(s) = 1 + Ts 1 1 + Ts Figure G-8 Bode plots of G1s2 1 Ts 1 and G1s2 .50 0.G-3 Bode Plot (Corner Plot or Asymptotic Plot) 40 G-11 20 |G( jv) | (dB) G(s) = 1 + Ts Asymptoes 0 G(s) = –20 1 1 + Ts – 40 0.4 45.005 .00 log10 T 2 1 0.0 40.4 84.0 1.00 1. 11 Ts2 0.0 20.65 2.4 1 jVT dB 0.00043 100.0 20.0 Error (dB) 0.043 40.117 0.7 26.0 2.1 1 vT 10 100 90 60 G( jv) (deg) 30 0 –30 G(s) = –60 –90 0.41 1.01 0.0 1 jVT 1.3 89.5 5.00043 (1 j T) (deg) 0.043 1 2 3 4.3 6.

The data in Table G-1 are still useful for the simple-pole case if appropriate sign changes are made to the numbers. (G-49). (G-42) are shown in Fig. yielding the corner frequency at n. the magnitude in dB of G( j ) in Eq. v vn W 1. At the corner frequency. (G-45) is a straight line that lies on the 0-dB axis. (G-48) to Eq. G-3-5 Quadratic Poles and Zeros Now consider the second-order transfer function G1s2 s 2 v2 n 2zvns v2 n 1 1 12z vn 2s 11 v2 2s2 n (G-45) We are interested only in the case when 1. Eq. (G-45) becomes 0G1 jv2 0 dB 40 log10 1v vn 2 dB (G-49) This equation represents a straight line with a slope of 40 dB decade in the Bodeplot coordinates. except that at high frequencies the slope of the straight-line approximation is 20 dB/decade. (G-45) becomes G1 jv2 The magnitude of G( j ) in dB is 20 log10 0G1 jv2 0 0G1 jv2 0 dB 20 log10 2 31 31 1 1v vn 2 4 2 j2z1v vn 2 1v vn 2 2 4 2 20 log10 1 4z2 1v vn 2 2 0 dB (G-46) (G-47) At very low frequencies. (G-42) is still at 1 T. Eq. (G-47) can be approximated as 20 log10 G1 jv2 20 log10 21v vn 2 4 (G-48) Thus. and the (1 j ) columns should all be negative. Therefore. For instance. the error (db). By letting s j . The phase of G( j ) is 0 degrees at 0. and 90 when . the low-frequency asymptote of the magnitude plot of Eq. (G-41).G-12 Appendix G Frequency-Domain Plots G-3-4 Simple Pole. the numbers in the 01 jvT 0 dB. since otherwise G(s) would have two unequal real poles. At very high frequencies. it is simple to extend all the analysis for the case of the simple zero to the Bode plot of Eq. is given by the negative of the right side of Eq. (G-42). The asymptotic approximations of 0G( j )0 dB at low and high frequencies are vT V 1 vT W 1 0G1 jv2 0 dB 0G1 jv2 0 dB 0 dB 20 log10 vT (G-43) (G-44) Thus. The magnitude in dB and phase of the Bode plot of Eq. (G-37). The actual magnitude cuve of G( j ) in this case may differ strikingly from the asymptotic curve. 1 (1 j T) 1 jvT For the function G1 jv2 1 (G-42) the magnitude. |G( j )| in dB. and the Bode plot can be obtained by considering G(s) as the product of two transfer functions with simple poles. and the phase G( j ) is the negative of the angle in Eq. the error between the straight-line approximation and the actual magnitude curve is 3 dB. G-8. v vn V 1. the straight-line approximation of 01 jvT 0 dB. the corner frequency of the Bode plot of Eq. The intersection of the two asymptotes is found by equating Eq. The reason for this is that the amplitude and phase curves of the second-order G( j ) depend not only on the .

The phase of G( j ) is given by G1 jv2 tan 1 e 2zv c1 vn a v 2 b df vn (G-50) and is plotted as shown in Fig. The standard procedure of constructing the second-order 0G( j )0 dB is to first locate the corner frequency n.3 0.707 1. G-10 for the same set of values of . which does not enter the asymptotic curve. G-9 for various values of .05 G( jv) (deg) – 45 = 1.1 0. The actual curve is obtained by making corrections to the asymptotes by using either the data from the error curves of Fig.2 0.0 –20 –30 – 40 0.0 –90 –135 –180 0.4 G-13 | G( jv) | (dB) 0. The analysis of the Bode plot of the second-order transfer function of Eq.G-3 Bode Plot (Corner Plot or Asymptotic Plot) 40 30 = 0.05 20 10 0 –10 = 0.6 0.01 0. G-9 for the corresponding .01 0. but also on the damping ratio . The errors between the two sets of curves are shown in Fig. The actual and the asymptotic curves of 0G( j )0 dB are shown in Fig.0 u = v/vn 10 100 Figure G-9 Bode plot of G1s2 1 1 2z1s vn 2 1s vn 2 2 . and 40-dB/decade line to the right of n. For G1s2 1 2z s vn 1 2 s v2 n (G-51) .3 0. corner frequency n.1 1. G-9 for several values of .4 0.2 0.1 1.0 u = v/vn 10 100 0 = 0.5 0.1 0. G-10 or the curves in Fig. (G-45) can be applied to the second-order transfer function with two complex zeros.

1v2 j0.1 1. passing through the 1 rad/sec point on the 0-dB axis. G-9. EXAMPLE G-5 As an illustrative example on the manual contruction of Bode plot. (G-23) and set s for computer plotting.0 u = v/vn 10 100 Figure G-10 Errors in magnitude curves of Bode plots of G1s2 1 1 2z1s vn 2 1s vn 2 2 . The errors between the actual and the asymptotic curves in Fig. The magnitude of the pure time delay term is equal to unity for all values of . G-3-6 Pure Time Delay. e jVTd • The magnitude of the pure time delay term is unity for all w.2v2 j0.01 0. G-10 are also inverted.707 1.0 0. 5. The complete Bode plot of the magnitude and phase of . The phase plot G( j ) is obtained by subtracting Td radians from the phase curve of G1(j ) at various .3 0. we have G1 jv2 1011 jv11 j 0. The pole at s 0 gives a magnitude curve that is a straight line with slope of 20 dB/decade.G-14 Appendix G Frequency-Domain Plots 25 20 = 0.5 0. the magnitude and phase curves are obtained by inverting those in Fig. The phase of the pure time delay term is e jvTd vTd jvTd (G-52) which decreases linearly as a function of . Thus. this step is unnecessary).1 10 |G( jv) | dB 5 0 –5 –10 –15 0. for the transfer function G1 jv2 the magnitude plot 0G( j )0 dB is identical to that of 0G1( j )0 dB.05 15 = 0.4 0.2 0.5v211 (G-55) Equation (G-54) shows that G( j ) has corner frequencies at 2. and 10 rad/sec.6 0. consider the function G1s2 101s s1s 102 52 221s (G-54) j (keeping in mind that G1 1 jv2e (G-53) The first step is to express G(s) in the form of of Eq.

G-11. the property of adding the curves of the individual components of the transfer function in the Bode plot does not carry over to this case. G-11. One of the most important applications of this type of plot is that when G( j ) is the forward-path transfer function of a unity-feedback control system.G-4 Magnitude-Phase Plot 40 –20 dB/sec 20 –40 dB/sec 0 |G( jv) | (dB) –20 –40 –60 –80 –100 0. G-4 MAGNITUDE-PHASE PLOT The magnitude-phase plot of G( j ) is a plot of the magnitude of G( j ) in dB versus its phase in degrees.78 rad/sec 1 2 5 10 v (rad/sec) 100 1000 Figure G-11 Bode plot of G1s2 101s s1s 102 52 221s . the plot can be superposed on the Nichols chart (see Chapter 9) to give information on the relative stability and frequency response of the system.10 Phase crossover 5. it is best to make the magnitudephase plot by computer or transfer the data from the Bode plot. G( j ) is obtained by adding the component curves together. Thus. When the gain factor K of the transfer function varies. The actual curves can be obtained by a computer program and are shown in Fig. as shown in Fig. .10 –90 –120 –150 G( jv) (deg) Gain crossover 3. However. in the construction of the plot. with as a parameter on the curve.88 rad/sec –60 dB/sec G-15 –40 dB/sec 1 2 5 10 v (rad/sec) 100 1000 –180 –210 –240 –270 0. the plot is simply raised or lowered vertically according to the value of K in dB. point by point.

G-12 and G-13. G-11.88 rad/sec 1 Re G Figure G-12 Polar plot of 101s 102 G1s2 .0 v –90.78 rad/sec Gain crossover v = 3.0 –202.88 rad/sec v = 10 rad/sec v = 30 rad/sec v = 100 rad/sec –247. and G-13. G-12. (G-54) are shown in Fig.0 –112.G-16 Appendix G Frequency-Domain Plots EXAMPLE G-6 As an illustrative example.78 rad/sec v= –1 0 Gain crossover 3.5 –225.0 . j Im G G-plane Phase crossover 5.5 Phase (deg) v –135. The relationships among these three plots are easily identified by comparing the curves in Figs.0 –157. s1s 221s 52 0 v = 1 rad/sec Phase crossover v = 5. respectively. G-11. s1s 221s 52 – 80 –270.5 –180. the polar plot and the magnitude-phase plot of Eq. The Bode plot of the function is already shown in Fig.5 40 30 20 10 0 0 v Magnitude (dB) – 10 – 20 – 30 – 40 – 50 – 60 – 70 Figure G-13 Magnitudephase plot of 101s 102 G1s2 .

3. given the function G1 jv2 1 1 jvT (G-56) the magnitude of G( j ) is the same whether T is positive (nonminimum phase) or negative (minimum phase). For instance. Additional properties of the minimum-phase transfer functions are as follows: 1. excluding the poles at s 0. the phase of G( j ) is different for positive and negative T. Gain-Crossover Point: The gain-crossover point on the frequency-domain plot of G( j ) is the point at which 0G( j )0 1 or 0G( j )0 dB 0 dB. given a minimumare uniquely related. G-6 MINIMUM-PHASE AND NONMINIMUM-PHASE FUNCTIONS A majority of the process transfer functions encountered in linear control systems do not have poles or zeros in the right-half s-plane. The phase- Bode Plot: The gain-crossover point (or points) is where the magnitude curve 0G( j )0 dB crosses the 0-dB axis. 2. . phase function G(s). G-13). The frequency at the gain-crossover point is called the gain-crossover frequency g. 1. when s j and as varies from to 0. given G( j ). In other words.G-6 Minimum-Phase and Nonminimum-Phase Functions G-17 G-5 GAIN. G( j ).and phase-crossover points on the frequency-domain plots are important for analysis and design of control systems. Nonminimum-phase transfer functions do not have the unique magnitude-phase relationships. the total phase variation of G( j ) is (n m) 2. Conversely. G-12). A nonminimum-phase transfer function will always have a more positive phase shift as is varied from to 0. characteristics of a Minimum-phase transfer functions have an important property in that their magminimum-phase function nitude and phase characteristics are uniquely related. it is called a nonminimum-phase transfer function. G-11). Phase-Crossover Point: The phase-crossover point on the frequency-domain plot of G( j ) is the point at which G( j ) 180°. These are defined as follows. This class of transfer functions is called the minimum-phase transfer functions. The phase-crossover point (or points) is where the phase curve crosses the 180 axis (see Fig. if any. When a transfer function has either a pole or a zero • The magnitude and phase in the right-half s-plane. However. The frequency at the phase-crossover point is called the phase-crossover frequency p. The gain and phase crossovers are interpreted with respect to three types of plots: Polar Plot: The gain-crossover point (or points) is where |G( j )| crossover point (or points) is where G( j ) 180° (see Fig. knowing its magnitude characteristics |G( j )| completely defines the phase characteristics. The phase-crossover point (or points) is where the G( j ) curve crosses the 180 axis (see Fig. For a minimum-phase transfer function G(s) with m zeros and n poles. The value of a minimum-phase transfer function cannot become zero or infinity at any finite nonzero frequency. Magnitude-Phase Plot: The gain-crossover point (or points) is where the G( j ) curve crosses the 0-dB axis. |G( j )| is completely defined.AND PHASE-CROSSOVER POINTS Gain.

transfer function for Bode diagrams of nonminimum-phase forward-path transfer functions should not be used for stastability studies. G-14(a). consider that the zero of the transfer function of Eq. when used along with a nonminimum-phase the Nyquist criterion discussed in Chapter 9. that is. • Do not use the Bode plot Care should be taken when using the Bode diagram for the analysis and design of systems with and the gain-phase plot of nonminimum-phase transfer functions. whereas the minimum-phase transfer function if Eq. (G-54). and the polar plot is shown in Fig. as shown in Fig. (G-54) has a net phase change of only 90 (from 180 to 90 ) over the same frequency range. is more convenient for nonminimum-phase systems. G-14(b). bility analysis of closed-loop control systems.1 1 10 v (rad/sec) (a) j Im G 0 v 100 1000 G-plane v= 0 Re G (b) Figure G-14 (a) Phase curve of the Bode plot. The same is true for the magnitude-phase plot.G-18 Appendix G Frequency-Domain Plots EXAMPLE G-7 As an illustrative example of the properties of the nonminimum-phase transfer function. G-11. G1s2 101s s1s 102 52 221s (G-57) The magnitude plot of the Bode diagram of G( j ) is identical to that of the minimum-phase transfer function in Eq. 90 45 G( jv) (deg) 0 – 45 – 90 –135 –180 0. (G-57) is shown in Fig. For stability studies. (G-54) is in the right-half s-plane. The phase curve of the Bode plot of G( j ) of Eq. G1s2 101s s1s 102 52 221s . (b) Polar plot. . the polar plot. Notice that the nonminimum-phase function has a net phase shift of 270 (from 180 to 90 ) as varies from to 0.

KUO FARID GOLNARAGHI JOHN WILEY & SONS. .Appendix H General Nyquist Criterion TO ACCOMPANY AUTOMATIC CONTROL SYSTEMS EIGHTH EDITION BY BENJAMIN C. INC.

electronic. recording. MA 01923. photocopying. ISBN 0-471-13476-7 . No part of this publication may be reproduced. fax (508)750-4470. stored in a retrieval system or transmitted in any form or by any means. without either the prior written permission of the Publisher. Inc. Inc. (508)750-8400. 605 Third Avenue.. except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act. or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center. fax (212) 850-6008. John Wiley & Sons. (212) 850-6011. All rights reserved. Requests to the Publisher for permission should be addressed to the Permissions Department. E-Mail: PERMREQ@WILEY. 222 Rosewood Drive. New York.COM. NY 10158-0012. mechanical. Danvers. To order books or for customer service please call 1-800-CALL WILEY (225-5945).Copyright © 2003 John Wiley & Sons. scanning or otherwise.

H-1(a) is the original one shown in Fig. j0) point. then the Nyquist path of Fig. or of 1 L(s). then plotting the Nyquist plot of L(s) only for s j to s j0 and not enclosing the ( 1. that is. number of poles of L(s).as well as nonminimum-phase loop transfer functions by using only the positive part of the j -axis of the Nyquist path. number of times the ( 1. the original Nyquist criterion requires that the L(s) plot that corresponds to the entire Nyquist path in Fig. However. Z P P N1 number of zeros of 1 L(s) that are in the right-half s-plane. H-1(a) and H-1(b). Therefore. but also all the poles and zeros of L(s) on the j -axis. With modern computer facilities and software. However. 9-20 be made. j0) point of the L(s)-plane that is encircled by the Nyquist plot of L(s) corresponding to s1. if there are any. in Fig. the analyst should not be burdened with the chores of manual plotting. 9-20 must have small indentations around them on the j -axis. Let us consider the two Nyquist paths shown in Figs. L(s) has either poles and/or zeros in the right-half s-plane.APPENDIX H General Nyquist Criterion H-1 FORMULATION OF NYQUIST CRITERION The original Nyquist criterion presented in Chapter 9 is cumbersome to apply when the loop transfer function is of the nonminimum-phase type. j0) point is enclosed. For a system with a nonminimum-phase loop transfer function L(s). if the system is of the minimum-phase type. 9-20. Yeung [1] introduced a general and yet simplified version of the Nyquist criterion that allows the determination of stability of closed-loop systems of minimum. that are on the j -axis. if the ( 1. Let us define the following quantities. H-1(b) encircles not only the entire right-half s-plane. including the origin. if the ( 1. the test of whether the ( 1. j0) point in the L(s)-plane is only a necessary but not sufficient condition for closed-loop stability. we are introducing a simplified Nyquist criterion that can be applied by using only the positive j -axis of the Nyquist path and then observing its behavior with reference to the ( 1. The rest of the Nyquist plot that corresponds to the small indentations and the large semicircle on the Nyquist path have to be plotted manually. Our MATLAB Toolbox (ACSYS) or other computer software can all be used to construct the plots of only functions that correspond to the positive j -axis of the s-plane. then an additional angle condition is all that must be satisfied by the Nyquist plot of L(s) for the system to be stable. in Fig. or of 1 L(s). the system is still unstable. If the loop transfer function L(s) has poles or zeros on the j -axis. the Nyquist path s1. j0) point is enclosed is still simpler to apply. Apparently. whereas the path s2. H-1 . This adds even more complexity to the construction of the L(s) plot. We shall show that if the loop transfer function is of the nonminimum-phase type. number of poles of L(s). We shall show that for nonminimum-phase systems. that are in the right-half s-plane. j0) point is not enclosed.

j0) point. £1 £2 N1 N2 360° 360° 1Z 1Z P2360° P Pv 2360° s1 Z P (H-1) (H-3) (H-4) Let us consider that each of the Nyquist paths and s2 is composed of three portions: 1. Educ. S. Then. H-1. the angles traversed by the Nyquist plots are identical for positive and negative values of . j0) point of the L(s)-plane that is encircled by the Nyquist plot of L(s) corresponding to s2. Then.” IEEE Trans. H-1 are symmetrical about the real axis in the s-plane. E-28. Yeung. (b) An alternative Nyquist path. Vol. j0) point. 58–60. respectively. . The portion from s j to j along the semicircle with infinite radius 2.H-2 Appendix H General Nyquist Criterion jv s-plane jv s-plane R R s s 0 0 s1 s2 (a) (b) Figure H-1 (a) Nyquist path. with reference to the two Nyquist paths in Fig. excluding the small indentations. N1 and N2 Z P Pv (H-2) Let 1 and 2 represent the net angles traversed by the Nyquist plot of L(s) with respect to the ( 1. and according to the Nyquist criterion. All the small indentations on the j -axis Since the Nyquist paths in Fig. 1 and 2 are written £1 £2 where 11 2£ 11 2£ 11 £ 12 £ 12 £ 13 £ 13 (H-5) (H-6) angle traversed by the Nyquist plot of L(s) with respect to the ( 1. Thus. (Source: K.) N2 number of times the ( 1. excluding all the small indentations 3. “A Reformulation of Nyquist’s Criterion. February 1985. pp. The portion along the j -axis. corresponding to the positive j -axis or the –j -axis of the s-plane.. corresponding to s1 and s2.

Eq. equals The number of zeros of 1 L(s) in the right-half s-plane the number of poles of L(s) in the right-half s-plane 0. j0) point. (H-8) gives the number of roots of the characteristic equation that are in the right-half s-plane. . it corresponds to the ( 1. Since on s2 the directions of the small indentations are opposite to that of s1. the angle 13 traversed by the phasor drawn from the ( 1. j0) point. the Nyquist plot of L(s) that corresponds to the infinite semicircle must either be a point on the real axis or a trajectory around the origin of the L(s)-plane. (H-5) is negative. Nyquist criterion can be carried out by drawing the Nyquist plot of L( j ) that corresponds to to 0. the closed-loop system is 11 270 is not the same as unstable. if the ( 1. excluding the small indentations. (H-9) for the system to be stable. j0) point is not enclosed by the Nyquist plot of L( j ). j0) point to the Nyquist plot along the semicircle with infinite radius is always zero. corresponding to the semicircle with infinite radius on the Nyquist paths. by knowing the values of 11. With reference to the Nyquist plot of L( j ) and the ( 1. if the closed-loop system is unstable. Thus. the angle variation 11 still has to satisfy Eq. the sign of 12 in Eq. For the closed-loop system to be stable. 11 However.H-1 Formulation of Nyquist Criterion H-3 12 13 angle traversed by the Nyquist plot of L(s) with respect to the ( 1.5(the number of poles of L(s) on the j -axis)180 • In general. Thus. angle traversed by the Nyquist plot of L(s) with respect to the ( 1. the Nyquist stability criterion can be carried out by constructing only the Nyquist plot that corresponds to the s j to s 0 portion on the Nyquist path.5Pv P2180° (H-10) Since P and P cannot be negative. (H-6) and using Eqs. Z must equal zero. the condition that the Nyquist plot of L( j ) not enclosing the ( 1. However. is positive with respect to the ( 1. 90 . j0) point is a necessary condition for closed-loop stability for nonminimum-phase systems. • The angle 11 is an angle variation. j0) point to the L(s) Nyquist plot that corresponds to the portion on the positive j -axis of the s-plan. Now adding Eq. (H-5) to Eq. the last equation indicates that if the phase traversed by the Nyquist plot of L( j ) as varies from to 0. for the nonminimum-phase system to be closed-loop stable. the Nyquist criterion for stability of the closed-loop system is £ 11 10. if any. corresponding to the small indentations on the j -axis of s1. P . if 11 is negative.5Pv 2180° (H-8) The equation states: The total angle traversed by the phasor drawn from the ( 1. we get £1 Solving for 11. j0) point. it still has to satisfy Eq. j0) point. £2 4£ 11 12Z 2P 1Z P Pv 2360° (H-7) we get £ 11 0. and P. (H-9) Thus. (H-9). (H-3) and (H-4). we see that when the angle variation 11 is positive. j0) point being enclosed. For a transfer function L(s) that does not have more zeros than poles. Furthermore. Thus. so that 11 .

j0) point is not enclosed by the Nyquist plot of L(s). j0) point. For improper transfer functions. then P L(s) that are at the origin. H-1-2 Systems with Improper Loop Transfer Functions Equation (H-8) is derived based on the condition that 13 0. 11 will always be given by Eq. that is. (H-12). s1s2 6s 52 v . the condition that the ( 1. and yet the system is unstable. it is easy to see that if the ( 1. EXAMPLE H-1 Consider that the loop transfer function of a control system is given by L1s2 s2 s1s2 s 1 6s 52 (H-13) j Im L L( jv)-plane v= –1 F 11 = –90° 0 Re L Enclosed area Figure H-2 Nyquist plot s2 s 1 of L1s2 . it has more poles than zeros.5Pv 2180° 90° 0. (H-8) becomes £ 11 For closed-loop stability. which is true only if L(s) is strictly proper. (H-11) becomes £ 11 Pv Since P denotes the number of poles of L(s) that are at the origin. when L(s) is of the minimum-phase type. Eq.H-4 Appendix H General Nyquist Criterion H-1-1 System with Minimum-Phase Loop Transfer Functions If L(s) is of the minimum-phase type. H-2 ILLUSTRATIVE EXAMPLES—GENERAL NYQUIST CRITERION MINIMUM AND NONMINIMUM TRANSFER FUNCTIONS In the following example we shall show that the Nyquist plot of a nonminimum-phase transfer function does not enclose the ( 1. Z 1Z 0 and P denotes the number of poles of (H-11) (H-12) 0. j0) point not be enclosed by the Nyquist plot is a necessary and sufficient condition for closed-loop stability. Thus. Eq. we can again use the method discussed in Section 9-6 by plotting the Nyquist plot of 1 L(s).

(H-14) The Nyquist plot of L(j ) for to 0 is plotted as shown in Fig. 1b2 K 7 240. Z must equal zero. from to 0. Thus. j0) point is not enclosed by the Nyquist plot. the loop transfer function of the system. Substituting these quantities into Eq. must equal 90 . which means that there are two closed-loop poles in the right-half s-plane. j0) point is to the right of the intersect as shown in Fig. the general Nyquist criterion does.H-2 Illustrative Examples Since L(s) has one pole at the origin. L( jv)-plane j Im L L( jv)-plane j Im L –1 F 11 = –90° v = 0 Re L v= –1 0 F 11 = +270° Re L Enclosed area Enclosed area 0 v (a) 0 v (b) Figure H-3 Nyquist plot of L1s2 s1s K 221s 102 . EXAMPLE H-2 Consider the system described in Example 9-1. in this case. the system would be unstable. but when the system is unstable. we get Z 2. we have Z 2. Substituting. 11 is indeed 90 .5Pv P2180° 450° H-5 2. and not 450 . (H-9). H-2. L(s). j0) point is to the left of the intersect of L( j ) with the real axis. the ( 1. 9-25. it does not tell how many characteristic equation roots are in the right-half plane. or (more easily observed when the critical point is enclosed). P 0. is given in Eq. However. thus. (H-16). (9-59). This means that the phasor drawn from the ( 1. Apparently. which means that the characteristic equation has two roots in the right-half s-plane. thus the last equation gives 11 90 . the Nyquist criterion with the “enclosure” test is easier to observe. j0) to the Nyquist plot. P 1 and P From Eq. or 90 in the CW direction.52180° (H-16) For closed-loop stability. we have £ 11 1Z 0. Now we shall apply the generalized Nyquist criterion to the system. L1 jv2 s1s K 221s 102 (H-15) The Nyquist plot of L( j ) is shown in Fig. Since L(s) is of the minimum-phase type. 11 is 270 . and is repeated below. then. since 11 is 90 . the closed-loop system is stable if the following condition is satisfied: £ 11 10. On the other hand. Substituting 11 90 into Eq. the ( 1. if the ( 1. for systems with minimum-phase loop transfer functions. 11 270 into Eq. and two poles in the right-half s-plane. 1a2 K 6 240. H-3(b) when the value of K is greater than 240. the system is unstable. . and it has one pole at the origin. P 1. It is shown in Example 9-1 that for closed-loop stability. Figure H-3(a) shows that if the ( 1. (9-82) and solving for Z. j0) point in the L( j )-plane must be to the left of the intersect of the Nyquist plot with the real axis. (H-11).

The characteristic equation of the system is s2 We can easily verify that stability requires 0 7 K 7 1 (H-27) 11 K2s K 0 (H-26) NYQUIST PLOT FOR K 0. the net angle 11 traversed by the phasor drawn from the ( 1. L1 j02 K ` jv v 90° 0 K1 jv jv1 jv 12 12 K1 jv v 2 12 jv (H-19) K ` jv v 0 90° (H-20) (H-21) To find the intersect of the L( j ) plot on the real axis. (H-7). j0) is enclosed by the Nyquist plot. We get L1 jv2 When . which means that the characteristic equation of the closed-loop system has one root in the right-half s-plane. the stability criterion requires that the phasor drawn from the ( 1. (H-19) by L1 jv2 K1 jv 121 v2 v 4 jv2 v 2 K 32v 2 j11 2 v2 2 4 v1v 12 (H-22) Setting the imaginary part of L( j ) to zero and solving for v2 For 1. (H-8).5Pv P2180° 1Z 0. since it has a zero at s 1. Thus. The function L(s) is of the nonminimumWe observed from the last equation that P phase type. we set s j in Eq. and so the same conclusion on closed-loop stability can be drawn.52180° 90° (H-25) Thus. £ 11 1Z 0. L1 j 2 When 0. the Nyquist plot of L( j ) that corresponds to the positive portion of the j -axis is sketched as shown in Fig. and P 0. Notice that the plot is obtained by rotating the L( j ) plot of Fig. we rationalize the function by multiplying 2 j . j0) point to the Nyquist plot is 90 . the requirement for closed-loop stability is £ 11 10. From Eq. j0) point to the Nyquist plot of L( j ) should traverse 90 as varies from to 0. we have (H-23) (H-24) NYQUIST PLOT FOR K 0. the system is unstable because 11 is positive. the Nyquist criterion on enclosure cannot be used adequately in this case. To sketch the Nyquist plot of L(s) that corresponds to the postive portion of the j -axis of the s-plane. Based on the preceding information. We can also readily see that ( 1. (H-17). We have the numerator and the denominator of Eq. Figure H-4 shows that as varies from to 0 along the Nyquist plot. L1 j12 K 1 rad/sec . Figure H-5(a) shows the Nyquist plot of L( j ) when K lies between 0 and 1. Thus.H-6 Appendix H General Nyquist Criterion EXAMPLE H-3 Consider that a control system has the loop transfer function L1s2 K1s s1s 12 12 (H-17) 1.5Pv P2180° 90° (H-18) Thus. Z 1. From Eq. H-4 by 180 . H-4 for K 0.

we have 11 270 . H-5(a) has a net rotation of 90 .52180° 270° (H-28) Thus. 1b2 K 6 1. and the closed-loop system is stable. Figure H-5(b) shows the Nyquist plot when K 1. so system is unstable. which means that the characteristic equation has two roots in the right-half s-plane. when K changes sign. It should be reminded that since L( j ) is of the minimum-phase type. j0) is not the reason that the system is stable. Checking the value of 11. which agrees with the stability requirement in Eq. which differs from the required 90 .H-2 Illustrative Examples j Im L 0 v H-7 L( jv)-plane Enclosed area F 11 = +90° v= –1 0 K Re L Figure H-4 Nyquist plot of the system in Example H-3. . K1s 12 L(s) . In general. H-5. the angle 11 in Fig. Equation (9-40) can be written as 1 L1s2 1 KL1 1s2 0 (H-29) j Im L L( jv)-plane –1 < K < 0 j Im L L( jv)-plane K < –1 v= –1 F 11 = –90° K v=1 0 Re L K v= –1 0 F 11 = +270° Re L v=1 Enclosed area Enclosed area v (a) Figure H-5 Nyquist plots of the system in Example H-3. £ 11 1Z 0.5Pv P2180° 1Z 0. j0) point is enclosed. L(s) 1a2 1 6 K 6 0. As is varied from to 0. v 0 0 (b) K1s s1s 12 12 . H-5(a) does not enclose the ( 1. it is not necessary to redraw the Nyquist plot as shown in Fig. s1s 12 about the origin. Z 2. (H-8). the fact that the Nyquist plot of L( j ) of Fig. Using Eq. Now we see that the ( 1. (H-18).

which is the required value. (H-17) for K 0. j0) point as the critical point for stability analysis. j0) point is regarded as the critical point. j0) point is the critical point. (H-31) shows that when K is negative. but designate the ( 1. . Eq. Thus. for 1 K 0.414 – j1 K = –1 K< 0 Figure H-7 Complete root loci of system in Example H-3. H-5 when the Nyquist plots for K 0 were actually constructed.H-8 Appendix H General Nyquist Criterion j Im L v 0 L( jv)-plane Enclosed area K < –1 Enclosed area F 11 = –90° –K 1 –K Re L –1 < K < 0 v 0 v= 0 Figure H-6 Nyquist plot of K1s 12 L1s2 with K 6 0. and the system is stable. the ( 1. the last equation can be written as 1 or KL1 1s2 KL1 1s2 1 0 (H-30) (H-31) where K is now positive. It is of interest to compare the Nyquist stability analysis with the root-locus analysis.414 K K=0 –1 0 K=0 K> 0 K= ± 1 K< 0 K K< 0 – s –0. s1s 12 The ( 1. These results agree with those obtained from Fig. the ( 1. 11 is 90 . H-6. As shown in Fig. When K 1. and the system is unstable. When K is negative. we can still use the L( j ) plot for positive K. Figure H-7 shows the root loci of the characteristic equation of the system with the loop transfer function jv s-plane K = –1 K< 0 j1 2. where K is positive. Figure H-6 shows the Nyquist plot of Eq. For negative K. j0) point is enclosed by the Nyquist plot.

The RL between 0 and 1 on the real axis indicates that the system is unstable for 0 K .162 K (H-39) R(s) + – E(s) K(s + 2 ) + – 10 s2(s + 3) Y(s) Figure H-8 Block diagram of the control system in Example H-4. The transfer function L(s) is of the nonminimum-phase type. (H-32) becomes L1 jv2 At .16 rad sec. H-8. we already have L( j0) 2K in Eq. It is desired to determine the range of K for which the system is stable. (H-36). and 0. EXAMPLE H-4 Consider the control system shown in Fig. Or. P 2. which are the frequencies at which the L( j ) plot intersects the real axis of the L( j )-plane. which are the values of at which the Nyquist plot of L( j ) intersects the negative real axis. and the system is stable only for the range of 1 K 0.6. and P 0.361 j1. The loop transfer function of the system is L1s2 Y1s2 E1s2 s 10K1s 3 22 2 3s 10 (H-32) The poles of L(s) are found to be at s 3. L1 j 2 At 0. (H-17). The RL indicates that for negative values of K. we have v110 3v2 2 The solutions of the last equation are 0 and v 110 3. From Eq. the system is unstable for K 1. When 3. the requirement for the closed-loop system to be stable is £ 11 Setting s j .16 rad/sec. (H-9).6.361 j1. we can use the Routh-Hurwitz criterion to verify that L(s) has two poles in the right-half s-plane. L1 j02 2K (H-36) 0 180° (H-35) 110 10K1 jv 3v 2 2 10. Eq.72. The stability condition of the system as a function of K is clearly indicated by the root-locus diagram.5Pv P2180° 360° (H-33) 22 jv3 (H-34) To find the intersect on the real axis of the L( j )-plane.H-2 Illustrative Examples H-9 given in Eq. The root loci cross the j -axis at 1 rad/sec. 0. 2v3 0 (H-38) L1 j3. When 0. Thus. . we rationalize L( j ) as L1 jv2 10K 52 110 3v2 2 110 v4 3v2 2 2 j3v110 v6 3v2 2 2v2 4 6 (H-37) Setting the imaginary part of L( j ) to zero.

not K is greater than unity. Since the ( 1. (a) 1 K 0. the closed-loop system is unstable.36 – j1. thus.72 K > 0 –2 0.6 – K K< 0 K=0 K= ± 0 – 0. the angle 11 rotates a total of 360 .5 K< 0 K – s –3.6 K=1 –j 10 Figure H-10 Complete root loci of the system in Example H-4. Figure H-9(b) shows the Nyquist plot of L( j ) when 11 is 0 .564 K=0 0. Figure H-9(a) shows the Nqyuist plot of L( j ) for 0 K 1.43K v = 1. K .43K v = 1. We can also show that the angle traversed by 360 .54 L ( jv)-plane j Im L j 5. In this case.54 L ( jv)-plane v= F11 = 0° 10 –K v = –1 v= 10 v=0 2K Re L F11 = –360° –K –1 v= 0 v=0 2K Re L 0 (a) (b) Figure H-9 Nyquist plots of the system in Example H-4. as required in Eq. j0) point is enclosed by the Nyquist plot. the system is stable. (H-33).H-10 Appendix H General Nyquist Criterion j Im L j 5.36 + j1. (b) K 1. jv K K=1 j 10 s-plane K=0 0.

j0) point lies between 0 and 2K. v L( jv)-plane F11 = 0° Critical point K < –8 Unstable F11 = –180° Critical point 0 > K > –8 Stable v= –1 0 +1 K/8 +1 Re L v=0 45° Enclosed area v 2 . j Im L 2 Enclosed area Critical point K>0 F11 = +180° Unstable Figure H-11 Nyquist plot of the control system in Example H-5. the system is unstable. and is not enclosed by L( j ). which corresponds to the frequency at which the L( j ) plot intersects the negative real axis. one branch of the RL will always stay in the right-half plane.16 rad sec. The system is stable only for K 1. 11 0 . H-9.H-2 Illustrative Examples H-11 When K is negative. the portion that corresponds to s j to j0 is plotted as shown in Fig. Thus. Instead of constructing the entire Nyquist plot. Clearly. and the root loci cross the j -axis at 3. and regard the ( 1. In this case. j0) point as the critical point. The value of K at the crossing point on the j -axis is 1. The conclusion is that the system is stable for K 1. and P 0. EXAMPLE H-5 Consider a control system with the loop transfer function L 1s2 1s K 221s2 42 (H-40) 2. For stability. The root loci of the system are shown in Fig. To apply the which has a pair of imaginary poles at s j2 and j2. 11 must equal 360 . and is enclosed by the Nyquist plot. when K is negative. j0) point is to the right of the point 2K. 11 The ( 1. the system is unstable. The data for this Nyquist plot are easily obtained using any of the frequency-domain program mentioned earlier. P Nyquist criterion in the original form. we can use the plots in Fig. H-10. and the system is unstable. but 180 . we would have to define the Nyquist path with small indentations around these poles. H-11. The following stability conditions are observed: 2K 1 1: 2K 0: The ( 1.

(H-41). and for the portion of 2 to 0. jv K K> 0 0 – K K < s-plane K=0 j2 K = –8 0 K = 0 –j2 ` K K> 0 K=0 K K< 0 – s K 0 > K< 0 K K Figure H-12 Complete root loci of the system in Example H-5. The system is unstable for all positive values of K. the total 11 is 180 . The system is stable for 0 K 8. from to 2 is 45 . H-11 is at ( 1. When K is negative. when varies from to 2. it is enclosed by the Nyquist path. The summary of the Nyquist criterion application to this system is as follows. and the system is unstable. Thus. and from 2 to 0 is 135 . For the angle check. 11 Range of K K K 8 0 8 K 0 (deg) for 2 to 0 135 45 45 11 (deg) for 2 to Total (deg) 180 0 180 11 Critical Point 1 point enclosed 1 point enclosed Stability Condition Unstable Unstable Stable 45 45 135 The complete root loci of the characteristic equation of the system are constructed in Fig. Thus the total 11 as varies 11 from to 0 is 180 .5Pv P2180° 180° (H-41) 11 As seen from Fig. Thus. – . When the ( 1. the critical point ( 1. which agrees with the value required in Eq. the value of £ 11 required for stability is 10. H-11. H-12 using the pole-zero configuration of Eq. j0) point is to the right of the K 8 point. j0) point lies between 0 and K 8. and the system is unstable. not 180 . the angle 11 is 135 . j0) is not enclosed by the Nyquist plot. j0). the magnitude of L( j ) goes to infinity when 2 rad/sec. The stability condition of 8 K 0 is easily viewed from the root loci. the system is unstable for K 8. When K is positive. 11 is 45 . (H-40). Figure H-11 shows that if the ( 1. (H-9).H-12 Appendix H General Nyquist Criterion From Eq. the critical point in Fig.

However. it is important to forewarn or take precautionary measures to prevent opening the loops during operation. root loci. In general. one at a time.414 rad sec. if such a situation exists. It does not matter whether the system is with single loop or multiple loops. stability analysis can be conducted using either the Routh-Hurwitz criterion. more insight may be gained on the stability of the individual loops of the system. Since the plot intersects the real axis at the 1 point at 1. the system will oscillate continuously with a frequency of 1. EXAMPLE H-6 Figure H-13 shows the block diagram of a system that controls the gun turret of a tank. one should conduct a systematic stability analysis of all the inner loops of the system. For multiloop feedback systems. During servicing of the turret control system.1) s + 10 Gc (s) + 6 s(s + 1)(s + 2) – G(s) Y(s) Figure H-13 Multiloop feedback control system for tank turret control. the inner loop is marginally stable. it may be advantageous to analyze the stability of the system by working from the inner loop toward the outer loop. The following example will illustrate this approach. as long as the overall system is stable. since once the loop transfer function is obtained. and finally flew apart. The loop transfer function of the overall system is L1s2 Gc 1s2G1s2 1 G1s2 1s 1001s 1021s3 0. It is admissible to have unstable inner loops. Therefore. for a multiloop control system. This way. With the power turned on. The purpose of this example is to show that it is inadequate to investigate just the stability of the overall system. or the Nyquist criterion. if the outer loop of the system is opened. .67(s + 0. The loop transfer function of the inner loop is G1s2 s1s 6 121s 22 (H-42) Figure H-14 shows the Nyquist plot of G(s). the mechanic accidentally opened the outer loop of the system. the gun turret went out of control.H-3 Stability Analysis of Multiloop Systems H-13 H-3 STABILITY ANALYSIS OF MULTILOOP SYSTEMS The Nyquist stability analyses conducted in the preceding sections are all directed toward the loop transfer function L(s).12 3s2 2s 62 (H-43) R(s) + – E(s) 16.414 rad sec.

2 0 –0. the total value of 11 for to 0 is 180 .4 0.5 1.414 rad sec is 90 . Thus.H-14 Appendix H General Nyquist Criterion 0.8 0. Since L(s) has two poles on the j -axis and the rest are in the left-half s-plane. the proper way is to start with the stability of the innermost loop by opening all the outer loops.5 v= –1 –0.2 –0.5 L( jv)-plane 1.5 v –1 –0.414 –1 0 G( jv)-plane j Im G v= Re G 0 –1. The angle 11 for to 1.414 rad sec to 0 is 90 . £ 11 10. v 41 4 1 0.5 0 0.6 –0. In general.4 14 v 0 v=0 1 6 –0.8 v = 1. Thus. (H-9) for the stability test of the overall system.12 3s2 2s 62 .4 0 Re L –1 –0.4 –0. so we must use Eq. and then add one loop at a time. P 2 and P 0.5 Figure H-15 Nyquist plot of L1s2 1s 1001s 1021s3 0. and the overall system is stable. and from 1. H-15.29 v = 5. when more than two loops are involved. j Im L 1.5Pv P2180° 180° (H-44) The Nyquist plot of L(s) is plotted as shown in Fig.5 Figure H-14 Nyquist plot of G1s2 s1s 6 121s 22 . until the outermost loop is closed.6 0. The function is also of the nonminimum-phase type.5 0 0. .

For each case. February 1985. K1s 22 K (a) L1s2 (b) L1s2 2 s1s 1021s 22 s1s 12 K1s2 5s 22 K1s 12 (c) L1s2 (d) L1s2 s1s3 3s 12 s1s3 2s2 2s 102 2 12 1s 22 K1s K1s2 5s 12 (e) L1s2 (f) L1s2 2 s1s s 12 s1s 121s2 42 H-3. Figure HP-5 shows the Nyquist plots of loop transfer function L( j ) for 0 to for single-feedback-loop control systems. PROBLEMS • Stability analysis with Nyquist criterion.. The loop transfer functions of single-feedback-loop control systems are given in the following equations. Determine the stability of the closedloop system by applying the Nyquist criterion. Educ. Check the answers by means of the Routh-Hurwitz criterion. and P refers to the number of poles of 1 1 L( j ) that are in the right-half s-plane. and in the right-half s-plane. give the number of zeros of 1 L(s) that are in the right-half s-plane.11s2 121s 22 s1s2 s 12 H-2. Determine the stability of the closed-loop system. find the number of poles of the closed-loop transfer function that are in the right-half s-plane. P. 51s 22 50 (a) L1s2 (b) L1s2 s1s 12 1s 12 s1s 521s 12 (c) L1s2 (e) L1s2 31s s1s3 s2 s1s 3 22 3s 5s 2s2 2s 12 2 102 (d) L1s2 (f) L1s2 100 s1s 121s2 22 0.” IEEE Trans. criterion to determine the values of K for system stability. Figure HP-3 shows the Nyquist plots of the loop transfer function L( j ) for 0 to for single-feedback-loop control systems. Sketch the Nyquist plot of L( j ) for 0 to . H-5. If the system is unstable. 58–60. P . Apply the Nyquist and Routh-Hurwitz criteria. Determine the stability of the systems described by the function 1 L( j ) shown in Fig. Sketch the Nyquist plot of L( j ) with K 1 for 0 to . Solve for the intersect of L( j ) on the negative real axis of the L( j )-plane analytically. where P refers to the number of poles of 1 L( j ) that are on the j -axis. The characteristic equations of linear control systems are given below. it is necessary to plot the Nyquist plot of 1 L( j ) to apply the simplified Nyquist criterion. are indicated for each case. • Application of Nyquist H-6. pp. The number of poles of L( j ) that are on the j -axis and in the right-half s-plane are indicated in each case. The loop transfer functions L(s) of single-feedback-loop systems are given in the following equations.Problems H-15 REFERENCE 1. Yeung. E-28. Vol. H-4. You may use a computer program to plot the Nyquist plots. K. (a) s3 4Ks2 1K 52s 10 0 (b) s3 K1s3 2s2 12 0 2 2 (c) s1s 12 1s (d) s3 2s2 20s 10K 0 42 K1s 12 0 (e) s1s3 3s 32 K1s 22 0 . the values of P and P for the function 1 L( j ) are given. “A Reformulation of Nyquist’s Criterion. The number of poles of L( j ) that are on the j -axis. Determine the range(s) of K for closed-loop system stability. HP-4. S. It was mentioned in the text that when the function L( j ) has more zeros than poles. For the unstable systems. H-1. You may construct the Nyquist plot of L( j ) using any computer program. Apply the Nyquist criterion and determine the values of K for the system to be stable. The gain K appears as a multiplying factor in L(s).

P = 0 v (b) –1 v ` 0 Re L j Im L L-plane v=0 v ` j Im L L-plane v=0 Re L –1 0 v ` Re L –1 P = 0. P = 0 v (a) v ` 0 Re L P = 2. P = 0 v 0 Re 1 L –1 Pv = 2.H-16 Appendix H General Nyquist Criterion j Im L v 0 j Im L L-plane L-plane 0 v –1 P = 3. P = 0 v 0 v (e) P = 1. P = 0 j Im 1 L 1 -plane L v ` 0 Re 1 L Figure HP-4 (a) (b) . P = 0 v (f) Figure HP-3 v 0 j Im 1 L 1 -plane L v ` –1 P = 3. P = 0 v (d) j Im L L-plane v ` –1 2 v j Im L v 1 v=0 2 18° Re L –1 L-plane v ` 0 Re L 0 P = 2. P = 2 v (c) P = 0.

P = 2 v P = 1.Problems j Im L j Im L L-plane 0 v 0 v v 0 H-17 v 0 –2K L-plane Re L –2K P = 0. P = 0 v Figure HP-5 (c) (d) .5K Re L –2K v= v=0 0 Re L 0 v P = 0. P = 0 v (a) j Im L L-plane v 0 –2K –0. P = 1 v (b) j Im L L-plane Re L P = 3.

INC.Appendix I Discrete-Data Control Systems TO ACCOMPANY AUTOMATIC CONTROL SYSTEMS EIGHTH EDITION BY BENJAMIN C. KUO FARID GOLNARAGHI JOHN WILEY & SONS. .

John Wiley & Sons. (508)750-8400. 605 Third Avenue.Copyright © 2003 John Wiley & Sons. Inc. ISBN 0-471-13476-7 . To order books or for customer service please call 1-800-CALL WILEY (225-5945).. All rights reserved. New York. Inc. fax (212) 850-6008. scanning or otherwise. or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center. No part of this publication may be reproduced. except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act. photocopying. NY 10158-0012. recording.COM. E-Mail: PERMREQ@WILEY. without either the prior written permission of the Publisher. mechanical. MA 01923. fax (508)750-4470. Danvers. electronic. Requests to the Publisher for permission should be addressed to the Permissions Department. (212) 850-6011. 222 Rosewood Drive. stored in a retrieval system or transmitted in any form or by any means.

2. …. The system is characterized by digitally coded signals at various parts of the system. In addition. I-1. However. z-transform is an operational method of solving linear time-invariant difference equations. linear digital control systems are described by difference equations (see Appendix B). there are distinct advantages working with digital versus analog signals. One important property of the z-transform is that it can r(t) DIGITAL CONTROLLER r*(t) D/A h(t) PROCESS y(t) Figure I-1 Block diagram of a typical digital control system. mainly because of advances made in microprocessors and microcomputers. I-2-1 Definition of the z-Transform Consider the sequence y(k). Therefore. k 0. a digital control system often requires the use of digitalto-analog (D/A) and analog-to-digital (A/D) converters. The block diagram of a typical digital control system is shown in Fig. Similarly. I-2 THE z-TRANSFORM Just as linear continuous-data systems are described by differential equations.APPENDIX I Discrete-Data Control Systems I-1 INTRODUCTION In recent years discrete-data and digital control systems have become more important in industry. where y(k) could represent a sequence of numbers or events. I-1 . such as a dc motor. driven by analog signals. 1. The significance of this definition will become clear later. We have seen that Laplace transform is a powerful method of solving linear time-invariant differential equations. the output device of the system is usually an analog component. The z-transform of y(k) is defined as Y(z) z-transform of y(k) a y1k2z k 0 k Z[y(k)] (I-1) where z is a complex variable with real and imaginary parts.

and solving for Y(z). (I-8) with Eq. Then. we can relate the sequence y(kT ) with a signal that can be expressed as y*1t2 a y1kT 2d1t 0 k kT 2 (I-7) Taking the Laplace transform on both sides of Eq. we have y1k2 1 k 0. The impulse at the kth time instant. we see that the z-transform may be related to the Laplace transform through z eTs (I-9) . (I-1). The latter is defined as the sampling period. Multiplying both sides of the last equation by e a z 1. the z-transform of y(k) is Y1z2 which converges for 0z 0 7 1. the latter is expressed in closed form as Y1z2 for 0e az 1 1 1 e az z 1 z e a (I-4) 0 6 1. 1. 2. carries the value of y(kT). we have Y*1s2 L3y*1t2 4 a y1kT 2e 0 kTs (I-8) k Comparing Eq. the z-transform of y(k) is written Y1z2 ae k 0 ak z k 1 e a z 1 e 2a z 2 p (I-3) which converges for 0e a z 1 0 6 1. (I-1). The following examples illustrate the derivation of the z-transforms of two simple functions. It may be useful to represent the sequence y(kT). 2. (I-7). EXAMPLE I-1 Consider the sequence y1k2 where e ak k 0. engineers feel more at home dealing with signals. Thus.I-2 Appendix I Discrete-Data Control Systems convert a sequence of numbers in the real domain into an expression in the complex z-domain. p (I-2) is a real constant. if which represents a sequence of ones. (t kT). 1. p (I-5) EXAMPLE I-2 In Example I-1. 1 z 1 z 2 p z z 1 (I-6) I-2-2 Relationship between the Laplace Transform and the z-Transform While the mathematicians like to talk about sequences. Applying Eq. 0. 2. subtracting the resulting equation from Eq. … as a train of impulses separated by the time interval T. k 0. This situation occurs quite often in digital and sampled-data control systems in which a signal y(t) is digitized or sampled every T seconds to form a time sequence that represents the signal at the sampling instants. 1. (I-3).

The definition of the z-transform in Eq. For uniformity. (I-1) may be regarded as a special case with T 1. Multiplication by a Constant Z3ay1kT2 4 where is a constant. the z-transforms of more complex functions may be obtained with the help of some of the z-transform theorems that follow. Theorem 1. we can write Y1z2 Z3y1t2 4 Z3Y1s2 4 (I-11) Z3y1kT 2 4 Z3y*1t2 4 Z3Y*1s2 4 (I-10) with the understanding that the function y(t) is first sampled or discretized to get y*(t) before taking the z-transform. Represent the values of y(t) at the time instants t y*(t): y*1t2 2. giving the z-transform. I-2-3 Some Important Theorems of the z-Transform Some of the commonly used theorems of the z-transform are stated in the following without proof. (I-13): Y*1s2 ae k 0 akT e kTs ae k 0 1s a2kT (I-14) 3.I-2 The z-Transform I-3 In fact. (I-9). Express Y*(s) in closed form and apply Eq. …. Thus. ae k 0 akT 0. EXAMPLE I-3 Consider the time function y1t2 e at us 1t2 kT. respectively. k kT 2 (I-12) The z-transform of y(t) is obtained by performing the following steps: 1. (I-9) allows us to treat sampled systems and perform digital simulation of continuous-data systems. 1. Just as in the case of the Laplace transform. Y1z2 z z e aT (I-15) In general. the z-transform as defined in Eq. 2. T can be set to unity. For engineering purposes. a z-transform table such as that in Appendix J may be used to transform from y(k) to Y(z). we can summarize the definition of the z-transform as Y1z2 Or. Addition and Subtraction (I-16) If y1(kT) and y2(kT) have z-transforms Y1(z) and Y2(z). then Y1 1z2 Y2 1z2 Z3y1 1kT 2 y2 1kT 2 4 Theorem 2. to form the function d1t (I-13) Take the Laplace transform on both sides of Eq. aZ3y1kT2 4 aY1z2 (I-17) . and if a sampling period is not involved. these theorems are useful in many aspects of the z-transform analysis. the real sequence is expressed as y(kT).

. Thus. is defined only for k 0. (I-19) denotes that of a time sequence shifted to the left by nT. One exception to this in the case of the z-transform is if one of the two functions is the delay e NTs. The following examples illustrate the usefulness of some of these theorems. Y(z) is the z-transform of y(kT ). and Eq. similar to the Laplace transform. • The final-value theorem is valid only if (1 z 1)Y(z) does not have poles on or inside the unit circle |z| 1. lim y1kT2 zS lim Y1z2 (I-21) if the limit exists. (I-19) is not just znY(z) is because the one-sided z-transform. the z-transform of the product of two real functions y1(k) and y2(k) is not equal to the product of the z-transforms Y1(z) and Y2(z). Theorem 6. Real Translation (Time Delay and Time Advance) Z3y1kT nT 2 4 z n c Y1z2 z nY1z2 (I-18) and Z3y1kT where n is a positive integer. Thus. Theorem 4. Real Convolution Z c a y1 1kT2y2 1NT k 0 N Y1 1z2Y2 1z2 Z3y1 1kT 2 * y2 1kT 2 4 kT 2 d Z c a y2 1kT 2y1 1NT k 0 N kT 2 d (I-23) where “*” denotes real convolution in the discrete-time domain. The reason that the right-hand side of Eq. (I-19) simply represents the sequence that is lost after it is shifted to the left of k 0. then Z3e NTs Y1s2 4 Z3e NTs 4Z3Y1s2 4 z NY 1z2 (I-24) Table I-1 summarizes the theorems on the z-transform just given. where N is a positive integer. Complex Translation Z3e where akT nT 2 4 n k k a y1kT 2z 4 0 1 (I-19) y1kT 2 4 Y1ze aT 2 (I-20) is a constant.I-4 Appendix I Discrete-Data Control Systems Theorem 3. we see that as in the Laplace transform. Equation (I-18) represents the z-transform of a time sequence that is shifted to the right by nT. Theorem 7. the second term on the right-hand side of Eq. Initial-Value Theorem kS0 Theorem 5. Final-Value Theorem z 1)Y(z) has no poles on or outside the unit circle 0z 0 kS lim y1kT2 lim 11 zS1 z 1 2Y1z2 (I-22) 1 in the if the function (1 z-plane.

.I-2 The z-Transform TABLE I-1 Theorems of z-Transforms Y1 1z2 Y2 1z2 Z3y1 1kT2 y2 1kT2 4 Z3ay1kT2 4 aZ3y1kT2 4 aY1z2 Z3y1k n2T4 z nY1z2 Z3y1k n2T4 zn c Y1z2 k a y1kT 2z d k 0 n 1 I-5 Addition and subtraction Multiplication by a constant Real translation (time delay) (time advance) Complex translation Initial-value theorem Final-value theorem where n positive integer Y1ze aT 2 Z3e akTy1kT2 4 lim y1kT 2 lim Y1z2 kS0 kS lim y1kT 2 if (1 Y1 1z2Y2 1z2 Real convolution z 1)Y(z) has no poles on or inside 0z0 zS1 lim 11 zS z 1 2Y1z2 1. and then the final time response determined by the inverse z-transform. Since the function (1 z 1)Y(z) does not have any pole on or outside the unit circle 0z0 the z-plane. then (I-25) Using the complex translation theorem in Eq. (I-20). not on y(t). the z-transform carries • The inverse z-transform of Y(z) is y(kT). In general. In other words.2082 (I-27) determine the value of y(kT) as k approaches infinity. the final-value theorem in Eq. not y(t). Z c a y1 1kT 2y2 1NT k 0 n k 0 n kT 2 d Z c a y2 1kT 2y1 1NT Z3y1 1kT 2 * y2 1kT 2 4 kT 2 d EXAMPLE I-4 (complex translation theorem) To find the z-transform of y(t) F1z2 Z3tus 1t2 4 Z1kT 2 1z Tz te 12 2 t .146z 0.792 z 0.792 z2 121z2 0. t 0. the inverse z-transform of a function Y(z) yields information on y(kT) only.208 (I-28) I-2-4 Inverse z-Transform Just as in the Laplace transform.416z 0. let f (t) t. Thus. (I-22) can be applied. we obtain Y1z2 Z3te at us 1t2 4 F 1zeaT 2 1z Tze e aT aT 2 2 (I-26) EXAMPLE I-5 ( final-value theorem) Given the function Y1z2 1z 0. one of the major objectives of the z-transform is that algebraic manipulations can be made first in the z-domain. kS 1 in lim y1kT 2 lim zS1 z 2 0.

Expanding Y(z) z by partial-fraction expansion. we should expand Y(z) into the form of Y1z2 z K1z e aT K2z z e p bT (I-29) To do this. first expand Y(z) z into fractions and then multiply by z to obtain the final expression. example illustrates this situation. Partial-Fraction Expansion Method The z-transform function Y(z) is expanded by partial-fraction expansion into a sum of simple recognizable terms. the partial-fraction expansion of Y(z) is carried out directly. the partialfraction expansion of Y(z) z should first be performed. the inverse z-transform can be carried out by one of the following three methods: 1. we note that practically all the z-transform functions have the term z in the numerator. 2. we have Y1z2 z The final expanded expression for Y(z) is Y1z2 z z 1 z z e aT 1 z 1 z 1 e aT (I-31) (I-32) From the z-transform table in Appendix J. 1. In carrying out the partial-fraction expansion. this usually means that the time sequence has a delay. In this case. Therefore. We have Y1z2 1 z 1 z 1 e aT (I-35) . p (I-33) It should be pointed out that if Y(z) does not contain any factors of z in the numera• If Y(s) does not have any zeros at z 0. EXAMPLE I-6 Given the z-transform function Y1z2 1z 11 2z e 121z aT e aT 2 (I-30) find the inverse z-transform. the corresponding inverse z-transform of Y(z) is found to be y1kT 2 1 e akT k 0. 3. The following example will illustrate this procedure. there is a slight difference between the z-transform and the Laplace transform procedures. The following expansion of Y(z) directly. With this in mind. Partial-fraction expansion Power-series method The inverse formula • If Y(s) has at least one zero at z 0. and the partial-fraction expanperform the partial-fraction sion of Y(z) should be carried out without first dividing the function by z. 2. then tor. EXAMPLE I-7 Consider the function Y1z2 1z 11 2 e aT 121z e aT 2 (I-34) which does not contain any powers of z as a factor in the numerator.I-6 Appendix I Discrete-Data Control Systems information only at the sampling instants. With reference to the z-transform table. and the z-transform table is used to determine the corresponding y(kT).

the second term on the right-hand side of Eq. p (I-36) Similarly. EXAMPLE I-8 Consider the function Y(z) given in Eq. if we expand Y(z) into a power series in powers of z k. the computer programs designed for performing the partial-fraction of Laplace transform functions can still be applied. p (I-39) which is the same result as in Eq. (2-10). (I-30). it is apparent that y1kT 2 e akT k 0. As a simple example.I-2 The z-Transform I-7 Although the z-transform table does not contain exact matches for the components in Eq. (I-1) gives a straightforward method of carrying out the inverse z-transform. Thus. let us consider the first-order unforced difference equation y1k 12 y1k2 0 (I-41) . 1. Inversion Formula The time sequence y(kT) can be determined from Y(z) by use of the inversion formula: y1kT2 1 Y1z2z k 1dz 2pj C (I-40) which is a contour integration along the path . The inversion formula of the z-transform is similar to that of the inverse Laplace-transform integral given in Eq. that is. (I-33). a circle of radius 0z 0 ecT centered k 1 at the origin in the z-plane. 2. 2. 2. Based on Eq. (I-35) can be identified with a time delay of T seconds. 2. we recognize that the inverse z-transform of the first term on the right-hand side can be written as Z 1 c 1 z 1 d Z 1 uk 3 1k cz 1 z d az 1 z 1 k 1 12T4 k 1. we can find the values of y(kT ) for k 0. and c is a value such that the poles of Y(z)z are inside the circle. (I-40) is to use the residue theorem of complex-variable theory (the details are not covered here). The result is 2z 1 11 1 e 2aT 2z 2 p 11 e kaT 2z k p (I-38) Thus. (I-1) we can clearly see that in the sampled case the coefficient of z k in Y(z) is simply y(kT). p (I-37) I-2-5 Computer Solution of the Partial-Fraction Expansion of Y (z) z Whether the function to be expanded by partial fraction is in the form of Y(z) z or Y(z). Power-Series Method The definition of the z-transform in Eq. (I-35). 1. the inverse z-transform of Y(z) is written y1kT 2 11 e a 1k 12T 2u 3 1k 12T4 k 1. One way of evaluating the contour integration of Eq. Thus. …. I-2-6 Application of the z-Transform to the Solution of Linear Difference Equations The z-transform can be used to solve linear difference equations. which can be expanded into a power series of z Y1z2 11 e aT 1 by dividing the numerator polynomial by the denominator polynomial by long division.

We have a y1k k 0 12z k a y1k2z k 0 k 0 (I-42) By using the definition of Y(z) and the real translation theorem of Eq.2832 4 2. Taking the inverse z-transform of Y(z). 1. p (I-46) 11 z 1 z 2 z 3 p 2x102 (I-45) Equation (I-41) is recognized as a single state equation.2Y1z2 U1z2 (I-49) The z-transform of u(k) is U(z) z (z 1). We have Y1z2 Thus.5y1k 12 0. The following example shows the z-transform solution of a second-order difference equation.5z 0. By this. we get y1k2 0.2832 k 0 (I-52) .2y1k2 u1k2 (I-47) The initial conditions of y(k) are: y(0) 0 and y(1) 0.07210.37 (I-51) where the exponents in the numerator coefficients are in radians. EXAMPLE I-9 Consider the second-order difference equation y1k where u1k2 u1k2 1 for k 0. we get Y1z2 z z 1 y102 (I-44) y102 4 Y1z2 0 (I-43) The inverse z-transform of the last equation can be obtained by expanding Y(z) into a power series in z 1 by long division.2832 e j 12. 2. p (I-48) 22 0.588 1.283 z 0.25 j1.283 j0.4472 k cos12.I-8 Appendix I Discrete-Data Control Systems To solve this equation. The z-transform solution of high-order discrete-data systems described by state equations is treated in Section I-3.165k 1.036e j1. (I-47).588 z 1 1. we take the z-transform on both sides of the equation.4472 k 3e j 12.588 0. (I-19) for time advance. we have Y1z2 1z z 121z2 0.036e z 0. we get 3 z2 Y1z2 z2 y102 zy112 4 0. y(k) is written y1k2 1 12 k y102 k 0. we mean that we multiply both sides of the equation by z k and take the sum from k 0 to k . Taking the z-transform on both sides of Eq.165k 1.25 j0. Substituting the initial conditions of y(k) and the expression of U(z) into Eq.03610.165k 1. 2.53zY1z2 zy102 4 0. the last equation is written z3Y1z2 Solving for Y(z). 1. (I-49) and solving for Y(z).37 1.22 (I-50) The partial-fraction expansion of Y(z) z is Y1z2 z 0.

Note that the ideal sampler is not a physical entity. can be controlled either by a controller that outputs analog signals or by a digital controller that outputs digital data. Since.I-3 Transfer Functions of Discrete-Data Systems I-9 r(t) T r*(t) ZOH h(t) Figure I-2 Sample-and-hold (S/H) device. a dc motor. and the controlled processes often contain analog components. The S/H that is most often used for the analysis of discrete-data systems consists of an ideal sampler and a zero-order-hold (ZOH) device. I-2. I-3. an interface such as a digital-to-analog (D/A) converter is necessary to couple the digital component to the analog devices. In Fig. For linear operation. the system shown in Fig. The input and output of the discrete-data system in Fig. It is used only for the representation of discrete data. Figure I-3 shows the typical operation of an ideal sampler and a ZOH. I-1 can be functionally represented by the block diagram in Fig. (c) Output signal of a zero-orderhold (ZOH) device. For instance. the D/A converter can be represented by a sample-and-hold (S/H) device. I-1 can be represented by number sequences with the numbers separated by the sampling period T. the lengths of the arrows simply represent the areas r(t) 0 (a) r*(t) t 0 10T 11T 12T 13T T 2T 3T 4T 5T 6T 7T 8T (b) 9T t h(t) 0 T 2T 3T 4T 5T 6T 7T 8T 9T (c) t Figure I-3 (a) Input signal to an ideal sampler. by definition. the arrows at the sampling instants represent impulses. It is devised simply to represent the discrete-time signal mathematically. Discrete-data control systems have the unique features that the signals in these systems either are in the form of pulse trains or are digitally coded. which is an analog device. Thus. an impulse has zero pulse width and infinite height. The output of the ideal sampler r*(t) is a train of impulses with the magnitudes of r(t) at T carried by the strengths of the impulses. In the latter case. (b) Output signal of an ideal sampler. The continuous data r(t) is sampled with a sampling period T by the ideal sampler. I-3 TRANSFER FUNCTIONS OF DISCRETE-DATA SYSTEMS • The ideal sampler is not a physical entity. . which consists of a sampler and a data-hold device.

kT. We begin by writing r*1t2 where T (t) r 1t2dT 1t2 kT 2 (I-54) is the unit-impulse train. say. S2 y*(t) S1 r(t) T r*(t) ZOH h(t) CONTROLLED PROCESS T y(t) Figure I-5 Discrete-data system with a fictitious sampler. a typical open-loop discrete-data system is modeled as shown in Fig. The following derivation is based on the Fourier-series representation of the signal r*(t). I-4. r*(t). The ZOH simply holds the magnitude of the signal carried by the incoming impulse at a given time instant. for the entire sampling period t until the next impulse arrives at t (k 1)T. As the sampling period T approaches zero. and is given by Cn where s 1 T T 0 dT 1t2e jnvs t dt (I-57) 2 T is the sampling frequency in rad/sec. dT 1t2 a d1t k (I-55) Since T(t) is a periodic function with period T. that is. its limit as T approaches zero does not have any physical meaning.I-10 Appendix I Discrete-Data Control Systems G(s) r(t) T r*(t) ZOH h(t) y(t) CONTROLLED PROCESS Figure I-4 Block diagram of a discrete-data system. under the impulses and are the magnitudes of the input signal r(t) at the sampling instants. There are several ways of deriving the transfer-function representation of the system in Fig. h(t) approaches r(t). Based on the preceding discussions. I-5. TS0 lim h1t2 r1t2 (I-53) However. the output of the ZOH. it can be expressed as a Fourier series: dT 1t2 j2pnt T a Cn e n (I-56) where Cn is the Fourier coefficient. . is an impulse train. The output of the ZOH is a staircase approximation of the input to the ideal sampler. since the output of the sampler. r(t).

The fictitious sampler S2 has the same sampling period T and is synchronized to the original sampler S1. (I-61). (I-60). we get r*1t2 1 r1t2e jnvs t T na (I-59) Taking the Laplace transform on both sides of Eq. (I-60) to y*(t). (I-63). and is called the pulsetransfer function of G(s). (I-65) are in sampled form. (I-59). and using the complex shifting property of Eq. (I-56). we get R*1s2 1 R1s T na jnvs 2 1 R1s T na jnvs 2 (I-60) Equation (I-60) represents the Laplace transform of the sampled signal r*(t). as shown in Fig. and Y*(s) all have the form of Eq. Eq. (I-63). (I-8). To overcome this problem. The Laplace transform of the system output y(t) is written Y1s2 G1s2R*1s2 (I-63) Although the output y(t) is obtained from Y(s) by taking the inverse Laplace transform on both sides of Eq. I-4. this step is difficult to execute because G(s) and R*(s) represent different types of signals. we have Y*1s2 1 G1s T na jnvs 2R*1s jnvs 2 (I-64) In view of the relationship in Eq. (I-62). where m is any integer. (I-8). and then the latter in Eq. we have R*1s jmvs 2 to . G*(s). we apply a fictitious sampler at the output of the system. where R*(s). (I-58) in Eq. (I-60). It is an alternative expression to Eq. I-5. (64) is written Y*1s2 R*1s2 1 G1s T na jnvs 2 R*1s2G*1s2 (I-65) where G*(s) is defined the same way as R*(s) in Eq. z-Transfer Function Now that all the functions in Eq. and using Eq. Cn is written Cn 1 dS0 Td lim d and a height of 1 1 T e 0 jnvs t dt dS0 lim 1 e jnvsd jnvsTd (I-58) Substituting Eq. From Eq. (2-23). The sampled form of y(t) is y*(t). Applying Eq. Since the unit impulse is defined as a pulse with a width of and d S 0. R*(s) can be written as R*1s2 a r1kT 2e 0 kTs (I-61) jm s k Since the summing limits of R*(s) range from Eq.I-3 Transfer Functions of Discrete-Data Systems I-11 . if s is replaced by s R*1s2 in (I-62) Pulse-Transfer Function Now we are ready to derive the transfer function of the discrete-data system shown in Fig. we can take the z-transform on both sides of the . (I-54).

I-12

Appendix I Discrete-Data Control Systems

y*(t) S1 r(t) R(s) T r*(t) R*(s) G1(s) d(t) D(s) T S2 d*(t) D*(s) G2(s) T Y*(s) y(t) Y(s)

(a)

T r(t) R(s) T r*(t) R*(s) G1(s) d(t) D(s) G2(s)

y*(t) Y*(s) y(t) Y(s)

(b)

Figure I-6 (a) Discrete-data system with cascaded elements and a sampler separating the two elements. (b) Discrete-data system with cascaded elements and no sampler in between.

equation by substituting z

eTs. We have Y1z2 G1z2R1z2 (I-66)

where G(z) is defined as the z-transfer function of G(s), and is given by G1z2 a g1kT2z
k 0 k

(I-67)

Thus, for the discrete-data system shown in Figs. I-5 and I-6, the z-transform of the output is equal to the z-transfer function of the process and the z-transform of the input.

I-3-1

Transfer Functions of Discrete-Data Systems with Cascade Elements
The transfer-function representation of discrete-data systems with elements connected in cascade is slightly more involved than that for continuous-data systems, because of the variation of having or not having samplers in between the elements. Figure I-6 shows two different discrete-data systems that contain two elements connected in cascade. In the system of Fig. I-6(a), the two elements are separated by the sampler S2, which is synchronized to, and has the same period as, the sampler S1. The two elements in the system of Fig. I-6(b) are connected directly together. It is important to distinguish these two cases when deriving the pulse-transfer function and the z-transfer function. For the system in Fig. I-6(a), the output of G1(s) is written D1s2 and the system output is Y1s2 G2 1s2D*1s2 G*1s2R*1s2 1 (I-69) G1 1s2R*1s2 (I-68)

Taking the pulse transform on both sides of Eq. (I-68), and using Eq. (I-62), we have D*1s2 (I-70)

I-3 Transfer Functions of Discrete-Data Systems

I-13

• The z-transform of two systems separated by a The corresponding z-transform expression of Eq. (I-71) is sampler is equal to the product of the z-transforms Y1z2 G1 1z2G2 1z2R1z2 of the two systems.

Now substituting Eq. (I-70) in Eq. (I-69) and taking the pulse transform, we get Y*1s2 G* 1s2G* 1s2R*1s2 1 2

(I-71)

(I-72)

We conclude that the z-transform of two systems separated by a sampler is equal to the product of the z-transforms of the two systems. The Laplace transform of the output of the system in Fig. I-6(b) is Y1s2 Y*1s2 where 3 G1 1s2G2 1s2 4* G1 1s2G2 1s2R*1s2 (I-73) (I-74) jnvs 2 Taking the pulse transform on both sides of the last equation, we get 3G1 1s2G2 1s2 4*R*1s2

1 G 1s T na 1

jnvs 2G2 1s

(I-75)

Notice that since G1(s) and G2(s) are not separated by a sampler, they have to be treated as one system when taking the pulse transform. Taking the z-transform on both sides of Eq. (I-74) gives Y1z2 Let Z5 3G1 1s2G2 1s2 4*6R1z2 G1G2 1z2 (I-76) (I-77)

Z5 3G1 1s2G2 1s2 4*6 Then, Eq. (I-76) is written Y1z2

G1G2 1z2R1z2

(I-78)

I-3-2

Transfer Function of the Zero-Order-Hold
Based on the description of the ZOH given earlier, its impulse response is shown in Fig. I-7. The transfer function of the ZOH is written Gh 1s2 L3gh 1t2 4 1 e s
Ts

(I-79)

Thus, if the ZOH is connected in cascade with a linear process using transfer function Gp(s), as shown in Fig. I-5, the z-transform of the combination is written G1z2
gh (t)

Z3Gh 1s2Gp 1s2 4

Za

1

e s

Ts

Gp 1s2b

(I-80)

1

0

T

t

Figure I-7 Impulse response of the ZOH.

I-14

Appendix I Discrete-Data Control Systems

By using the time-delay property of z-transforms, Eq. (I-18), Eq. (I-80) is simplified to G1z2 11 z 1 2Z a Gp 1s2 s b (I-81)

EXAMPLE I-10 Consider that for the system shown in Fig. I-5,
Gp 1s2 1 s1s 0.52 (I-82)

The sampling period is 1 second. The z-transfer function of the system between the input and the output is determined using Eq. (I-81). G1z2 11 11 b s 1s 0.52 4 2 4 z 1 2Z a 2 b s s 0.5 s 1
2

z 1 2Z a

0.426z 0.361 z2 1.606z 0.606

(I-83)

I-3-3

Transfer Functions of Closed-Loop Discrete-Data Systems
The transfer functions of closed-loop discrete-data systems are derived using the following procedures: 1. 2. 3. 4. Regard the outputs of samplers as inputs to the system. All other noninputs of the system are treated as outputs. Write cause-and-effect equations between the inputs and the outputs of the system using the SFG gain formula. Take the pulsed transform or the z-transform of the equations obtained in step 3, and manipulate these equations to get the pulse-transfer function or the z-transfer function.

Reference [1] describes the sampled signal flow graph that can be used to implement step 4 using the SFG gain formula. The following examples illustrate the algebraic procedure of finding the transfer functions of closed-loop discrete-data systems. EXAMPLE I-11 Consider the closed-loop discrete-data system shown in Fig. I-8. The output of the sampler is regarded as an input to the system. Thus, the system has inputs R(s) and E*(s). The signals E(s) and Y(s) are regarded as the outputs of the system. Writing the cause-and-effect equations for E(s) and Y(s) using the gain formula, we get E1s2 Y1s2 R1s2 G1s2H1s2E*1s2 G1s2E*1s2
y*(t) Y*(s) y(t) Y(s)

(I-84) (I-85)

T r(t) R(s) + – e(t) E(s) e*(t) E*(s) G(s)

T

H(s)

Figure I-8 Closed-loop discrete-data system.

I-9. this may not be possible for all discrete-data systems. Y*1s2 R*1s2 1 3G1s2H1s2 4* G1z2 1 GH1z2 G*1s2 (I-88) Taking the z-transform on both sides of the last equation. Taking the pulse transform on both sides of Eq. we get Y1s2 1 3G1s2H1s2 4* G1s2 R*1s2 (I-87) Taking the pulse transform on both sides of Eq. we have Y1z2 R1z2 (I-89) EXAMPLE I-12 We show in this example that although it is possible to define an input-output transfer function for the system in Fig. we get Y1s2 E1s2 G1s2E1s2 R1s2 H1s2Y*1s2 (I-90) (I-91) Taking the pulse transform on both sides of the last two equations and after simple algebraic manipulations. Writing E(s) and Y(s) in terms of the inputs using the gain formula. the pulse transform of the output is written Y*1s2 1 3G1s2R1s2 4* 3G1s2H1s2 4* (I-92) Note that the input R(s) and the transfer function G(s) are now combined as one function. and using Eq. we get R*1s2 (I-86) E*1s2 1 3G1s2H1s2 4* Substituting E*(s) from Eq. and we cannot define a transfer function in the form of Y*(s) R*(s). . Y(s) and E(s) are regarded as the outputs. In this case. which has a sampler in the feedback path. (I-62). Let us consider the system shown in Fig. The z-transform of the output is written Y1z2 GR1z2 1 GH1z2 (I-93) T r(t) R(s) + – H(s) y*(t) Y*(s) e(t) E(s) G(s) y*(t) Y*(s) y(t) Y(s) T Figure I-9 Closed-loop discrete-data system. (I-86) into Eq. we arrive at the pulsetransfer function of the closed-loop system.I-3 Transfer Functions of Discrete-Data Systems I-15 Notice that the right-hand side of the last two equations contains only the inputs R(s) and E*(s) and the transfer functions. (I-60) and solving for E*(s). (I-87). the outputs of the sampler Y*(s) and R(s) are the inputs of the system. I-8. (I-85). [G(s)R(s)]*.

C. as defined in Eq. the algebraic method may become tedious. (5-44). we let t (k 1)T and t0 kT. Typical signals that appear at various points in the system are shown in the figure. we can write h1t2 h1kT 2 r 1kT 2 (I-94) for kT t (k 1)T. but because of the discrete-time data. the differential equations are discretized to yield a set of difference equations. I-10. when dealing with discrete-data systems. The first one is that the system contains continuous-data components. respectively. I-4-1 Discrete State Equations Let us consider the discrete-data control system with an S/H device. As described earlier.I-16 Appendix I Discrete-Data Control Systems Although we have been able to arrive at the input-output transfer function and transfer relation of the systems in Figs. The matrices A. the state transition equation is x1t2 F1t t0 2x1t0 2 t F1t t0 t2Bh1t2dt (I-97) for t t0. for more complex system configurations. Equation (I-98) is written x3 1k or 12T4 F1T 2x1kT 2 x3 1k 12T4 1k kT 12T F3 1k 12T t4Bdt h1kT 2 (I-99) (I-100) F1T 2x1kT 2 U1T 2h1kT 2 . 1. the modern way of modeling a discrete-data system is by discrete state equations. k 0. I-4 STATE EQUATIONS OF LINEAR DISCRETE-DATA SYSTEMS Just as for continuous-data systems. and D are coefficient matrices. Now we let the linear process G be described by the state equation and output equation: dx1t2 dt y1t2 Ax1t2 Cx1t2 Bh1t2 Dh1t2 (I-95) (I-96) where x(t) is the n n state vector. and h(t) and y(t) are the scalar input and output. B. we often encounter two situations. By using Eq. I-8 and I-9 by algebraic means without difficulty. (I-98) can be taken outside of the integral sign. h(t). the input h( ) in Eq. 2. (I-95). In this case. is a sequence of steps. Then Eq. Therefore. The second situation involves systems that are completely discrete with respect to time. as shown in Fig. …. (I-97) becomes x 3 1k 12T4 F1T2x1kT 2 1k 12T kT F3 1k 12T t4Bh1t2dt (I-98) Since h(t) is piecewise constant. The signalflow graph method may be extended to the analysis of discrete-data systems. and the system dynamics should be difference equations from the outset. the reader may refer to [1] for details. the components of the system are still described by differential equations. The output of the S/H. but the signals at certain points of the system are discrete with respect to time because of sample-and-hold (S/H) operations. The output signal y(t) ordinarily is a continuous-data signal. If we are interested only in the responses at the sampling instants.

2. (I-100) can be solved by using a simple recursion procedure.I-4 State Equations of Linear Discrete-Data Systems I-17 r(t) T r*(t) ZOH h(t) G y(t) r (t) 0 t r *(t) 0 T 2T 3T 4T 5T 6T 7T t h(t) 0 T 2T 3T 4T 5T 6T 7T t y(t) 0 t Figure I-10 Discrete-data system with sample-and-hold (S /H). By setting k 0. (I-100). … successively in Eq. and is referred to as the vector-matrix discrete state equation. the following . I-4-2 Solutions of the Discrete State Equations: Discrete State-Transition Equations The discrete state equations represented by Eq. 1. where u1T2 1k kT 12T F3 1k T 12T t4Bdt 0 F1T t2Bdt (I-101) Equation (I-100) is of the form of a set of linear first-order difference equations in vectormatrix form.

I-18

Appendix I Discrete-Data Control Systems

equations result: k k k o k n 1: x1nT2 0: 1: 2: x1T2 x12T2 x13T2 o F1T2x102 U1T2h102 F12T2 F1T2x1T2 U1T2h1T2 F13T2 F1T2x12T2 U1T2h12T2 F1T2x3 1n 12T4 U1T2h3 1n 12T4 (I-102) (I-103) (I-104) (I-105)

Substituting Eq. (I-102) into Eq. (I-103), then Eq. (I-103) into Eq. (I-104), and so on, we obtain the following solution for Eq. (I-100): x1nT2 Fn 1T2x102
n 1

aF
i 0

n

i

1

1T2u1T2h1iT2

(I-106)

(nT). where, from Eq. (5-36), n(T) [ (T)]n Equation (I-106) is defined as the discrete state-transition equation of the discretedata system. It is interesting to note that Eq. (I-106) is analogous to its continuous-data counterpart in Eq. (5-41). The state-transition equation of Eq. (I-97) describes the state of the system of Fig. I-10 for all values of t, whereas the discrete state-transition equation in Eq. (I-106) describes the states only at the sampling instants t nT, n 0, 1, 2, …. With nT considered as the initial time, where n is any positive integer, the statetransition equation is x3 1n N2T4 FN 1T2x1nT2
N 1 N aF i 0 i 1

1T2U1T2h3 1n

i2T4

(I-107)

where N is a positive integer. The output of the system at the sampling instants is obtained by substituting t and Eq. (I-106) into Eq. (I-96), yielding y1nT2 Cx1nT2 Dh1nT2 C a F3 1n
i 0 n 1

nT

CF1nT2x102

i

12T4U1T2h1iT2

Dh1nT2

(I-108)

An important advantage of the state-variable method over the z-transform method is that it can be modified easily to describe the states and the output between sampling instants. In Eqs. (I-97), if we let t (n )T, where 0 1 and t0 nT, we get x 3 1n ¢2T4 F1¢T2x1nT2 F1¢T2x1nT2
1n ¢2T

nT

F3 1n

¢2T

t4Bdth1nT2 (I-109)

U1¢T2h1nT2

By varying the value of between 0 and 1, the information on the state variables between the sampling instants is completely described by Eq. (I-109). When a linear system has only discrete data through the system, its dynamics can be described by a set of discrete state equations: x 3 1k and output equations: y1kT2 Cx1kT2 Dr1kT2 (I-111) 12T4 Ax1kT2 Br1kT2 (I-110)

where A, B, C, and D are coefficient matrices of the appropriate dimensions. Notice that Eq. (I-110) is basically of the same form as Eq. (I-100). The only difference in the two

I-4 State Equations of Linear Discrete-Data Systems

I-19

situations is the starting point of system representation. In the case of Eq. (I-100), the starting point is the continuous-data state equations of Eq. (I-95); (T) and (T) are determined from the A and B matrices, and must satisfy the conditions and properties of the state transition matrix. In the case of Eq. (I-110), the equation itself represents an outright description of the discrete-data system, and there are no restrictions on the matrices A and B. The solution of Eq. (I-110) follows directly from that of Eq. (I-100), and is
n 1

x1nT2 where

Anx102

aA
i 0

n

i

1

Br1iT2

(I-112)

An

AAAA p A 0d n S0

(I-113)

I-4-3 z-Transform Solution of Discrete State Equations
In Section I-1-5, we illustrated the solution of a simple discrete state equation by the ztransform method. In this section, the discrete state equations in vector-matrix form of an nth-order system are solved by z-transformation. Consider the discrete state equations x 3 1k zX1z2 1zI 12T4 Ax1kT2 Br1kT2 (I-114)

Taking the z-transform on both sides of the last equation, we get zx102 AX1z2 1zI BR1z2 (I-115)

Solving for X(z) from Eq. (I-115), we get X1z2 A2
1

zx102

A2

1

BR1z2 BR1z2 4

(I-116)

Taking the inverse z-transform on both sides of Eq. (I-116), we have x1nT2 Z 1 3 1zI A2
1

z4x102

Z 1 3 1zI

A2

1

(I-117)

In order to carry out the inverse z-transform operation of the last equation, we write the z-transform of An as Z1An 2
n aAz n 0 n

I

Az

1

A2z

2

p

(I-118)

Premultiplying both sides of Eq. (I-118) by Az 1 and subtracting the result from the last equation, we get 1I Az 1 2Z1An 2 I (I-119) Therefore, solving for Z(An) from the last equation yields Z1An 2 or An 1I Az 1 2 Z 1 3 1zI
1

1zI A2 1z4

A2 1z

(I-120) (I-121)

Equation (I-121) represents a way of finding An by using the z-transform method. Similarly, we can prove that Z 1 3 1zI A2
1

BR1z2 4

n i

1 n i 1

aA
0

Br1iT2

(I-122)

I-20

Appendix I Discrete-Data Control Systems

Now we substitute Eqs. (I-121) and (I-122) into Eq. (I-117), x(nT) becomes
n 1 n i 1

x1nT2 which is identical to Eq. (I-112).

Anx102

aA
i 0

Br1iT2

(I-123)

I-4-4

Transfer-Function Matrix and the Characteristic Equation
Once a discrete-data system is modeled by the dynamic equations of Eqs. (I-110) and (I-111), the transfer-function relation of the system can be expressed in terms of the coefficient matrices. By setting the initial state x(0) to zero, Eq. (I-116) becomes X1z2 3C1zI G1z2 or G1z2 C3adj1zI 1zI A2
1

A2

1

BR1z2

(I-124)

Substituting Eq. (I-124) into the z-transformed version of Eq. (I-111), we have Y1z2 B D4R1z2 G1z2R1z2 (I-125)

where the transfer-function matrix of the system is defined as C1zI A2
1

B

D A 0 D4

(I-126)

A2B 0zI 0zI A 0 A0 0

(I-127)

The characteristic equation of the system is defined as 0zI y3 1k (I-128)

In general, a linear time-invariant discrete-data system with one input and one output can be described by the following difference equation with constant coefficients: n2T4 an 1y3 1k n 12T4 an 2 y3 1k n p a1y3 11k 122T4 a0y1kT2 bm 1r3 1k m 12T4 bmr3 1k m2T4 p b1r3 11k 122T4 b0r 1kT2 n m 22T4 (I-129)

Taking the z-transform on both sides of Eq. (I-129) and setting zero initial conditions, the transfer function of the system is written Y1z2 bmz m bm 1z m 1 p b1z b0 n m (I-130) R1s2 z n an 1z n 1 p a1z a0 The characteristic equation is obtained by equating the denominator polynomial of the transfer function to zero. zn an 1z n
1

p

a1z

a0

0

(I-131)

EXAMPLE I-13 Consider that a discrete-data system is described by the difference equation
y 1k 22 5y 1k Y1z2 R1z2 z
2

12

3y 1k2 z 5z 2

r 1k

12

2r 1k2

(I-132)

The transfer function of the system is simply 3 (I-133)

. the elements of a discrete state diagram resemble the computing elements of a digital computer. a discrete state diagram may be constructed for the system. Shifting or time delay: x2 1kT2 X2 1z2 or X1 1z2 z 1X2 1z2 x1 102 (I-146) x1 3 1k zX1 1z2 12T4 zx1 102 (I-144) (I-145) x1 1kT2 X1 1z2 x3 1kT2 X3 1z2 (I-142) (I-143) ax1 1kT2 aX1 1z2 (I-140) (I-141) The state diagram representation of these operations are illustrated in Fig. The mathematical description of these basic digital computations and their corresponding z-transform expressions are as follows: 1. Multiplication by a constant: x2 1kT2 X2 1z2 2. Then the equations represent the discrete-time state transition from t t0. Similar to the relations between the analog-computer block diagram and the state diagram for continuous-data systems. The initial time t 0 in Eqs. I-11. Some of the operations of a digital computer are multiplication by a constant.I-4 State Equations of Linear Discrete-Data Systems The characteristic equation is z2 5z 3 0 I-21 (I-134) The state variables of the system may be defined as x1 1k2 x2 1k2 x1 1k x2 1k y1k2 x1 1k r 1k2 (I-135) (I-136) 12 Substituting the last two equations into Eq. (I-134) is obtained by using 0zI I-4-5 State Diagrams of Discrete-Data Systems When a discrete-data system is described by difference or discrete state equations. (I-139) The same characteristic equation as in Eq. and time delay or shifting. addition of several variables. (I-132) gives the two state equations as 12 12 x2 1k2 r 1k2 3x1 1k2 5x2 1k2 3r 1k2 (I-137) (I-138) from which we have the matrices A and B: A c 0 3 1 d 5 B c 1 d 3 A0 0. (I-145) and (I-146) can be generalized to t t0. The discrete state diagram can be used to determine the transfer functions as well as for digital implementation of the system. Summing: x2 1kT2 X2 1z2 3.

(I-132).I-22 Appendix I Discrete-Data Control Systems a X1(z) X2(z) = aX1(z) X0(z) 1 X2(z) X2(z) 1 X1(z) X2(z) = X0(z) + X1(z) x1(0) 1 z–1 X2(z) X1(z) = z X2(z) + x1(0) –1 X1(z) Figure I-11 Basic elements of a discrete state diagram. which is y1k 22 5y1k 12 3y1k2 r 1k 12 2r 1k2 (I-147) • The state variables are defined as the outputs of the delay units in the state diagram. One way of constructing the discrete state diagram for the system is to use the state equations. (I-137) and (I-138) is shown in Fig. . EXAMPLE I-14 Consider again the difference equation in Eq. In this case. The state-transition equations of the system can be obtained directly from the state diagram using the SFG gain formula. I-12. (I-132) or by the state equations of Eqs. the state equations are already defined in Eqs. By using essentially the same principle as the state diagram for continuous-data systems. (I-137) and (I-138). I-12. x2(0). the state diagram for Eqs. The state variables are defined as the outputs of the delay units in the state diagram. and R(z) as input nodes in Fig. the state-transition equations are written as c X1 1z2 d X2 1z2 1 1 5z 1 z 1 x1 102 c dc d x2 102 ¢ 3z 1 1 1 1 z 1 11 5z 1 2 3z c ¢ 3z 1 3z 2 2 d R1z2 (I-148) x2(0) 1 –3 r(k) R(z) x2(k + 1) –5 z–1 x2(k) X2(z) 1 x1(k + 1) z –1 x1(0) 1 1 x1(k) X1(z) y(k) Y(z) –3 Figure I-12 Discrete state diagram of the system described by the difference equation of Eq. (I-137) and (I-138). The time delay unit z 1 is used to relate x1(k 1) to x2(k). By referring to X1(z) and X2(z) as the output nodes and x1(0).

3 (b) Cascade decomposition x1(0) z –1 1 2 X2(z) – 0. .I-4 State Equations of Linear Discrete-Data Systems I-23 x2(0) 1 x1(0) 1 1 R(z) –5 –3 (a) Direct decomposition z –1 1 z X2(z) –1 2 X1(z) Y(z) 1 x2(0) x1(0) 1 1 R(z) – 4.7 z –1 1 X1(z) Y(z) 1 z–1 1 –4. 2 R1z2 1z 5z 32 where ¢ 1 5z 1 3z 2 (I-149) The same transfer function between R(z) and Y(z) as in Eq.36 X1(z) x2(0) Y(z) 0. using the decomposition schemes (Fig. (I-133) can be obtained directly from the state diagram in Fig. and so the details are not repeated here. I-13 by applying the SFG gain formula between these two nodes. the discrete state diagram can be drawn directly from the difference equation via the transfer function. As an alternative. The decomposition of a discrete-data transfer function follows basically the same procedure as that of an analog transfer function covered in Section 5-9.64 Figure I-13 State diagrams of the transfer function Y1z2 1z 22 . I-13).3 R(z) 1 1 z–1 X2(z) –0.7 (c) Parallel decomposition 0.

The state-diagram representation of the zero-order-hold is shown EXAMPLE I-15 As an illustrative example on how the state diagram of a sampled-data system is constructed. Y1s2 1 e s Ts 1 s 1 E*1s2 (I-153) Taking the z-transform on both sides of Eq. I-15. let us consider the system in Fig. (I-153). which is a train of impulses. and the output by h(t). a state diagram model for the sample-and-hold (zero-order-hold) must be established. the relation is simply h1t2 e1kT 2 (I-151) for kT t (k 1)T.I-24 Appendix I Discrete-Data Control Systems 1 s H(s) e(kT ) + Figure I-14 State-diagram representation of the zero-order-hold (ZOH). The discrete dynamic equations of the system are written directly from the state diagram. the signal h(t) is a sequence of steps. e1kT 2 s (I-152) 1)T. (I-154). x1 3 1k 12T4 y1kT 2 e Tx1 1kT 2 x1 1kT 2 11 e T 2e1kT 2 (I-155) (I-156) e(t) T e*(t) ZOH h(t) G(s) = 1 s+1 y(t) Figure I-15 Sampled-data system. The input-output relation in the Laplace domain is H1s2 1 e s Ts E*1s2 (I-150) In the time domain. For this purpose. we need the relation between H(s) and e(kT ). Since the ZOH simply holds the strength of the input impulse at the sampling instant until the next input comes. In the state-diagram notation. . I-14. Consider that the input of the ZOH is denoted by e*(t). We shall demonstrate the various ways of modeling the inputoutput relations of the system. with the two types of elements separated by sample-and-hold devices. (I-151) to give H1s2 for kT t (k in Fig. First. we take the Laplace transform on both sides of Eq. the Laplace transform of the output of the system is written in terms of the input to the ZOH. I-4-6 State Diagrams for Sampled-Data Systems When a discrete-data system has continuous-data as well as discrete-data elements. we get Y1z2 1 z e e T T E1z2 (I-154) Figure I-16 shows the state diagram for Eq.

zero-input stability can also be referred to as asymptotic stability. no multiple-order roots on the unit circle. that the roots of the characteristic equation lie inside the unit circle 0z 0 This is not surprising. i 1. …. As with the treatment in Section 6-1. the output sequence of the system must satisfy the following conditions: 1. …. 2. output. I-17. The possible stability conditions of the system are summarized in Table I-2 with respect to the roots of the characteristic equation. and g(kT ) be the input. i 1. 2. y(kT). or simply stable. the system is said to be BIBO stable. n (all roots inside the unit circle) 0 zi 0 1 for any i for simple roots. I-5-1 BIBO Stability Let u(kT ). 0y1kT2 0 M 6 lim 0y1kT2 0 0 (I-158) (I-159) kS Thus. and impulse sequence of a linear timeinvariant SISO discrete-data system. With zero initial conditions. I-15. 2. Let the characteristic equation roots of a linear discrete-data timeinvariant SISO system be zi. the following condition must be met: a 0g1kT2 0 6 0 (I-157) k I-5-2 Zero-Input Stability For zero-input stability. The regions of stability and instability for discrete-data systems in the z-plane are shown in Fig. 2. …. if its output sequence y(kT) is bounded to a bounded input u(kT). 2. we can show that for the system to be BIBO stable. and no 0 zi 0 1 for i 1. i 1. TABLE I-2 Stability Conditions of Linear Time-Invariant Discrete-Data SISO Systems Root values 0 zi 0 1 for all i. since the j -axis of the s-plane is mapped onto the unit circle in the z-plane. or 0 zi 0 1 for any multiple-order root. n (at least one simple root. n. and no roots outside the unit circle) 0 zi 0 1 for any i.I-5 Stability of Discrete-Data Systems x1(0) 1 1 – e –T x1(k + 1)T e(kT +) –e –T I-25 z–1 x1(kT) X1(z) Figure I-16 Discrete state diagram of the system in Fig. I-5 STABILITY OF DISCRETE-DATA SYSTEMS The definitions of BIBO and zero-input stability can be readily extended to linear timeinvariant SISO discrete-data control systems. …. n (at least one simple root outside the unit circle and at least one multiple-order root on the unit circle) Stability condition Asymptotically stable or simply stable Marginally stable or marginally unstable Unstable . We can show that both the BIBO stability and the zero-input stability of discrete-data systems require 1 in the z-plane. respectively.

.221z 0. originally devised for continuous-data systems. which are the characteristic equation roots. The Nyquist criterion.82 51z 1. can all be extended to the stability studies of discrete-data systems.82 5z 1. Bilinear Transformation Method [1] We can still apply the Routh-Hurwitz criterion to discrete-data systems if we can find a transformation that transforms the unit circle in the z-plane onto the imaginary axis of another complex plane. One exception is the RouthHurwitz criterion.I-26 Appendix I Discrete-Data Control Systems jIm z z-plane Unstable Unstable Unit circle Stable –1 0 Stable 1 Re z Unstable Unstable Figure I-17 Stable and unstable regions for discrete-data systems in the z-plane. and thus can be applied only to continuous-data systems. which in its original form is restricted to only the imaginary axis of the s-plane as the stability boundary.12 Stable system Unstable system due to the pole at z Marginally stable due to z 1 1 1.82 51z 12 121z 0. We cannot use the z-transform relation z exp(Ts) or s (In z) T. and Bode diagram. and the stability condition of the system. root-locus diagram.22 12 2 1z 0.2 Unstable due to second-order pole at z I-5-3 Stability Tests of Discrete-Data Systems We pointed out in Section I-5 that the stability test of a linear discrete-data system is essentially a problem of investigating whether all the roots of the characteristic equation are inside the unit circle 0z 0 1 in the z-plane.221z 0. EXAMPLE I-16 M1z2 M1z2 M1z2 M1z2 1z 1z z1z z2 1z 5z 0. The following example illustrates the relationship between the closed-loop transferfunction poles.

d are real constants. if computer programs are available for the transformations. we substitute z into Eq. the difference is insignificant. the relationship between w and . Once the characteristic equation in z is transformed into the r domain using Eq. and we get w 2 cos wT T cos vT j sin vT j sin vT 1 1 (I-165) e jvT cos vT j sin vT (I-164) Rationalizing the last equation. Another transformation that is often used in discrete-data control-system design in the frequency domain is z or w 2z Tz 1 1 (I-163) 12 T2 12 T2 w w (I-162) which is called the w-transformation. (I-163). The r-transformation given in Eq. there are many bilinear transformations of the form of ar b z (I-160) cr d where a. and r is a complex variable.I-5 Stability of Discrete-Data Systems I-27 since it would transform an algebraic equation in z into a nonalgebraic equation in s. However. and the Routh test still cannot be applied. the unit circle in the z-plane is mapped onto the imaginary axis w w-plane. of course. . (I-161) is probably the simplest form that can be used for manual transformation of an equation F(z) to an equation in r. b. especially since the sampling period T appears in Eq. that will transform circles in the z-plane onto straight lines in the r-plane. One such transformation that transforms the interior of the unit circle of the z-plane onto the left half of the r-plane is z 1 1 r r (I-161) which is referred to as the r-transformation. the real frequency. The correlation between and w is that they both go to 0 and at the same time. the Routh-Hurwitz criterion can again be applied to the equation in r. the w-transformation is more difficult to use. The advantage of the w-transformation over the rtransformation is that the imaginary axis of the w-plane resembles that of the s-plane. (I-163). For Routh-Hurwitz criterion. is vw 2 vT tan T 2 vs pv tan p vs in the (I-167) where s is the sampling frequency in rad/sec. Note that the w-transformation becomes the r-transformation when T 2. we get w jvw 2 vT j tan T 2 j (I-166) w Thus. However. To show this. and simplifying. (I-161). c.

We can show that these numbers cannot be all negative. EXAMPLE I-17 Consider that the characteristic equation of a discrete-data control system is z3 5.146 14.0. z 3. Eq. EXAMPLE I-18 Let us consider a design problem using the bilinear transformation and Routh-Hurwitz criterion. (I-161) into the last equation and simplifying. This corresponds to Eq.0461. the roots are: z 2.128r 3 11. we have the following conditions: 1 K 7 0 1 3K 7 0 K 7 0 3 K 7 0 which lead to the condition for stability: 0 6 K 6 1 (I-172) Direct Stability Tests There are stability tests that can be applied directly to the characteristic equation in z with reference to the unit circle in the z-plane. we get 3.128 2. The characteristic equation of a linear discrete-data control system is given as F1z2 z3 z2 z K 0 (I-170) where K is a real constant.984.344r 14.67. and r 0.27 0 (I-169) Routh’s tabulation of the last equation is r3 Sign change r2 Sign change r1 r0 6. (I-161).7z 0.27 3. One of the first methods that gives the necessary . We first transform F(z) into an equation in r using the bilinear transformation of Eq.9117. since the conditions contradict each other. The result is 11 K2r 3 r3 r2 r1 r0 11 3K2r 2 311 K2r 3 K 0 (I-171) Routh’s tabulation of the last equation is 1 K 1 3K 8K11 K2 1 3 K 3K 3(1 3 0 K) K For a stable system. The problem is to find the range of values of K so that the system is stable. and z 0. the numbers in the first column of the tabulation must be of the same sign. (I-168).I-28 Appendix I Discrete-Data Control Systems The following examples illustrate the application of the r-transformation to a characteristic equation in z so that the equation can be tested by Routh-Hurwitz criterion in the r-domain.344 Since there are two sign changes in the first column of the tabulation. respectively.74r 2 2. The three corresponding roots in the r-plane are: r 3. For Eq. (I-169) has two roots in the right half of the r-plane.27 0 11.0.74 14.368 0 (I-168) Substituting Eq.94z2 7. (I-168) having two roots outside the unit circle in the z-plane. r 1. Next. for all the numbers to be positive. This result can be checked by solving the two equations in z and r.

I-5 Stability of Discrete-Data Systems

I-29

and sufficient conditions for the characteristic equation roots to lie inside the unit circle is the Schur-Cohn criterion [2]. A simpler tabulation method was devised by Jury and Blanchard [3, 4] and is called Jury’s stability criterion [6]. R. H. Raible [5] devised an alternate tabular form of Jury’s stability test. Unfortunately, these analytical tests all become very tedious for equations higher than the second order, especially when the equation has unknown parameter(s) in it. Then, there is no reason to use any of these tests if all the coefficients of the equation are known, since we can always use a root-finding program on a computer. Weighing all the pros and cons, this author believes that when the characteristic equation has at least one unknown parameter, the bilinear transformation method is still the best manual method for determining stability of linear discrete-data systems. However, it is useful to introduce the necessary condition of stability that can be checked by inspection. Consider that the characteristic equation of a linear time-invariant discrete-data system is F 1z2 anz n an 1z n
1

p

a1z

a0

0

(I-173)

where all the coefficients are real. Among all the conditions provided in Jury’s test, the following necessary conditions must be satisfied for F(z) to have no roots on or outside the unit circle. F112 7 0 F1 12 7 0 F1 12 6 0 0a0 0 6 an if n if n even integer odd integer

(I-174)

If an equation of the form of Eq. (I-173) violates any one of these conditions, then not all of the roots are inside the unit circle, and the system would not be stable. Apparently, these necessary conditions can be checked easily by inspection. EXAMPLE I-19 Consider the equation
F1z2 F 112 0a0 0 z3 z2 0.5z 0.25 0 (I-175)

Applying the conditions in Eq. (I-174), we have 2.75 7 0 0.25 6 a3 and 1 F 1 12 0.25 6 0 for n 3, which is odd

Thus, the conditions in Eq. (I-174) are all satisfied, but nothing can be said about the stability of the system.

EXAMPLE I-20 Consider the equation
F 1z2 The conditions in Eq. (6-58) are for n 3, which is odd F1 12 0.75 7 0 0a0 0 1.25, which is not less than a3, which equals 1. Since for odd n F( 1) must be negative, the equation in Eq. (I-176) has at least one root outside the unit circle. The condition on the absolute value of a0 is also not met. z3 z2 0.5z 1.25 0 (I-176)

Second-Order Systems The conditions in Eq. (I-174) become necessary and sufficient when the system is of the second order. That is, the necessary and sufficient conditions for the second-order equation F1z2 a2z2 a1z a0 0 (I-177)

I-30

Appendix I Discrete-Data Control Systems

to have no roots on or outside the unit circle are F112 7 0 F1 12 7 0 0a0 0 6 a2 EXAMPLE I-21 Consider the equation
F 1z2 F112 0a0 0 2.25 7 0 0.25 6 a2 F 1 12 1 z2 z 0.25 0 (I-179)

(I-178)

Applying the conditions in Eq. (I-178), we have 0.25 7 0 for n 2, which is even

Thus, the conditions in Eq. (I-178) are all satisfied. The two roots in Eq. (I-179) are all inside the unit circle, and the system is stable.

I-6 TIME-DOMAIN PROPERTIES OF DISCRETE-DATA SYSTEMS
I-6-1 Time Response of Discrete-Data Control Systems
To carry out the design of discrete-data control systems in the time domain or the zdomain, we must first study the time- and z-domain properties of these systems. We learned from the previous sections that the output responses of most discrete-data control systems are functions of the continuous-time variable t. Thus, the time-domain specifications such as the maximum overshoot, rise time, damping ration, and so forth, can still be applied to discrete-data systems. The only difference is that in order to make use of the analytical tools such as z-transforms, the continuous data found in a discretedata system are sampled so that the independent time variable is kT, where T is the sampling period in seconds. Also, instead of working in the s-plane, the transient performance of a discrete-data system is characterized by poles and zeros of the transfer function in the z-plane. The objectives of the following sections are as follows: To present methods of finding the discretized time responses of discrete-data control systems 2. To describe the important characteristics of the discretized time response y(kT) 3. To establish the significance of pole and zero locations in the z-plane 4. To provide comparison between time responses of continuous-data and discretedata control systems Let us refer to the block diagram of the discrete-data control system shown in Fig. I-18. The transfer function of the system is Y1z2 R1z2 1 G1z2 GH1z2 (I-180) 1.

where GH(z) denotes the z-transform of G(s)H(s). Once the input R(z) is given, the output sequence y(kT) can be determined using one of the following two methods: 1. Take the inverse z-transform of Y(z) using the z-transform table. 2. Expand Y(z) into a power series of z k.

I-6 Time-Domain Properties of Discrete-Data Systems

I-31

y*(t) T r(t) R(s) + – e(t) E(s) T e*(t) E*(s) ZOH Gp(s) G(s) Y(z)

y(t) Y(s)

Figure I-18 Block diagram of a discretedata control system.

H(s)

The z-transform of the output is defined as Y1z2 a y1kT2z
k 0 k

(I-181)

The discrete-time response y(kT ) can be determined by referring to the coefficient of z k for k 0, 1, 2, . . . . Remember that y(kT), k 0, 1, 2, . . . contains only the sampled information on y(t) at the sampling instants. If the sampling period is large relative to the most significant time constant of the system, y(kT) may not be an accurate representation of y(t). EXAMPLE I-22 Consider that the position-control system described in Section 7-7 has discrete data in the forward
path, so that the system is now described by the block diagram of Fig. I-18. For K fer function of the controlled process is Gp 1s2 s 1s 65,250 361.22 Gp 1s2 s d 14.5, the trans-

(I-182)

The forward-path transfer function of the discrete-data system is Gh0 Gp 1z2 For a sampling period of T Z 3 Gh0 Gp 1s2 4 11 z 1 2Z c

(I-183)

0.001 second, the z-transfer function in Eq. (I-183) is evaluated as Gh0Gp 1z2 Gh0Gp 1z2 z2 0.029z 0.0257 1.697z 0.697 (I-184)

The closed-loop transfer function of the system is Y1z2 R1z2 1 Gh0Gp 1z2 z2 0.029z 0.0257 1.668z 0.7226 (I-185)

where R(z) and Y(z) represent the z-transforms of the input and the output, respectively. For a unitstep input, R(z) z (z 1). The output transform Y(z) becomes Y1z2 1z z10.029z 121z 2 0.02572 0.72262 1.668z (I-186)

The output sequence y(kT ) can be determined by dividing the numerator polynomial of Y(z) by its denominator polynomial to yield a power series in z 1. Figure I-20 shows the plot of y(kT) (dots) versus kT, when T 0.001 second. For comparison, the unit-step response of the continuous-data system in Section 7-6 with K 14.5 is shown in the same figure. As seen in Fig. I-19, when the sampling period is small, the output responses of the discrete-data and the continuous-data systems are very similar. The maximum value of y(kT ) is 1.0731, or a 7.31 percent maximum overshoot, as against the 4.3 percent maximum overshoot for the continuous-data system.

• When one or more characteristic equation roots lie on the negative axis in the z-plane.3712 Discrete-data system y(kT )T = 0. the forward-path transfer function of the discrete-data system is Gh0Gp 1z2 and the closed-loop transfer function is Y1z2 R1z2 • Two equal characteristic equation roots on the negative real axis in the z-plane do not correspond to critical damping. the discrete-data system will be unstable. 3. The root at 1. when one or more characteristic-equation roots lie on the negative real axis of the z-plane. For the discrete-data system. 1.01 second 1. the system response will oscillate with positive and negative peaks.2 uy(t). For discrete-data systems.027 (I-187) (I-188) The output sequence y(kT) with T 0. From Section 7-6.01608 second.03 0.000.148 Figure I-19 Comparison of unit-step responses of discrete-data and continuous-data systems.01658 second. one root will move outside the unit circle.04 0. When the sampling period is increased to 0. Thus.6 y(T) 1. .001 second y(5 T ) 1.027z 0. and if the value of T is too large. 0 0. the two roots become equal and real and are negative. but the true maximum overshoot is considerably higher than that. the two characteristic-equation roots are very close to the z 1 point and are complex.9028 y(2T ) 1.01 second is shown in Fig.000 causes the step response of the system to oscillate with a constant amplitude.8 uy(t) y(3 T ) 0. 2. Figure I-21 shows the oscillatory response of y(kT) when T 0. Thus. for all sampling periods greater than 0. I-19 with k 0.4379 z 0.01658 second.3198z 0.0 0.01 0.01658 second.3198z 0.494 and z 1. and the system will become unstable. but the sampled output no longer gives an accurate measure of the true output. the second-order system can become unstable. Notice that the maximum value of y(kT) is 1.01 second.3712. which is the critical value for stability. response will oscillate with positive and negative peaks.3198 1. 1.7426 0.02 Time(sec) 0.4939 0 (I-189) which has roots at z 0. Unlike the continuous-data system. the sample-and-hold has the effect of making the system less stable. Notice that when the sampling period is very small.4649 2 1.05 When the sampling period is increased to 0. 4. and 5. the case of two identical roots on the negative real axis in the z-plane does not correspond to critical damping.I-32 Appendix I Discrete-Data Control Systems 2. y(kT) 1. Beyond this value of T.0 Discrete-data system T = 0.4938z 0. the characteristic equation of the discrete-data system is z2 1. we learned that the second-order continuous-data system is always stable for finite positive values of K. The true continuous-time output of the discrete-data system is shown as the dotted curve. Figure I-20 shows the trajectories of the two characteristic-equation roots of the discrete-data system as the sampling period T varies. the larger sampling period only makes the system less stable.4 y(4 T ) 0.4379 z 2 1. When T 0. and the system is marginally stable.2929z 0.

I-6 Time-Domain Properties of Discrete-Data Systems j Im z z-plane Unstable region Unstable region I-33 T = 0.666 T = 0.0 .1464 + j0.01608 second T = 0.834 ± j0.0 1. Unstable region Unstable region 3.494 T = 0.666 Unit circle Figure I-20 Trajectories of roots of a second-order discrete-data control system as the sampling period T varies.5 y(kT) 1.0 2.01658 second 1 Re z T = 0.01658 second –1 0 –0.001 second 0.1649 T = 0.01 second –0.5 0 0 3 6 9 12 15 18 21 24 27 30 Figure I-21 Oscillatory response of a discrete-data system with a sampling period T 0.01658 second.5 2.1464 – j0. Number of sampling periods –0.5 –1.0 0.01 second – 0.

given any point s1 in the s-plane. In Section I-2. . I-22(a). . that is. it is important to study the relation between the location of the characteristic-equation roots in the z-plane and the time response of the discrete-data system. . –1 s = – s jvs / 2 (s > 0) s = –s < 0 1 s=s>0 Re z (b) Figure I-22 Periodic strips in the s-plane and the corresponding points and lines between the s-plane and the z-plane. m = 1. In other words. (I-62). 2. as shown in Fig. the function R*(s) has the same value at all periodic points s s1 jm s. the s-plane is divided into an infinite number of periodic strips.I-34 Appendix I Discrete-Data Control Systems I-6-2 Mapping between s-Plane and z-Plane Trajectories For analysis and design purposes. The strip between jv s-plane j5vs / 2 Complementary strip j 2vs j3vs / 2 Complementary strip jvs jvs / 2 Primary strip 0 – jvs / 2 s Complementary strip – jvs – j3vs / 2 Complementary strip –j 2vs – j5vs / 2 (a) j Im z z-plane s = jv s = – s jvs / 2 (s < 0) s= jvs / 2 s=– s = 0. . R*(s jm s) R*(s). Thus. jmvs. the periodic property of the Laplace transform of the Laplace transform of the sampled signal R*(s) is established by Eq. where m is an integer.

The origin.I-6 Time-Domain Properties of Discrete-Data Systems I-35 s 2 is called the primary strip. The same parameters can be defined for discrete-data systems with respect to the characteristic-equation roots in the z-plane. The loci of the constant. 4. jv s-plane Unit circle jIm z z-plane ea1T –a2 a1 e–a2T 0 s –1 0 1 Re z Constantdamping loci (a) (a) Figure I-23 Constant-damping loci in the s-plane and the z-plane. s 0. and constant. The portion inside the unit circle corresponds to 0. 2. I-24. m 0.n in the z-plane are described in the following sections. and the portion outside the unit circle corresponds to 0. and the details are explained as follows. The portion inside the unit circle corresponds to 0. The boundaries of the period strips. Constant-Damping Loci: For a constant-damping factor responding trajectory in the z-plane is described by z eaT in the s-plane. z 0. 3. and the natural undamped frequency n to characterize the system dynamics. . are mapped onto the positive real axis of the z-plane. The center lines of the periodic strips. 6. the damping ratio . 3. The corresponding z-plane locus is a straight line emanating from the origin at an angle of 1T radians. and the portion outside the unit circle corresponds to 0. s jm s. in the z-plane corresponds to s in the s-plane. Constant-Frequency Loci: The constant-frequency 1 locus in the s-plane is a horizontal line parallel to the -axis. and all others at higher frequencies are called the complementary strips. I-23. …. 5... we devise the damping factor . 2.. the cor(I-190) which is a circle centered at the origin with a radius of e T. in the x-plane. constant. In the time-domain analysis of continuous-data systems. constant. as shown in Fig. 5. measured from the real axis. …. 1. are mapped onto the negative real axis of the z-plane. The point z 1 in the z-plane corresponds to the origin. Figure I-22(b) shows the mapping of the periodic strips from the s-plane to the z-plane. The j -axis in the s-plane is mapped onto the unit circle 0 z 0 1 in the z-plane. as shown in Fig. Regions shown in the periodic strips in the left-half s-plane are mapped onto the interior of the unit circle in the z-plane. s jm s 2. 4. m 1.

Only the loci inside the unit circle are shown.I-36 Appendix I Discrete-Data Control Systems jv s-plane 1 jv 2 s jv2 jv1 0 –jv1 0 – v1T 1 vT 2 2 jIm z z-plane v1T v2T Re z (a) (b) Figure I-24 Constant-frequency loci in the s-plane and the z-plane.n loci in the s-plane are concentric circles with the center at the origin. Constant Natural-Undamped Frequency Loci: The constant.n loci in the z-plane are shown in Fig. .loci in the z-plane are described by z eTs e 2p1tanb2 vs 2pv vs (I-192) jv s-plane vn = vs /2 vn = 3vs /8 vn = vs /4 vn = vs / 8 vn = vs /16 vn = 3vs /8 jIm z z-plane vn = vs /4 vn = vs /8 vn = vs /2 –1 0 1 vn = vs /16 Re z 0 s (a) (b) Figure I-25 Constant-natural-undamped frequency loci in the s-plane and the z-plane. and the radius is n. I-25 for n s 16 to s 2. the s-plane loci are v tan b jv (I-191) The constant. The corresponding constant. Constant-Damping Ratio Loci: described by s For a constant-damping ratio .

I-6 Time-Domain Properties of Discrete-Data Systems jv = 60° s-plane 45° 30° 20° 10°3° 15° 5° I-37 – 1 jvs 2 90° – 0 (a) z-plane jIm vs /4 = 20˚ = 0.loci in the top half of the z-plane.0 v=0 1 s (b) Figure I-26 Constant-damping-ratio loci in the s-plane and the z-plane. (I-193). Figure I-26 shows several typical constant.locus in the z-plane.866 s = 0˚ = 0 = 3˚ = 0. I-6-3 Relation between Characteristic-Equation Roots and Transient Response Based on the discussions given in the last section. the constant. . is a logarithmic spiral for 0 90 .259 vs/2 –1 0 = 90˚.5 = 45˚ = 0. = 1.342 = 30˚ = 0.174 = 15˚ = 0.052 = 5˚ = 0.087 = 10˚ = 0. we can establish the basic relation between the characteristic-equation roots and the transient response of a discrete-data system.707 = 60˚ = 0. where b sin 1z constant (I-193) For a given value of . described by Eq.

Roots on the Negative Real Axis in the z-Plane: The negative real axis of the z-plane corresponds to the boundaries of the periodic strips in the s-plane. I-27 and I-28. For example. Roots outside the unit circle correspond to unstable systems.I-38 Appendix I Discrete-Data Control Systems jv Complementary strip t 2 jvs /2 3 2 Primary strip 1 t – jvs /2 1 3 4 4 t t jvs s-plane Complementary strip – jvs (a) jIm z z-plane 3 2 4 1 t t 1 t t Figure I-27 (a) Transient responses corresponding to various pole locations of Y*(s) in the s-plane (complex-conjugate poles only). the response has a constant amplitude. and the responses will increase with kT. in general. of the system. but not on the stability. (b) Transientresponse sequence corresponding to various pole locations of Y(z) in the z-plane. The roots closer to the unit circle will decay slower. Roots on the Positive Real Axis in the z-Plane: Roots on the positive real axis inside the unit circle of the z-plane give rise to responses that decay exponentially with an increase of kT. the zeros of the closed-loop transfer function will also play an important role on the response. when . Typical responses relative to the root locations are shown in Figs. –1 Re z Unit circle (b) keeping in mind that. When the root is at z 1.

Thus. the complex-conjugate points are on the boundaries of the primary s 1 strip in the s-plane. the output sequence will occur in alternating positive and negative pulses.I-6 Time-Domain Properties of Discrete-Data Systems s-plane jvs t 1 jv 2 s t T –s2 t –s1 0 s I-39 jv – 1 jvs 2 (a) jIm z t z-plane Unit circle t –1 – e– s1T– e– s2T 0 1 Re z t (b) Figure I-28 (a) Transient responses corresponding to various pole locations of Y*(s) in the s-plane (complex-conjugate poles on the boundaries between periodic strips). the output sequence will have exactly one sample in each one-half period of the envelope. For the frequency of s 2. (b) Transient-response sequence corresponding to various pole locations of Y(z) in the z-plane. j s 2. as shown in Fig. The corresponding z-plane points are z e s1T e jvs T 2 e s1T (I-194) which are on the negative real axis of the z-plane. I-28(b). .

It should be pointed out that since the true error of the system is e(t). so that the input and output are represented in sampled form. Eq. as shown in the block diagram of Fig. …. r(kT) and y(kT). e*1t2 or e1kT2 r1kT2 y1kT2 (I-197) The steady-state error at the sampling instants is defined as e* ss lim e*1t2 tS r*1t2 y*1t2 (I-196) lim e1kT2 kS (I-198) By using the final-value theorem of the z-transform. I-19. Thus. (I-199) is written e* ss lim e1kT2 kS lim 11 zS1 z 12 1 R1z2 GhoGp 1z2 (I-200) This expression shows that the steady-state error depends on the reference input R(z) as well as the forward-path transfer function GhoGp(z) Just as in the continuous-data systems. the error signal should still be defined as e1t2 r 1t2 y1t2 (I-195) where r(t) is the input. Refer to Figs. Due to the discrete data that appear inside the system. we shall consider only the three basic types of input signals and the associated error constants and relate e* to these and the type of the system. respectively. e*s predicts only the steady-state error of the system at the sampling instants. 1. 2. That is. As the roots move toward the second and the third quadrants. the error signal is more appropriately represented by e*(t) or e(kT).I-40 Appendix I Discrete-Data Control Systems Complex-Conjugate Roots in the z-Plane: Complex-conjugate roots inside the unit circle in the z-plane correspond to oscillatory responses that decay with an increase in kT. I-18 be of the form K11 Ta s211 Tb s2 p 11 Tm s2 Gp 1s2 (I-201) s j 11 T1s211 T2s2 p 11 Tns2 where j 0. z-transform or difference equations are often used. the frequency of oscillation of the response increases. I-7 STEADY-STATE ERROR ANALYSIS OF DISCRETE-DATA CONTROL SYSTEMS Since the input and output signals of a typical discrete-data control system are continuoustime functions. ss Let the transfer function of the controlled process in the system of Fig. Roots that are closer to the unit circle will decay slower. s By expressing E(z) in terms of R(z) and GhoGp(z). I-27 and I-28 for typical examples. The error analysis conducted here is only for unity-feedback systems with H(s) 1. y(t) is the output. the steady-state error is e* ss lim e1kT2 kS lim 11 zS1 z 1 2E1z2 (I-199) provided that the function (1 z 1)E(z) does not have any pole on or outside the unit circle in the z-plane. The transfer function GhoGp(z) is Gh0Gp 1z2 11 z 1 2Z c K11 s j 1 11 Tas211 Tbs2 p 11 Tms2 d T1s211 T2s2 p 11 Tns2 (I-202) .

I-18 is a step function with magnitude R. (I-205). the step-error constant is written K* p lim GhoGp 1z2 zS1 1) in the denomi- lim 11 zS1 z 12 Kz z 1 K (I-209) Similarly. GhoGp(z) will have an s2 term in the denominator that corresponds to a term (z 1)2. I-18 is related to the step-error constant K* in the same way as in the continuous-data case. we get GhoGp 1z2 11 11 z 1 2Z c z 2c 1 K s 1 terms due to the nonzero poles d terms due to the nonzero poles d (I-208) Kz z Since the terms due to the nonzero poles do not contain the term (z nator. p We can relate K* to the system type as follows. This causes the step-error constant K* to be infinite. in Fig. for a type-1 system.I-7 Steady-State Error Analysis of Discrete-Data Control Systems I-41 Steady-State Error Due to a Step-Function Input When the input to the system. except p that K* is given by Eq. p For a type-0 system. (I-202). and the equation becomes GhoGp 1z2 11 z 1 2Z c K11 s11 Ta s211 T1s211 Tb s2 p 11 T2s2 p 11 Tm s2 d Tns2 (I-207) Performing partial-fraction expansion to the function inside the square brackets of the last equation. we see that the steady-state error of the discrete-data control system in Fig. The summary of the error constants and the steady-state error due to a step input is as follows: System Type 0 1 2 K* p K e* ss R (1 0 0 K) . The p same is true for any system type greater than 1. we get e* ss lim zS1 1 Rz z 1 (I-203) R GhoGp 1z2 1 R lim GhoGp 1z2 zS1 (I-204) Let the step-error constant be defined as K* p Equation (I-204) becomes e* ss R 1 K* p (I-206) lim GhoGp 1z2 zS1 (I-205) Thus. r(t). the z-transform of r(t) is R1z2 Substituting R(z) into Eq. j 0 in Eq. (I-200).

System Type 0 1 2 K* v 0 K e*s s R K 0 Steady-State Error Due to a Parabolic-Function Input When the input is a parabolic function. r(t) Rtus(t). The relations between the steady-state error e* . the steady-state error at the sampling instants is e* ss T2 lim 2 zS1 1z 1 lim 1z T 2 zS1 12 31 2 R1z GhoGp 1z2 4 12 (I-214) R 12 2GhoGp 1z2 By defining the parabolic-error constant as K* a 1 lim 3 1z T 2 zS1 12 2GhoGp 1z2 4 (I-215) the steady-state error due to a parabolic-function input is e* ss R K* a (I-216) . the z-transform of r(t) is R1z2 RT 2z1z 21z 12 12 3 (I-213) From Eq. (I-210) becomes e* ss R K* v (I-212) 1 lim 3 1z T zS1 12GhoGp 1z2 4 (I-211) * The ramp-error constant Kv is meaningful only when the input r(t) is a ramp function and if the function (z 1)GhoGp(z) in Eq.I-42 Appendix I Discrete-Data Control Systems Steady-State Error Due to a Ramp-Function Input When the reference input to the system in Fig. Eq. (I-211) does not have any poles on or outside the 1. (I-200). r(t) Rtus(t) 2. I-18 is a ramp function of magnitude R. Kv and the system unit circle 0z 0 * ss type when the input is a ramp function with magnitude R are summarized as follows. (I-200) becomes e* ss lim zS1 1z z T 12 31 RT GhoGp 1z2 4 (I-210) lim zS1 R 1 GhoGp 1z2 Let the ramp-error constant be defined as K* v Then. The steady-state error in Eq.

where n takes on all integers between and . The RL contain an infinite number of branches. (I-219). we get GH*1s2 1 T n a 1s K jnvs 21s jnvs 12 (I-221) which has poles at s jn s and s 1 jn s. By using the properties of the RL in the s-plane. G1s2H1s2 K s1s 12 (I-220) Substituting Eq. RL of 1 GH*(s) 0 are drawn as shown in Fig. This evidently makes the construction of the root loci of Eq.I-8 Root Loci of Discrete-Data Systems I-43 The relations between the steady-state error e* . I-30(b) for the sampling period T 1 s. I-29. consider that for the system of Fig. As an illustration. The characteristic equation roots of the system satisfy the following equation: 1 in the s-plane. Let us consider the discrete-data control system shown in Fig. System Type 0 1 2 3 K* a 0 0 K e* ss R K 0 I-8 ROOT LOCI OF DISCRETE-DATA SYSTEMS The root-locus technique can be can be applied to discrete-data systems without any complications. (I-213) are summarized as follows. the poles and zeros of GH*(s) in the s-plane will be infinite in number. (I-217) in the s-plane quite complex. K*. Thus. . (I-220) into Eq. and these Y*(s) T R(s) + – E(s) T H(s) E*(s) Y(s) G(s) Figure I-29 Discrete-data control system. I-30(a). and the system type when the input is a ss a parabolic function with its z-transform described by Eq. the root loci for discrete-data systems are plotted in the z-plane. or 1 GH1z2 0 (I-218) GH*1s2 0 (I-217) in the z-plane. rather than in the s-plane. The pole configuration of GH*(s) is shown in Fig. From Eq (I-64) GH*(s) is written GH*1s2 1 G1s T na jnvs 2H1s jnvs 2 (I-219) which is an infinite series. I-29. With the z-transformed transfer function.

In Fig. in general. I-29 with G1s2H1s2 . if the sampling period T is increased . (I-218) is. For the same system.3682 (I-222) The RL of the closed-loop characteristic equation are constructed based on the pole-zero configuration of GH(z). I-29 the loop transfer function in the z domain is GH1z2 1z 0. EXAMPLE I-23 Consider that for the discrete-data system shown in Fig.5 is drawn. and the intersection with the RL gives the desired value of K 1. amples illustrate the constructions of root loci for discrete-data systems in the z-plane. The same procedures of construction for continuous-data of discrete-data systems in systems are directly applicable in the z-plane for discrete-data systems.32 K –j3vs K=0 (a) (b) Figure I-30 Pole configuration of GH*(s) and the root-locus diagram in the s-plane for the K discrete-data system in Fig. Notice that when the value of K exceeds 4.32 K jvs j vs 2 K=0 K = 4.32.632Kz 121z 0. and the number of can be applied to root loci root loci is finite in the z-plane. s1s 12 clearly indicate that the closed-loop system is unstable for all values of K greater than 4. the constant-damping-ratio locus for 0. as shown in Fig.32 K=0 –1 0 K = 4. a rational function continuous-data systems in z with constant coefficients. it is well known that the same system without sampling is stable for all positive values of K. and the system becomes unstable. • The same procedures of The root-locus problem for discrete-data systems is simplified if the root loci are conconstruction of root loci of structed in the z-plane using Eq. The constant-damping-ratio locus may be superimposed on the RL to determine the required value of K for a specified damping ratio. T 1 sec.32 K –2vs –j 5vs 2 K=0 K = 4. (I-218).32. The following exthe z-plane.32 K jv jv s-plane K=0 K K=0 K K=0 j K=0 K = 4. one of the two roots moves outside the unit circle.32 K K –1 0 v –j s 2 –jvs –j 3vs 2 0 K K=0 K K=0 K K=0 K=0 K = 4. I-31. Since Eq.I-44 Appendix I Discrete-Data Control Systems 5vs s-plane j 3vs j 5vs 2 j 2vs 3vs 2 K=0 K K=0 K = 4. its poles and zeros are finite in number. I-31. In contrast.

since the marginal value of K for stability is 2.865Kz 121z 0. K G1s2H1s2 .624.32 K –1 0 K K=0 0.135 0. I-32. let us consider that a zero-order-hold is inserted between the sampler and the controlled process G(s) in the system of Fig.368 0. the system is actually less stable.1352 (I-223) The RL for this case are shown in Fig.367 K=0 1 Re z Figure I-32 Root-locus diagram of a discrete-data control system without zero-order-hold.I-8 Root Loci of Discrete-Data Systems j Im z I-45 z-plane K=1 K = 4. Note that although the complex part of the RL for T 2 seconds takes the form of a smaller circle than that when T 1 second. as compared with the marginal K of 4. to 2 seconds. the z-transform loop transfer function becomes GH1z2 1z 0.32 for T second. I-29. The loop transfer function of the system with the zero-orderhold is GhoGH1z2 K 3 1T 1 1z e T 2z Te e T T 121z 2 1 e T 4 (I-224) j Im z Unit circle z-plane K –1 K = 2. s1s 12 T 1 second. Next. s1s 12 T 2 seconds.62 path ( = 50%) K=0 1 Re z Unit circle Figure I-31 Root-locus diagram of a discrete-data control system without zero-order-hold. K G1s2H1s2 . .624 K= 0 K=0 0.

K G1s2H1s2 . I-9 FREQUENCY-DOMAIN ANALYSIS OF DISCRETE-DATA CONTROL SYSTEMS All the frequency-domain methods discussed in the preceding sections can be extended to the analysis of discrete-data systems. However. the absolute and relative stability conditions of the discrete-data system must be investigated with respect to the unit circle and the other interpretation of performance with respect to the regions in the z-plane. I-33(a) and I-33(b).I-46 Appendix I Discrete-Data Control Systems j Im z z-plane K = 2.526 K K= –1 –0. In conclusion.46 (b) Root loci for T = 2 seconds Figure I-33 Root-locus diagram of a discretedata control system with zero-order-hold.707 0 K=0 0. respectively. Comparing the root loci of the system with and without the ZOH.368 K=0 1 Re z K = 2.524 K=0 0 0. we see that the ZOH reduces the stability margin of the discrete-data system. the marginal value of stability for K is 2. s1s 12 The RL of the system with ZOH for T 1 and 2 seconds are shown in Fig. the root loci of discrete-data systems can be constructed in the z-plane using essentially the same properties as those of the continuous-data systems in the s-plane.46 –1. Consider the discrete-data system shown .135 K=0 1 Re z 0.3 for T 1 second and 1.478 K = 1.3 0. In this case.65 K = 0.46 for T 2 seconds.196 (a) Root loci for T = 1 second j Im z z-plane K = 1.3 K K= –0.

I-34. The closed-loop transfer function of the system is Y1z2 R1z2 GhoG1z2 1 GhoG1z2 (I-225) where GhoG(z) is the z-transform of Gho(s)G(s). I-9-1 Bode Plot with the w-Transformation The w-transformation introduced in Eq. the plot of GhoG(e j T) in the polar coordinates for 0 to s 2 needs to be plotted. 1. as shown in Fig. 2. Thus. Since the ping the points on the unit circle. Since the positive j -axis of the s-plane corresponds to real frequency. …. the frequency-domain plot of G(e j T) repeats for n s to (n 1) s. the absolute and relative stability conditions of the closed-loop discrete-data system can be investigated by making the frequency-domain plots of GhoG(z). Unit circle v = 3vs /2 . Just as in the case of continuous-data systems. I-35. (I-162) can be used for frequency-domain analysis and design of discrete-data control systems. This is also equivalent to map1.I-9 Frequency-Domain Analysis of Discrete-Data Control Systems I-47 Y *(s) T R(s) + _ E(s) T E*(s) ZOH Gho (s) G(s) Y(s) Figure I-34 Closed-loop discrete-data control system. in Fig. 0z 0 unit circle repeats for every sampling frequency s( 2 T). when is varied along the j -axis. In fact. it is necessary to plot GhoG(e j T) only for the range of 0 to s. in the z-plane onto the GhoG(e j T)-plane. The transformation is z 12 T2 12 T2 w w jIm z s-plane s = jv jvs jvs /2 v = vs /2 0 s –1 0 v = vs /4 z-plane (I-226) jv v = vs v=0 1 Re z Figure I-35 Relation between the j -axis in the s-plane and the unit circle in the z-plane. the frequency-domain plots of GhoG(z) are obtained by setting z e j T and then letting vary from 0 to . n 0. since the unit circle in the z-plane is symmetrical about the real axis.

1 v (rad/sec) –90 1.0 2.01 0. (I-226) and (I-227) in G(z) to get G( j w).0 10.002 0.57 s1s 12 (I-228) and the sampling frequency is 4 rad/sec. (I-166)]. let the transfer function of the process in the system in Fig.71dB GM = 5. –202. T 1. I-34.0 . w jvw 2 vT j tan T 2 (I-227) For frequency-domain analysis of a discrete-data system.5 –225 0. so that 1.2082 60 50 With ZOH 40 | G( jv) | (dB) 30 20 10 0 Without ZOH GM = 0.01 0.91° Figure I-36 Bode plot of GhoG1z2 of the system in Fig.243z GhoG1z2 G1z2 (I-229) 1z 121z 0.57 3 s1s 12 4 . we set [Eq.1 v (rad/sec) 1.001 0.I-48 Appendix I Discrete-Data Control Systems In the frequency domain.0 –111. with G1s2 1.77 dB –10 0.0 2.002 0.5 With ZOH Without ZOH –135 G( jv) (deg) PM = 39° –157.57 sec.0 10.001 0. I-34 be G1s2 1. EXAMPLE I-24 As an illustrative example on frequency-domain plots of discrete-data control systems. Let us first consider that the system does not have a zeroorder-hold.5 –180 PM = 2. and with and without ZOH. the latter can be used to form the Bode plot or the polar plot of the system. we substitute Eqs.

71 Phase Margin (deg) 39.7306 0.57 G1s2 . s 2 For the sake of comparison. the system with the ZOH is less stable. phase margin. and with and without ZOH.58 22.77 dB 1 Re Gho G(e jvT ) PM = 39° Figure I-37 Frequencydomain plot of 1. With ZOH 0 v 0 v Without ZOH . The gain margin. Gain Margin (dB) Without ZOH With ZOH 5. For the system with ZOH. Thus. the forward-path transfer j Im Gh0 G(e jvT ) GM = 0. as shown in Fig.5145 v=2 GM = 5. Similarly.57 seconds. (I-228). The mirror image of the locus shown. and GhoG1e jvT 2 in degrees versus . I-37 for three decades of frequency with the plots ended at 2 rad/sec. the phase of the Bode plot of the system with the ZOH is more negative than that of the system without the ZOH. and peak resonance of the two systems are summarized as follows.2082 (I-230) The polar plot and the Bode plot of the last equation are shown in Figs. I-36. with the mirror placed on the real axis. The Bode plot of GhoG(e j T) consists of the graphs of 0 GhoG1e jvT 2 0 in dB versus .77 0. the Bode plot and polar plot of the forward-path transfer function can be done using the w-transformation of Eqs. Notice that the polar plot of the system with the ZOH intersects the negative real axis at a point that is closer to the ( 1. j0) point than that of the system without the ZOH.I-9 Frequency-Domain Analysis of Discrete-Data Control Systems I-49 The frequency response of GhoG(z) is obtained by substituting z e j T in Eq. The polar j T plot of GhoG(e ) for 0 to s 2 is shown in Fig. the forward-path transfer function of the system with a zero-orderhold is obtained: GhoG1z2 1.91 Mr 1. respectively.71 dB v = 2 rad/sec PM = 2. s1s 12 T 1.64 As an alternative. represents the plot for s 2 to s.91° –1 0 –0.2215z 1z 121z 0. I-36 and I-37. (I-226).0 2.

T 1.504w211 w11 1.5711 0.57 . w j w. . I-34 with G1s2 1.I-50 Appendix I Discrete-Data Control Systems function in the w-domain is GhoG1w2 1.197w2 (I-231) 60 50 With ZOH 40 Amplitude (dB) Without ZOH 30 20 10 0 –10 10–3 10–2 10–1 100 101 vw –100 Without ZOH –120 With ZOH –140 Degrees –160 –180 –200 –220 10–3 10–2 10–1 100 101 vw Figure I-38 Bode plot of GhoG(z) of the system in Fig. The plots are done with the w-transformation.57 s1s 12 seconds with and without ZOH.0913w2 1.

(I-227). The conclusion from this illustrative example is that once z is replaced by e j T in the z-domain transfer function. or if the w-transform is used. the controlled process is the same. (I-232).I-10 Design of Discrete-Data Control Systems For the system without ZOH. I-39. (b) Digital control system with state feedback. except in discrete-data systems the controller is designed to process sampled or digital data. r(t) + – e(t) T Digital Controller D(z) e*(t) D(s) T Controlled Process e*(t) ZOH GP(s) y(t) (a) r (k) + – T u(k) ZOH Controlled Process x(k) K (b) Figure I-39 (a) Digital control system with cascade digital controller.978w2 I-51 (I-232) Substituting w j w into Eq. (I-38). I-10 DESIGN OF DISCRETE-DATA CONTROL SYSTEMS I-10-1 Introduction The design of discrete-data control systems is similar in principle to the design of continuous-data control systems. Block diagrams of these systems are shown in Fig. the Bode plots are made as shown in Fig. An in-depth coverage of the subject may be found in books dedicated to digital control. . the design of discrete-data control systems can be carried out in either the frequency domain or the time domain. Using computer programs. digital control systems can be designed with a minimum amount of trial and error. GhoG1 jw2 1 0. in most situations.6163w2 w11 1. The two frequencies are related through Eq. Notice that the frequency coordinates in Fig. The design of discrete-data control systems treated in this chapter is intended only for introductory purposes. In fact. I-38 are w. Just as with the design of continuous-data control systems. all the frequency-domain analysis techniques available for continuous-data systems can be applied to discrete-data systems. In this chapter we deal only with the design of a control system with a cascade digital controller and a system with digital state feedback. whereas those in Fig. The design objective is basically that of determining the controller so that the system will perform in accordance with specifications. I-36 are the real frequency .

df 1t2 ` dt t 1 1 f 1kT2 T f 3 1k 12T4 2 (I-234) kT To find the z-transfer function of the derivative operation described before. In general. the z-transfer function of the digital differentiator is GD 1z2 KD z Tz 1 (I-236) where KD is the proportional constant of the derivative controller. if at all. Thus. and others can be approximated by digital controllers. but the availability and advantages of digital control suggest that the controller be implemented by digital elements. I-40. we can show that as the sampling period T approaches zero. and second. there are situations in which the analog controller is already designed. the problem of implementing digital controllers by digital processors. the system should be designed so that the dynamics of the controller can be described by a z-transfer function or difference equations. phase-lead or phase-lag controllers. (I-236).I-52 Appendix I Discrete-Data Control Systems I-10-2 Digital Implementation of Analog Controllers It seems that most people learn how to design continuous-data systems before they learn to design digital systems. (I-234). forward-rectangular integration. Therefore. These are described as follows. it is not surprising that most engineers prefer to design continuous-data systems. However. the choice of the sampling period is extremely important. We have Za df 1t2 ` dt t b 1 11 T z 1 2 F1z2 z Tz 1 F1z2 (I-235) kT Thus. as shown in Fig. if the designer intends to use digital control. GD(z) approaches KD s. the constant KP cannot be realized with infinite resolution. I-10-3 Digital Implementation of the PID Controller The PID controller in the continuous-data domain is described by Gc 1s2 KP KD s KI s (I-233) The proportional component KP is implemented digitally by a constant gain KP. Since a digital computer or processor has finite word length. which is the transfer function of the analog derivative controller. the problems discussed in this section are twofold: first how continuous-data controllers such as PID. The value of T should be sufficiently small so that the digital approximation is adequately accurate. The three basic methods of approximating the area of a function numerically are trapezoidal integration. Let the integral of f (t) evaluated at t kT be . The time derivative of a function f(t) at t kT can be approximated by the backwarddifference rule. using the values of f(t) measured at t kT and (k 1)T. Ideally. There are a number of numerical integration rules that can be used to digitally approximate the integral controller KI s. that is. Trapezoidal Integration The trapezoidal-integration rule approximates the area under the function f (t) by a series of trapezoids. and backward-rectangular integration. we take the z-transform on both sides of Eq. Replacing z by eTs in Eq.

the digital approximation rule is illustrated in Fig. (I-141). we have the transfer function of the digital integrator as GI 1z2 KI U1z2 F1z2 KIT 1z 12 21z 12 (I-238) where KI is the proportional constant.I-10 Design of Discrete-Data Control Systems I-53 f [(k – 1)T] f(t) f(kT) u[(k – 1)T] 0 (k – 1)T kT t Figure I-40 Trapezoidalintegration rule. The integral of f (t) at t kT is approximated by u1kT2 u3 1k 12T4 Tf 3 1k 12T4 (I-241) The z-transfer function of the digital integrator using the backward-rectangular integration rule is GI 1z2 KI U1z2 F1z2 z KIT 1 (I-242) f(kT) u[(k – 1)T] 0 (k – 1)T kT t Figure I-41 Forwardrectangular integration rule. I-41. Forward-Rectangular Integration For the forward-rectangular integration. I-42. designated as u(kT ). the integral of f (t) at t kT is approximated by u1kT2 u3 1k 12T4 Tf 1kT2 (I-239) By taking the z-transform on both sides of Eq. (I-237). Taking the z-transform on both sides of Eq. the transfer function of the digital integrator using the forward-rectangular rule is GI 1z2 KI U1z2 F 1z2 KITz z 1 (I-240) Backward-Rectangular Integration For the backward-rectangular integration. . Then. as shown in Fig. we approximate the area under f (t) by rectangles. u1kT2 u3 1k 12T4 T 5 f 1kT2 2 f 3 1k 12T4 6 (I-237) where the area under f(t) for (k 1)T t kT is approximated by the area of the trapezoid in the interval.

By combining the proportional. the transfer function of the digital PD controller is Gc 1z2 aKP KD bz T z KD T (I-246) KP z–1 f (kT) – KD T + + + + u(kT ) Figure I-43 Block diagram of a digitalprogram implementation of the PID controller. and integration operations described before.I-54 Appendix I Discrete-Data Control Systems f [(k – 1)T] 0 (k – 1)T kT t Figure I-42 Backward-rectangular integration rule. Trapezoidal Integration Gc 1z2 aKP TKI 2 KD 2 bz T TKI 2 z1z 12 a KP 2KD bz T KD T (I-243) Forward-Rectangular Integration Gc 1z2 aKP KD T TKI b z2 z1z aKP 12 2KD bz T KD T (I-244) Backward-Rectangular Integration Gc 1z2 When KI aKP KD 2 bz T aTKI z1z KP 12 2KD bz T KD T (I-245) 0. derivative. TKI 2 + z–1 + + + . the digital PID controller is modeled by the following transfer functions.

382 1 d 1. I-44. and Gc(z).61T s s1s 0. The system configuration in Fig. the equivalent digital controller Gc(z) can be obtained by the arrangement shown. the transfer function of the continuous-data controller.61 (I-247) From Fig. I-44 is represented by the transfer function Gc 1s2 s s 1 1.61T The digital-program implementation of Eq.62e U1z2 z z 12 Z c 1. I-44. In practice. (I-248) is shown in Fig. (I-248). f (kT ) + + z –1 0. Figure I-44 illustrates the basic scheme with Gc(s). the time delay is implemented by storing a variable in some storage location in the computer and then taking it out after T seconds have elapsed.612 (I-248) e 1. we can realize it by using an analog controller Gc(s) and sample-and-hold units.38 + 0. I-10-4 Digital Implementation of Lead and Lag Controllers In principle. as shown in Fig. I-44 actually suggests that given the continuous-data controller Gc(s). . the equivalent digital controller. given the digital controller Gc(z). I-45. I-55 f (t) T f*(t) ZOH Gc(s) Gc(z) u(t) T u*(t) ZOH To controlled process Once the transfer function of a digital controller is determined. On the other hand. the controller can be implemented by a digital processor or computer. the transfer function of the digital controller is written Gc 1z2 11 F1z2 z 10.62e –1. any continuous-data controller can be made into a digital controller simply by adding sample-and-hold units at the input and the output terminals of the controller and selecting a sampling frequency as small as is practical. The sampling period T should be sufficiently small so that the dynamic characteristics of the continuous-data controller are not lost through the digitization.I-10 Design of Discrete-Data Control Systems Figure I-44 Realization of a digital controller by an analog controller with sample-and-hold units. consider that the continuous-data controller in Fig. EXAMPLE I-25 As an illustrative example.61T + + u(kT ) e–1. The operator z 1 is interpreted as a time delay of T seconds. Figure I-43 illustrates a block diagram representation of the digital program of the PID controller using the trapezoidal-integration rule.61T Figure I-45 Digital-program realization of Eq.

(I-249). subtraction. The three basic methods of decomposition for digital programming are discussed in the following sections. multiplication by a constant. Changing the components of a continuous-data controller is rather difficult once the controller has been built. or digital signal processors (DSPs). A distinct advantage of digital controllers implemented by microprocessors or DSPs is that the control algorithm contained in the controller can be easily altered by changing the program. (I-249) by direct decomposition for m 2 and n 3. microprocessors. and shifting. b1/a0 b2/a0 b3/a0 1/a0 z –1 z –1 z –1 Figure I-46 Signal-flow graph of digital program by direct decomposition of Eq. Digital controllers can be realized by digital networks. e1 –a1/a0 –a2/a0 –a3/a0 e2 . digital computers. (I-249) with n 3 and m 2. then a0 0. I-11-1 Physical Realizability of Digital Controllers The transfer function of a digital controller can be expressed as E2 1z2 b0 b1z 1 p bm z Gc 1z2 E1 1z2 a0 a1z 1 p an z m n (I-249) where n and m are positive integers. If Gc(z) is expressed as bm z m bm 1z m 1 p b1z b0 Gc 1z2 (I-250) an z n an 1z n 1 p a1z a0 then the physical realizability requirement is n m. This means that the series expansion of Gc(z) should not have any positive powers in z. We consider that a digital program is capable of performing arithmetic operations of addition. The transfer function Gc(z) is said to be physically realizable if its output does not precede any input. In terms of the Gc(z) given in Eq. The decomposition techniques presented in Chapter 5 can be applied to realize the digital controller transfer function by a digital program. if b0 0. The branches with gains of z 1 represent time delays or shifts of one sampling period. (I-249).I-56 Appendix I Discrete-Data Control Systems I-11 DIGITAL CONTROLLERS • One advantage of the digital controller is that its program can be easily altered. Digital Program by Direct Decomposition Applying direct decomposition to Eq. we have the following equations: E2 1z2 X1z2 1 1b b1z 1 p a0 0 1 1 E1 1z2 1a z 1 a0 a0 1 bmz a2z m 2 X1z2 p anz n 2 X1z2 (I-251) (I-252) 2 Figure I-46 shows the signal flow graph of a direct digital program of Eq.

5z 1 2 0. Digital Program by Parallel Decomposition The transfer function in Eq.5z 1 2X1z2 0. the transfer function is physically realizable. Equation (I-249) is written in factored form as Gc 1z2 Real Pole and Zero Gci 1z2 Complex Conjugate Poles (No Zeros) Gci 1z2 Ki 1 di1z 1 Gc1 1z2Gc2 1z2 p Gcn 1z2 (I-253) where the individual factors can be expressed as Ki 1 1 ciz diz 1 1 (I-254) di2z 2 (I-255) Complex Conjugate Poles with One Zero Gci 1z2 Ki 1 1 ciz 1 di1z 1 di2z 2 (I-256) and several other possible forms up to the second order.2z 1 (I-261) .or second-order transfer functions.2z 1 2X1z2 (I-258) Expanding the numerator and denominator of the last equation and equating. we have 110 5z 1 2X1z2 E1 1z2 1. Gc 1z2 E1 1z2 E2 1z2 1 0.or secondorder terms by partial-fraction expansion.5z 1 1 z 1 1 10 0. The transfer function Gc(z) is realized by the three types of digital programs discussed earlier.I-11 Digital Controllers I-57 Digital Program by Cascade Decomposition The transfer function Gc(z) can be written as a product of first. (I-249) can be expanded into a sum of simple first.2z 1X1z2 0. These terms are then realized by digital programs connected in parallel. (I-257) is divided into two factors in one of several possible ways. Cascade Digital Programming The right-hand side of Eq.2z 2X1z2 (I-259) (I-260) The last two equations are realized by the digital program shown in Fig. Gc 1z2 E1 1z2 E2 1z2 11 1011 z 1 211 0. each realizable by a simple digital program. Direct Digital Programming Equation (I-257) is written Gc 1z2 E2 1z2 X1z2 E1 1z2 E2 1z2 11 1011 z 211 1 0. The digital program of the overall transfer function is then represented by these simple digital programs connected in cascade. I-47.2z 1 2 (I-257) Since the leading coefficients of the numerator and denominator polynomials in z 1 are all constants. EXAMPLE I-26 Consider the following transfer function of a digital controller.

5 10 z–1 e2 Figure I-48 Cascade digital program of Eq.75 0. Gc 1z2 E1 1z2 E2 1z2 18. (I-257).2 –0.75 1 z 1 1 8. (I-257).75 1 e1 e2 1 – 8.2z 1 (I-262) Figure I-49 shows the signal flow graph of the parallel digital program of the controller. I-12 DESIGN OF DISCRETE-DATA CONTROL SYSTEMS IN THE FREQUENCY DOMAIN AND THE z-PLANE The w-transformation introduced in Section I-5 can be used to carry out the design of discrete-data control systems in the frequency domain. Parallel Digital Programming The right-hand side of Eq.I-58 Appendix I Discrete-Data Control Systems 10 5 1 e1 z–1 1.2 z–1 e2 Figure I-47 Direct digital program of Eq.75 z–1 1 0. all the design techniques for continuous-data control systems can be applied to the design of discrete-data systems. Once the transfer function of the controlled process is transformed into the w-domain. (I-257) is expanded by partial fraction into two separate terms. (I-257). 1 1 e1 1 0. Figure I-48 shows the signal flow graph of the cascade digital program of the controller.2 Figure I-49 Parallel digital program of Eq.2 z–1 0. z–1 18. .

the z-domain controller is obtained by substituting the w-transformation relationship in Eq. The maximum overshoot is 66 percent. respectively. The transfer function of the controlled process is Gp 1s2 2500 s1s 252 (I-265) The z-transfer function of the forward path.01 second. 10-28 and 10-45 for a 1 and a 1.005w211 w11 0. The gain and phase margins of the uncompensated system are 6. that is. The sampling period is 0. the single-stage phase-lead and phase-lag controllers in the w-domain can be expressed by the transfer function Gc 1w2 1 1 a tw tw (I-263) where a 1 corresponds to phase lead and a 1 corresponds to phase lag. (I-162). and shown in Fig.01 second.1152z 0. Once the controller is designed in the w-domain.106 0. (I-163).1152z 1z 121z The closed-loop transfer function of the discrete-data system is 1 Gh0Gp 1z2 z2 0.77882 (I-267) (I-266) Carrying out the z-transform in the last equation with T Gh0Gp 1z2 ® o 1z2 ® r 1z2 Gh0Gp 1z2 0. we have Gh0Gp 1w2 10011 0. (I-268). including the sample-and-hold is. z 12 T2 12 T2 w w (I-269) Substituting Eq. 10-29. respectively.8848 (I-268) The unit-step response of the uncompensated system is shown in Fig. When w is replaced by j w. . we get 0. (I-263) with a 1.39 dB and 14. the Bode plots of Eq. I-50. 2z 1 (I-264) w Tz 1 The following example illustrates the design of a discrete-data control system using the w-transformation in the frequency domain and the z-plane.77 . I-51. Let us carry out the design in the frequency domain using the w-transformation of Eq.000208w2 0.0402w2 (I-270) The Bode plot of the last equation is shown in Fig. (I-263) are identical to those of Figs. EXAMPLE I-27 Consider the sun-seeker control system described in Section 4-9. Phase-Lag Controller Design in the Frequency Domain Let us first design the system using a phase-lag controller with the transfer function given in Eq.I-12 Design of Discrete-Data Control Systems in the Frequency Domain and the z-Plane I-59 I-12-1 Phase-Lead and Phase-Lag Controllers in the w-Domain Just as in the s-domain. Now let us assume that the system has discrete data so that there is a ZOH in the forward path.106 1. (I-269) into Eq. Let us require that the phase margin of the system be at least 50 . Gh0Gp 1z2 11 z 1 2Z a 2500 b s2 1s 252 0.6636z 0.

8.84w (I-274) 1). We set 20 log10 a 16. (I-274) is shown in Fig. we set 1 a to be at least one decade below the gain crossover point at w 12. a phase margin of 50 can be realized if the gain crossover point is at w 12. From the Bode plot in Fig.7 dB. we need 16. Gc 1z2 1) (z 0.0 Uncompensated system With phase-lead controller With phase-lag controller u0(t) 1.9985 0.0100 sec 0 0 16 32 40 64 80 Number of sampling periods Figure I-50 Step responses of discrete-data sun-seeker system in Example I-27.1462 (I-273) 1 (I-272) Substituting the z-w-transform relation. w controller in the z-domain is obtained. 1 t The phase-lag controller in the w-domain is Gc 1w2 1 1 a tw tw (2 T )(z z z 1 1 w 6.8 and the gain of the magnitude curve of GhoGp( j w) is 16. The phase margin of the compensated system is improved to 55 .0 T = 0.I-60 Appendix I Discrete-Data Control Systems 2.8.7 dB (I-271) from which we get a 0. The unit-step .7 dB of attenuation to bring the magnitude curve down so that it will cross the 0-dB axis at w 12. in Eq. Thus. the phase-lag a 0.1468 (I-275) The Bode plot of the forward-path transfer function with the phase-lag controller of Eq. I-51. I-51. Next. (I-274). We set 1 at Thus.99 0.1462.

7dB 0 Amplitude (dB) With phase-lead controller – 20 – 40 With phase-lag controller Uncompensated system – 60 – 80 100 0 101 12.83˚ Uncompensated system –250 Figure I-51 Bode plots of discrete-data sun-seeker system in Example I-27. (I-263). it is expected that some difficulty may be encountered in designing a phase-lead controller for the . Since the slope of the phase curve near the gain crossover is rather steep.I-12 Design of Discrete-Data Control Systems in the Frequency Domain and the z-Plane 40 I-61 20 16. The maximum overshoot is reduced to 16 percent. –300 100 101 102 vw 103 104 105 response of the phase-lag compensated system is shown in Fig. as shown in Fig. I-51. I-50. Phase-Lead Controller Design in the Frequency Domain A phase-lead controller is obtained by using a 1 in Eq. The same principle for the design of phase-lead controllers of continuous-data systems described in Chapter 10 can be applied here.8 102 vw 103 104 105 –50 With phase-lead controller –100 Degrees PM = 55˚ –150 –180 –200 With phase-lag controller PM = 50˚ PM = 50.

Thus. the new gain crossover should be located at the point where the magnitude curve is at 26 2 13 dB. we can show that the best value of KD from the standpoint of overshoot is 0. I-50.6636z (I-283) The root contours of the system for KD 0 are shown in Fig. The characteristic equation of the closed-loop system is z1z 2 1.88482 1. Nevertheless.77882 (I-281) We can use the root-contour method to investigate the effects of varying KD.001944w (I-278) The transfer function of the phase-lead controller in the z-domain is Gc 1z2 8. we can assign a relatively large value for a. (I-246). In fact. but the rise time is faster.I-62 Appendix I Discrete-Data Control Systems system. Digital PD-Controller Design in the z-Plane The digital PD controller is described by the transfer function in Eq.3888z 6.52KD 1z 121z 0. the geometric mean of the two corner frequencies of the phase-lead controller should be at the point where the magnitude of GhoGp( j ) is 13 dB. since the contours do not dip low enough toward the real axis.6636z 0. we set KP 1. 1 t 1151a 514 (I-277) The w-domain transfer function of the phase-lead controller is Gc 1w2 1 1 atw tw 1 1 0. and the maximum overshoot is 28 percent. at the same time. the equivalent forwardpath transfer function with KD appearing as a multiplying factor is Geq 1z2 11. I-51 this is found to be at w 115. Digital PI-Controller Design in the z-Plane The digital PI controller introduced in Section I-10-3 can be used to improve the steady-state performance by increasing the system type and. The phase margin of the compensated system is 50. Let us set a 20.7776 0. the forward-path transfer function of the compensated system is Gc 1z2GhoGp 1z2 11 100KD 2z z 100KD 0.52KD 1z z1z 2 121z 0. Thus.7776z 1.88482 11.03888w 0. I-51.83 .92172 0. The maximum overshoot is 27 percent. and is repeated as Gc 1z2 aKP KD bz T z KD T (I-280) To satisfy the condition that Gc(1) 1 so that Gc(z) does not affect the steady-state error of the system. These root contours show that the effectiveness of the digital PD controller is limited for this system. The unit-step response of the phase-lead compensated system is shown in Fig. Applying the digital PD controller as a cascade controller to the sun-seeker system. From Fig. improve the relative stability by using . The gain of the controller at high values of w is 20 log10 a 20 log10 20 26 dB (I-276) From the design technique outlined in Chapter 10.106 0.6112 (I-279) The Bode plot of the phase-lead compensated system is shown in Fig.1152z 1z 121z 0. I-52.022.92172 0 (I-282) Dividing both sides of the last equation by the terms that do not contain KD. rather than using the amount of phase lead required as a guideline.

(I-245).9217 0 KD = 0 0.8318 – j0.393 + j0.8318 j0. As shown earlier. Let us select the backward-rectangular integration implementation of the PID controller given by Eq. (I-267) is written GhoGp 1z2 1z K1z 121z 0.77882 (I-285) where K 0. the dipole principle. the transfer function in Eq.8318 j0. the maximum overshoot of the system is 66 percent. The roots of the characteristic equation when K 0.8318 + j0.1152 are 0.4392 Figure I-52 Root contours of sun-seeker system in Example I-27 with digital PD controller.4392 and 0. KD varies.92172 0.022 0. If K is reduced to 0. the .1152. I-53.I-12 Design of Discrete-Data Control Systems in the Frequency Domain and the z-Plane jIm z z-plane I-63 KD = 0.01152.4392. The root loci of the system are drawn as shown in Fig. the transfer function of the digital PI controller becomes Gc 1z2 KP z 1KP KI T 2 z KP a1 z KI T b KP 1 z 1 (I-284) The principle of the dipole design of the PI controller is to place the zero of Gc(z) very close to the pole at z 1.4392 KD = 0 KD = 1 Re z 0. With KD 0.468 KD = 0 KD KD = –1 –0. The effective gain provided by the controller is essentially equal to KP. To create a root-locus problem.

I-64 Appendix I Discrete-Data Control Systems j Im z Uncompensated system K = 0.8836 + j0.99521z 0. z 0. (I-284) should be set to 0. (I-265) is reduced to 250.0898 K=0 K=0 Re z 1 K= 0. At the same time.0926 and 0.1K1z 0. This means to realize a similar improvement on the maximum overshoot.995.0926.77882 K .8318 + j0.995 Root-loci compensated system Figure I-53 Root loci of sun-seeker system in Example I-27 with and without digital PI controller.1152 0. We can show that if the gain in the numerator of Eq. Thus. The forward-path transfer function of the system with the PI controller becomes 0.9217 0 Root-loci uncompensated system K=0 0.1152 0. The maximum overshoot is reduced to only 3 percent.8863 + j0. we let the zero of Gc(z) be at 0.8836 j0.4392 Uncompensated system K = 0.92172 Gc 1z2GhoGp 1z2 (I-287) 1z 12 2 1z 0.05.1. the maximum overshoot of the system would be reduced to 3 percent.0926 K= –1 –0.7788 K Compensated system K = 0.995 Gc 1z2 0.01152 0.8836 j0.1 (I-286) z 1 The corresponding value of KI is 0. characteristic equation roots are at 0. the value of KP in Eq.

and thus the effect on the transient response is negligible. The output response reaches the desired steady state with zero error in a minimum number of sampling periods without intersampling oscillations. the two dominant roots of the characteristic equation are at 0.8863 j0. . The switching operation of sampling allows the discrete-data systems to have a finite transient period. The third root is at 0. In contrast. (I-289). Sirisena [10] showed that given a discrete-data control system with the controlled process described by Gh0Gp 1z 1 2 Q1z 1 2 P1z 1 2 (I-288) for the system to have a deadbeat response to a step input. a continuous-data system reaches the final steady-state trajectory or value theoretically only when time reaches infinity. Figure I-54 shows a typical deadbeat response of a discrete-data system subject to a unit-step input. H.0898. The following example illustrates the design of a discrete-data system with deadbeat response using Eq. (I-287) is approximately 8 percent. No attempt was made in optimizing the system with respect to a set of performance specifications.1152. y(t) 1 0 T 2T 3T 4T 5T 6T 7T t Figure I-54 A typical deadbeat response to a unit-step input.0898 and 0. which is very close to the pole of Gc(z) at z 1. A deadbeat response is one that reaches the desired reference trajectory in a minimum amount of time without error. This design problem simply illustrates the mechanics of designing a phase-lead and phase-lag controller using the w-transformation method in the frequency domain and digital PD and PI controllers in the z-plane. R. We can show that the actual maximum overshoot of the system with the forward-path transfer function in Eq. The root loci of the compensated system are shown in Fig.9948. the transfer function of the digital controller is given by Gc 1z2 Q112 P1z 1 2 Q1z 1 2 (I-289) where Q(1) is the value of Q(z 1) with z 1 1. I-13 DESIGN OF DISCRETE-DATA CONTROL SYSTEMS WITH DEADBEAT RESPONSE One difference between a continuous-data control system and a discrete-data control system is that the latter is capable of exhibiting a deadbeat response. I-53. When K 0.8863 j0.I-13 Design of Discrete-Data Control Systems with Deadbeat Response I-65 where K 0.1152.

the output velocity becomes zero after the second sampling period.92172 0.106 0.1152z 1 11 11 z 1 211 0.1152z 1 11 11 z 211 1 0. (I-207) and is written as GhoGp 1z 1 2 Thus.7788z 1 2 (I-291) (I-292) Q1z 1 2 P1z 2 1 0.106 0.77882 0. The digital controller for a deadbeat response is obtained by using Eq. The z-transfer function of the output velocity is written o 1z2 ® r 1z2 Gc 1z211 1001z z2 Gc 1z2GhoGp 1z2 12 z 1 2Z c 2500 d s1s 252 (I-298) The output velocity response due to a unit-step input is o 1z2 100 z 100z 1 (I-299) • The root sensitivity of a system with deadbeat response is poor. we evaluate the output velocity of the system. that is.1152z (I-295) The closed-loop transfer function of the compensated system is Gc 1z2GhoGp 1z2 1 Gc 1z2GhoGp 1z2 0. Gc 1z 1 2 Thus.22138z2 0.22138 1z z 1 211 0. To show that the output response is without intersampling ripples. I-55.22138z2 0. the output response reaches the unit-step input in two sampling periods.106z (I-293) 121z 0.0.11521z 0.92172 2 (I-296) 0.52041z z For a unit-step function input.7788z 1 2 0.22138.5204z 1 z (I-297) z 3 p Thus. Thus. The responses of o(t) and o(t) are shown in Fig. (I-289).1152z 0. The characteristic of a system with deadbeat response is that the poles of the closed- . which proves that the position response is deadbeat without intersampling ripples.1152z (I-294) The forward-path transfer function of the compensated system is Gc 1z2GhoGp 1z2 ® o 1z2 ® r 1z2 0. The forward-path transfer function of the uncompensated system is given in Eq. o(t) d o(t) dt.52041z 0.92172 2 z1z 12 0.7788z 1 2 1 2 0.I-66 Appendix I Discrete-Data Control Systems EXAMPLE I-28 Consider the discrete-data sun-seeker system discussed in Example I-27. the output transform is ® o 1z2 0. Gc 1z2 P1z 1 2 Q112 Q1z 1 2 11 0.9217z 1 2 (I-290) and Q(1) 0.9217z 1 2 0. Q1z 1 2 P1z 1 2 0.

the n roots of the characteristic equation 0zI A BK 0 0 (I-303) can be arbitrarily placed. from the standpoint of root sensitivity discussed in Chapter 8. the root sensitivity of a system with deadbeat response is very high. Since these are multiple-order poles. the closed-loop system is represented by the state equation x3 1k 12T4 1A BK2x1kT2 (I-302) Just as in the continuous-data case treated in Section 10-12. loop transfer function are all at z 0. Let us consider the discrete-data system described by the following state equation: x3 1k where x(kT) is an n control is 12T4 Ax1kT2 Bu1kT2 (I-300) 1 state vector. (I-300). I-14 POLE-PLACEMENT DESIGN WITH STATE FEEDBACK Just as for continuous-data systems. a matrix K exists that can give an arbitrary set of eigenvalues of (A BK). The following example illustrates the design of a discrete-data control system with state feedback and pole placement. pole-placement design through state feedback can be applied to discrete-data systems. (I-301) into Eq.0 0. we can show that if the pair [A. and u(kT) is the scalar control. that is. The state-feedback u1kT 2 Kx1kT 2 r1kT 2 (I-301) where K is the 1 n feedback matrix with constant-gain elements.5204 I-67 0 T 2T 3T 4T 5T 6T 7T t v0(t) 100 0 T 2T 3T 4T 5T 6T 7T t Figure I-55 Output position and velocity responses of discrete-data sun-seeker system in Example I-28.I-14 Pole-Placement Design with State Feedback u0(t) 1. . By substituting Eq. B] is completely controllable.

sun-seeker system with state feedback.8032k1 22.7788 22.2212 0.7788 0.7788 4.12k1 z 0.1152k1 ` 22.7788 0.12k1 0.12k2 2z 0. I-56(b).I-68 Appendix I Discrete-Data Control Systems EXAMPLE I-29 Consider that the sun-seeker system described in Example 10-9 is subject to sampled data so that the state diagram of the system without feedback is shown in Fig.1152k2 2x2 1kT2 22.1152k2 d 22. we have the discrete-data state equations as x1 3 1k x2 3 1k 12T4 12T4 11 0. (I-305) and (I-306) and letting t (k 1)T. The closed-loop system with state feedback for the time interval (kT) t (k 1)T is portrayed by the state diagram shown in Fig.12k1x1 1kT2 10.12k2 0 Solving for k1 and k2 from the last two equations. The sampling period is 0. the coefficient matrix of the closed-loop system with state feedback is A BK (I-308) The characteristic equation of the closed-loop system is 0zI A BK 0 ` z 0.8032k1 22. I-56(a). we get k1 0. The conditions on k1 and k2 are (I-310) 1. I-56(b).1152k1 Let the desired location of the characteristic equation roots be at z 0.1152k1 0.2212 0. 2500 s –1 ZOH –25 (b) X2(s) X1(s) .7788 0. we have X1 1s2 X2 1s2 2500k2 2500k1 1 1 c d x1 1kT2 d x2 1kT2 s s1s 252 s2 1s 252 s2 1s 252 2500k1 2500k2 1 x1 1kT2 x 1kT2 x 1kT2 s1s 252 s1s 252 2 s 25 2 c (I-305) (I-306) Taking the inverse Laplace transform on both sides of Eqs.2212 22.7788 22.12k2 0 (I-311) 0.07936 (I-312) s–1 u(t) T u*(t) ZOH 2500 s–1 X2(s) –25 s–1 X1(s) (a) –k 1 –k 2 x2(kT ) s–1 s u(kT ) –1 x1(kT ) s–1 s –1 Figure I-56 (a) Signalflow graph of open-loop. sun-seeker system.12k2 (I-309) 0 z2 1 1. with X1(s) and X2(s) as outputs and x1(kT) and x2(kT) as inputs. (b) Signal-flow graph of discrete-data.01 second. discrete-data.1152k2 1 0.1152k1 22. 0. The dynamics of the ZOH are represented by the branch with a transfer function of 1 s.7788 4.1152k1 2x1 1kT2 10. where the feedback gains k1 and k2 form the feedback matrix K 3k1 k2 4 (I-304) Applying the SFG gain formula to Fig.12k2 2x2 1kT2 c 1 0.2033 and k2 0.12k2 (I-307) Thus.12k2 22.

R.22 1z 121z2 z 12 (c) F1z2 1z z 12 1z 0. Vol. 8. 3. 2nd ed. NJ. Automatic Control. Automatic Control. Perform the partial-fraction of the following functions. Discrete-Time Control Systems. Workman. 1947–1948. Powell. Nagle. K. Determine the z-transforms of the following sequences.52 • z-transform. Use the final-value theorem of the z-transform if it is applicable. R.368z (a) F1z2 (b) F1z2 1z 121z 12 1z 12 1z2 1. x112 0 . New York. Apply partial-fraction expansion to F(z) and then use the z-transform table. AC-18. 2. Vol. December 1961. 5. Automatic Control. O. OGATA.” IRE Proc. Englewood Cliffs. “A Simplified Schur-Cohen Test. J. E. final-value theorem I-5. 2nd ed. pp. p .” IEEE Trans. Given that Z[f(k)] F(z). Vol. 1 1 (a) F1s2 (b) F1s2 3 3 1s 52 s 1s 12 (c) F1s2 10 s1s 52 2 (c) F1s2 5 s1s 2 22 • Inverse z-transform I-4. Jury and B. Jury and J. Phillips and H. 49. Find the inverse z-transforms f(k) of the following functions. F.52 Check the answers by carrying out the long division of F(z) and express it in a power series of z 1. AC-19. B. 9.. Digital Control System Analysis and Design. AddisonWesley. C. 1990. I. “Ripple-Free Deadbeat Control of SISO Discrete Systems. 7.” IEEE Trans. 1984 H.. AC-30. Automatic Control.” IEEE Trans. Kuo. 10z 0. 6. 2. Find the z-transforms of the following functions. June 1974. Englewood Cliffs.” IEEE Trans. E. PROBLEMS I-1. 1987. pp. pp. 6. “A Simplification of Jury’s Tabular Form. 157–163.852 (d) F1z2 1z 10 121z 0. MA. (a) x1k 22 x1k 12 0.Problems I-69 REFERENCES 1. Reading. Sirisena. 12. and M. New York. (a) f 1kT2 kT sin 2kT (b) f 1k2 e 1 1 k k 0. even integers 1. 248–250. z-transform I-3.. 3. Prentice Hall. (a) f 1k2 ke 3k (b) f 1k2 k sin 2k (c) f 1k2 e 2k sin 3k (d) f 1k2 k2e 2k • z-transforms I-2. find the value of f(k) as k approaches infinity without obtaining the inverse z-transform of F(z). L. February 1985. Solve the following difference equations by means of the z-transform. Cohen. G. John Wiley & Sons. 168–170. Prentice Hall. No. “A Set of Stability Constraints on the Denominator of a Sampled-Data Filter. D. L. M. Vol.. pp. 1992. Digital Control Systems. Blanchard. 1964. p . April 1973. T. Jr. pp. I. H.364z 0. NJ. Anderson. Vol. C. Jury.52 (d) F1z2 1z z 121z 1. “A Stability Test for Linear Discrete Systems in Table Form. L.1x1k2 us 1k2 x102 x112 0 (b) x1k 22 x1k2 0 x102 1. 327–328. if applicable. April 1966. 4. Oxford University Press. and then find the z-transforms using the z-transform table. 10. • z-transform solutions I-6. 4. odd integers • Partial-fraction expansion. z 10z (a) F1z2 (b) F1z2 1z 12 1z 0. Franklin. Raible. Theory and Application of the z-Transform Method. I. E. 7. D.7322 (c) F1z2 1z z2 12 1z 0. AC-11. Digital Control of Dynamic Systems. 5.

2. 2. It is well known that the transfer function of an analog integrator is G1s2 Y1s2 X1s2 1 s where X(s) and Y(s) are the Laplace transforms of the input and the output of the integrator. (a) Let y(kT) denote the digital approximation of the integral of x(t) from t 0 to t kT. r 0. owe nothing) 0. 1. (1) and show that the transfer function of the digital integrator is G1z2 Y1z2 X1z2 z Tz 1 . I-11. Perform the partial-fraction expansion to the following z-transfer functions.82 2z z (c) G1z2 (d) G1z2 2 1z 121z 0. (a) Show that the difference equation that describes the loan process can be written as P1k where P(k) P(O) P(N) 12 11 r2P 1k2 u • z-transform applications amount owed after the kth period. and substituting successively.22 5z (a) G1z2 (b) G1z2 1z 12 1z 0. Use G4(s) as determined in part (b).000. the rectangular integration is described by the schemes shown in Fig.01 (1 percent per month).521z 0. T is the sampling period. The continuous signal x(t) is approximated by a staircase signal. Consider that a new car is purchased with a load of P0 dollars over a period of N months at a monthly interest rate of r percent. … . I-8. I-9. starting with k k 1.I-70 Appendix I Discrete-Data Control Systems I-7. The principal and interest are to be paid back in N equal payments of u dollars each. The integral of x(t). This problem deals with the application of the difference equations and the z-transform to a loan-amortization problem. Then y(kT) can be written as y1kT 2 y3 1k 12T4 Tx1kT 2 (1) where y[(k 1)T] denotes the area under x(t) from t 0 to t (k 1)T. IP-10. discrete-data system • Numerical integration when the system is subject to an input sequence described by r(kT) transfer function G(z) Y(z) R(s). discretedata system k 0. The sampling period is 0. Take the z-transform on both sides of Eq. In a basic computer course. IP-10. I-10.52 1z 121z2 z 12 (d) Find y(t) for t 0 when the input is a unit-step function. is approximated by the area under the rectangular approximation signal.5 second. 2. Find the transfer functions Y(z) R(z) of the discrete-data systems shown in Fig.12 1z 121z 0. p 1 for all 0. and N 48 months. which is the area under x(t). Find u. Find the • Transfer function. The last two conditions are also known as the boundary conditions. 1. (d) Consider that P0 $15. A linear time-invariant discrete-data system has an output that is described by the time sequence y1kT2 1 e 2kT • Time response. then (c) Solve the difference equation in part (a) by using the z-transform method. Show that the solution of the equation is u 11 11 r2 NP0r r2 N 1 0. … . N. k P0 initial amount borrowed 0 (after N periods. 10z1z 0. the monthly payment. respectively. There are many ways of implementing integration digitally. (b) Solve the difference equation in part (a) by the recursive method.

• Vector-matrix discrete state equations I-12. as shown in Fig. ….Problems I-71 r(t) T r*(t) 1 s(s + 2) y(t) (a) r(t) T r*(t) 1 s +1 (b) 10 s +2 y(t) r(t) T r*(t) 1 s +1 (c) T 10 s +2 y(t) r(t) T r*(t) ZOH (d) h(t) 5 s(s + 2) y(t) r(t) + – e(t) T e*(t) ZOH 5 s (s + 2) y(t) (e) r(t) + – e(t) T e*(t) ZOH 5 s (s + 1) (s + 2) Figure IP-10 (f) (b) The rectangular integration described in Fig. we can use a “forward” rectangular hold. The signal x(t) is first sent through an ideal sampler with sampling period T. IP-11(a) can be interpreted as a sample-andhold operation. The block diagram of a sampled-data system is shown in Fig. x(kT ). IP-12. The state equations of the controlled process are dx1 1t2 dt x2 1t2 dx2 1t2 dt 2x1 1t2 3x2 1t2 h1t2 . Verify the result obtained in part (a) for G(z) using the “backward” sample-and-hold interpretation. (c) As an alternative. IP-11(c). Find the transfer function G(z) for such a rectangular integrator. …. The output of the sampler is the sequence x(0). x(T). These numbers are then sent through a “backward” hold device to give the rectangle of height x(kT) during the time interval from (k 1)T to kT. IP-11(b). as shown in Fig.

I-72 Appendix I Discrete-Data Control Systems x(t) x(T ) x(2T ) x(3T ) x (4T ) x(kT ) x(k+1)T x(0) 0 T 2T 3T 4T (k – 1)T (a) kT (k + 1)T t x(t) T Ideal Sampler x*(t) BACKWARD RECTANGULAR HOLD 1 s = y(t) (b) x(t) x(T ) x(0) x(3T ) x(2T ) x(k+1)T x(kT ) 0 T 2T 3T 4T (k – 1)T (c) kT (k + 1)T t Figure IP-11 where h(t) is the output of the sample-and-hold. u(t) is constant during the sampling period T. that is. (b) Find the characteristic equation of the system described in Problem I-12. dx1 1t2 dt x1 1t2 dx2 1t2 dt u1t2 The sampling period is 0. . 1. (a) Find the transfer function X(z) U(z) for the system described in Problem I-12. transfer function • Discrete-data system. (a) Find the transfer function X(z) U(z) for the system described in Problem I-13. (b) Find the characteristic equation and its roots of the system described in Problem I-13. • Discrete-data system. transfer function I-14. Repeat Problem I-12 for the linear sampled-data system with the following state equations. (a) Find the vector-matrix discrete state equations in the form of x 3 1k 12T4 f1T 2x1kT 2 u1T 2u1kT 2 0. 2. I-15.001 second. … N. (b) Find x(NT) as functions of x(0) and u(kT) for k y*(t) u(t) u*(t) T ZOH h(t) T x = Ax + Bh y(t) Figure IP-12 • Vector-matrix discrete state equations I-13.

. IP-18. • Digital control.52 0 Determine the values of K for the system to be asymptotically stable. Draw a state diagram for the digital control system represented by the following dynamic equations: x1k 12 Ax1k2 Bu1k2 y1k2 x1 1k2 A 0 £0 5 1 1 3 1 2§ 1 B 0 £0§ 1 Find the characteristic equation of the system.5Kz 1K 0.5z 1 0 (b) z3 z2 3z 0.2 z–1 z–1 2 z–1 1 y(k) Figure IP-17 –0. Apply the w-transform to the following characteristic equations of discrete-data control systems.1 –0. state equations. r(t) + – e(t) T e*(t) ZOH G(s) y(t) Figure IP-18 • Stability of discrete-data systems G(s) = 1 s+1 T = 1 sec I-19. A digital control system is described by the state equation x1k 12 10. (a) z2 1. Determine the values of K for the system to be asymptotically stable. Find the transfer function Y(z) R(z). or unstable) using the Routh-Hurwitz criterion. statediagram characteristic equation I-73 I-16. • Stability of digital control system I-20. • Stability of digital control system I-21.Problems • Digital control. 1 1 r(k) –0. IP-17. state diagram.368 0. transfer function I-17.5 0 Check the answers by solving for the roots of the equations using a root-finding computer program. Draw a state diagram for the system. and x(k) is the state variable.2 0 (c) z3 1.632K2x1k2 Kr 1k2 where r(k) is the input. state diagram I-18. The state diagram of a digital control system is shown in Fig. The block diagram of a sampled-data system is shown in Fig. Write the dynamic equations. The characteristic equation of a linear digital control system is z3 z2 1.1 • Sampled-data system.2z2 2z 3 0 (d) z3 z2 2z 0. marginally stable. Write the discrete state equations of the system. and determine the conditions of stability (asymptotically stable.

• Stability of discrete-data control system r(t) +– e(t) T e*(t) ZOH K s(s + 1.I-74 Appendix I Discrete-Data Control Systems I-22. (c) Repeat part (a) when the sampling period is 1.2z 0.1 second and Kt (e) Repeat part (d) for T 0.05 second. find the critical value of K for system stability. and K*. The block diagram of a sampled-data control system is shown in Fig. (b) Repeat part (a) when the sampling period is 0.1 second.5z 0. (b) Compute the unit-step response y(kT) for k 0 to 100. IP-22. (c) For T 0.5 0 (d) z4 0.1 second. (c) Repeat parts (a) and (b) for T 0. error constants I-25.5z3 0. The block diagram of a sampled-data control system is shown in Fig.1 second and Kt 1. (a) For T 0.5) y(t) Figure IP-22 I-23.25 0 I-24.0 second. and determine the stability condition of the systems. Use a root-finding computer program to find the roots of the following characteristic equations of linear discrete-data control systems. (a) Derive the forward-path and the closed-loop transfer functions of the system in z-transforms. R(s) + – E(s) 10 U(s) + – T U*(s) ZOH 1 s 1 s Y(s) Kt Figure IP-25 .1 second.25z2 0.5 0 (c) 0.5z3 • Sampled-data system z2 1.5 0 (b) z3 z2 z 0. The sampling period is 0. The block diagram of a discrete-data control system is shown in Fig. p v a (b) Derive the transfer functions Y(z) E(z) and Y(z) R(z). K*. (a) z3 2z2 1. IP-25.1z 0. find the values of K so that the system is asymptotically stable at the sampling instants. (d) Compute the unit-step response y(kT) for k 0 to 50 for T 0. R(s) + – E(s) T E*(s) ZOH H(s) 5 s(s + 2) Y(s) Figure IP-24 • Sampled-data system. (a) Find the error constants K*.5 second. IP-24. 5.

(kT) k 0. The block diagram of a sampled-data control system is shown in Fig. IP-26(a). Ra r(t) MICROPROCESSOR e(kT) D/A AMPLIFIER K + ea – + eb M – vm Damper ROTOR ENCODER (a) Vm (s) * T R(s) + – E(s) T Amplifier-motor damper E*(s) ZOH Gp (s) Vm(s) (b) Figure IP-26 • Root loci of sampleddata system I-27.1 second. and a parabolic function t2us(t) 2.5 second. The digital control is modeled by the block diagram shown in Fig. The forward-path dc-motor control system described in Problem 5-33 is now incorporated in a digital control system. 1.01 and 0. (a) Construct the root loci in the z-plane for the system for K 0. . This generates the sequence of numbers. Find the marginal values of K for stability. I-75 I-26. Find the steady-state error e(kT) as k S when the p v a input r(t) is a unit-step function. IP-27. G1s2 K s1s 52 (b) Repeat part (a) when the system has a zero-order-hold. . Use any computer simulation program. as shown in Fig. 2. IP-27. a unit-ramp function. . The microprocessor then generates the error signal e(kT) r(kT) (kT). and K*. Show that the closed-loop system is unstable when T 0. as shown in Fig. The microprocessor takes the information from the encoder and computes the velocity information.001 second. without the zero-order hold. Find the characteristic equation and its roots. . (a) Find the transfer function (z) E(z) with the sampling period T 0. (c) Repeat parts (a) and (b) for T 0. Locate these roots in the z-plane. .1 second. and then with T 0. Use the parameter values given in Problem 5-33. (d) Find the error constants K*. K*.1 second. (b) Find the closed-loop transfer function (z) R(z).Problems • Digital dc-motor control systems. IP-26(b). when T 0.

2s s s I-33. (a) Backward-rectangular integration rule. The forward-path transfer function of the system is GhoG1z2 1 e s Ts a s1s 16. • Frequency-domain analysis of liquid-level control system with sampled data I-31. The characteristic equations of linear discrete-data control systems are given in the following equations. (b) Apply the w-transformation of Eq. Construct the Bode plot of GhoG(w) using the w-transformation of Eq. I-29.01s • Digital program implementation of digital controllers .1 second. IP-33. and determine the limiting value of N (integer) for the closed-loop system to be stable. Draw a digital-program implementation diagram for the digital controller.5K2 10 K1z 5 0 121z 0.5Kz 1K 12 0 (b) z2 (c) z 2 (d) z2 (e) 1z • Frequency-domain analysis of discrete-data system 10. G1s2 Construct the root loci in the z-plane for K • Root loci of discretedata systems Ke 0. Find the transfer function of the equivalent digital controller.9052 The sampling period is T 0.1s s1s 121s 22 0. Determine the marginal value of K for stability.52 The sampling period is 0.1 second. The forward-path transfer function of a unity-feedback discrete-data control system with sample-and-hold is GhoG1z2 1z 0. (b) forward-rectangular integration rule. Construct the root loci for K 0.67N b 121s 12.52 10 (a) Gc 1s2 (b) Gc 1s2 s 12 1s 102 s 1 0.0952z 121z 0. Find the gain and phase margins of the system. Consider that the liquid-level control system described in Problem 6-13 is now subject to sample-and-hold operation.1 second. Use the backward-difference rule for derivatives. 200 5 (i) Gc 1s2 2 (ii) Gc 1s2 10 0. (I-226). Carry out the analysis for the following continuous-data controllers. (a) Plot the plot of GhoG(z) and determine the stability of the closed-loop system. The system shown in Fig.5 4 0 0 0. The parameter N represents the number of inlet valves. Find the digital equivalents using the following integration rules for the controllers given.55 1 0. and (c) trapezoidal-integration rule.15K 1.05 second.4s (c) Gc 1s2 (d) Gc 1s2 s 1.72 0 I-30. with T 0.5 1 10. The sampling period is 0. 101s 1. (I-226) to GhoG(z) and plot the Bode plot of GhoG(w).42 0.52z 12z 0.1s (iii) Gc 1s2 1 0. (a) z3 Kz2 1. IP-27 has the following transfer function.1K 12 1z 10. A continuous-data controller with sample-and-hold units is shown in Fig.I-76 Appendix I Discrete-Data Control Systems R(s) + – E(s) T E*(s) ZOH G(s) Y(s) Figure IP-27 • Root loci of sampleddata system I-28.4 2 10.14K2z z 0. • Digital integration I-32.

5 and 0. • Inventory-control system with digital phase-lead controller • Aircraft attitude control system with digital controller I-37.5z 2 2 1.5z 1 z 2 (a) Gc 1z2 (b) Gc 1z2 1 2 3 z z 1. . • Inventory-control system with digital PD controller I-36.5 z2 z 1 z 1 1 (d) Gc 1z2 (f) Gc 1z2 z z 1 0. The sampling period is 0.5z 2z z 1 2 3 1 0. 2. Figure IP-35 shows the block diagram of the inventory-control system described in Problem 10-17 with a digital PD controller. Gc 1z2 r(t) + – e(t) Gc(s) T Gc(z) T ZOH aKP KD bz T z KD T 4 s2 y (t) Figure IP-35 Select a sampling period T so that the maximum overshoot of y(kT ) will be less than 1 percent.2z 1 0. .5z z 2 3 • Inventory-control system with digital PD controller I-35. The transfer function of the process of the inventory-control system described in Problem 10-17 is 4 Gp 1s2 s2 The block diagram of the system with a PD controller and sample-and-hold is shown in Fig.5z 1 z 1 2z 2 (c) Gc 1z2 (e) Gc 1z2 z3 0. 1. The process transfer function of the second-order aircraft attitude control system described in Problem 10-5 is 4500K Gp 1s2 s1s 361. Can you design a phase-lag controller in the w-domain? If not. 1011 0. Determine which of the following digital transfer functions are physically realizable. Plot the output response y(kT) for k 0.1z z 1. Find the third root. design a phase-lead controller using the w-transformation so that the phase-margin of the system is at least 60 .22 . (b) Set KP 1. I-38. IP-35. explain why not. Find the value of KD so that the maximum overshoot of y(kT) is a minimum.01 second.5. Consider that the digital PD controller has the transfer function Gc 1z2 KP KP 1z 12 Tz (a) Find the values of KP and KD so that two of the three roots of the characteristic equation are at 0.Problems S2 y*(t) S1 r(t) r *(t) ZOH T h(t) Gc(s) T y(t) I-77 Figure IP-33 • Physical realizability of digital transfer functions I-34. . For the inventory-control system described in Problem I-36. Find the transfer function of the digital PD controller using the following equation. .

I-78

Appendix I Discrete-Data Control Systems Consider that the system is to be compensated by a series digital controller Gc(z) through a sample-and-hold. (a) Find the value of K so that the discrete ramp-error constant K* is 100. v (b) With the value of K found in part (a), plot the unit-step response of the output y*(t) and find maximum overshoot. (c) Design the digital controller so that the output is a deadbeat response to a step input. Plot the unit-step response.

• Sun-seeker system with digital controller; deadbeat response • Sun-seeker system with state feedback • State-feedback control

I-39. The sun-seeker system described in Example I-5 is considered to be controlled by a series digital controller with the transfer function Gc(z). The sampling period is 0.01 second. Design the controller so that the output of the system is a deadbeat response to a unit-step input. Plot the unit-step response of the designed system. I-40. Design the state-feedback control for the sun-seeker system in Example I-5 so that the characteristic equation roots are at z 0.5, 0.5. I-41. where A c 0 1 1 d 1 B 0 c d 1 Consider the digital control system x 3 1k 12T4 Ax1kT2 Bu1kT2

The state-feedback control is described by u(kT ) Kx(kT), where K [k1 k2]. Find the values of k1 and k2 so that the roots of the characteristic equation of the closed-loop system are at 0.5 and 0.7.

Appendix J z-Transform Table

TO ACCOMPANY AUTOMATIC CONTROL SYSTEMS EIGHTH EDITION

BY BENJAMIN C. KUO FARID GOLNARAGHI

JOHN WILEY & SONS, INC.

Copyright © 2003 John Wiley & Sons, Inc. 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, scanning or otherwise, except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923, (508)750-8400, fax (508)750-4470. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 605 Third Avenue, New York, NY 10158-0012, (212) 850-6011, fax (212) 850-6008, E-Mail: PERMREQ@WILEY.COM. To order books or for customer service please call 1-800-CALL WILEY (225-5945). ISBN 0-471-13476-7

APPENDIX J z-Transform Table Laplace Transform 1 1 s 1 1 e 1 s2 1 s3 1 s s 1s s1s v s2 1s s 1s 2 n 1 Ts Time Function Unit impulse (t) Unit step us(t) dT 1t2 a d1t n 0 z-Transform 1 z z nT 2 z z 1z 21z lim Tz 12 2 12 12 3 z e d 1 1 t t2 2 tn n! e t T 2z1z 1 12 n 0 n c aS0 n! 0an z z t aT 1 a 1 a2 2 a a2 v2 v a2 2 s v a a2 2 2 2 z e aT te 1 e sin t t t 1z z2 2 11 Tze aT 1z e aT 2 2 e 121z aT 2z e aT 2 1 2aT z sin vT 2z cos vT v e sin t z ze aT sin vT 2ze aT cos vT z1z z 2 e 1 cos t t cos vT 2 2z cos vT s v 2 e cos t z 2 z2 ze aT cos vT 2ze aT cos vT e 2aT J-1 .

INC.Answers to Selected Problems TO ACCOMPANY AUTOMATIC CONTROL SYSTEMS EIGHTH EDITION BY BENJAMIN C. KUO FARID GOLNARAGHI JOHN WILEY & SONS. .

except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act. or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center. without either the prior written permission of the Publisher. New York. Danvers. Requests to the Publisher for permission should be addressed to the Permissions Department. John Wiley & Sons. (508)750-8400. MA 01923. ISBN 0-471-13476-7 . To order books or for customer service please call 1-800-CALL WILEY (225-5945). Inc.Copyright © 2003 John Wiley & Sons. No part of this publication may be reproduced. E-Mail: PERMREQ@WILEY. fax (508)750-4470. scanning or otherwise. 222 Rosewood Drive. Inc. electronic. All rights reserved. NY 10158-0012. mechanical.COM.. fax (212) 850-6008. 605 Third Avenue. stored in a retrieval system or transmitted in any form or by any means. (212) 850-6011. photocopying. recording.

. (b) Y1 ¢ ¢ 1 G1G2H1 G3H2 G4H3 G2G3G4G5H4 G2G6H1H4 G1G2G4H1H3 Y6 G1G2G3G4 G3G4G5 ` 3-12.198s 10 0 CHAPTER 4 Modeling of Physical Systems 1B1 B2 2 dy1 d2y1 4-1. . (c) System transfer function: 2 Er 1s2 s 6s 0. 2-3. (b) Torque equations: 1u u2 2 J 1 dt2 4-7.5s 3-2. 1 j.5s 2 1 2t 1 1 e e 3 2 3 0. . (b) Characteristic equation: s2 7s 25 0 3-22. (a) g1t2 (e) g1t2 1 e s GT 1s2 s 11 e s 2 211 e 0. Gd 1s2 2 E1s2 N 0 s 10 6s 20 120 25e 3-19. (b) 3-16.01s2 30. 1 j.57142e 0. (a) Y1 Y7 0 ¢ ¢ 1 G2H1 G4H2 G1G2G3G4H3 G3G4G5H3 G2G4H1H2 101s 42 s1s 102 Y1s2 ` 3-14. (c) Poles: s 0. 10. zeros: s 2. 1. K1u1 u2 2 B J dt K y B2 2 4-8. zero: s 2.5s 2 3t 1 11 s e s22 s 2 11 e 0.Answers to Selected Problems The answers given in the following section are for certain selected problems. 0. (a) He 1s2 Hi 1s2 1Ra Las2 (b) 1Jm JL m 1s2 r 1s2 G2G4G6H1H3H4 t 5e 5t 2us 1t2 dy2 dy1 B2 dy2 1 f. and thus are often shown for only certain parts of a problem. (b) Optimal gear ratio: n* 4-12. (a) Poles: s 0.002333K1s 8.5s E1s2 1 G3 1s2H1 1s2 G1 1s2G3 1s2 3-6. (a) G1s2 (b) G1s2 2-6. CHAPTER 2 Mathematical Foundation 2-1. (b) ¢ 1 2 3G1s2 4 2 3-20. .57142e 0. ` ¢ 1 G2 1s2G3 1s2 G3 1s2H1 1s2 R1s2 N 0 ¢ E1s2 1 ` N1s2 R 0 ¢ G1G2G3G4G5 G6 11 G3H2 G4H3 2 Y7 3-8. M dt M dt dt du2 1 T.00035K1s 8. (b) y1t2 3-21. (b) Transfer function: 1 KbHe 1s2 s3JTLas 2 4-14. (d) Forward-path transfer function: ® o 1s2 TL 0 ` Tm 1s2 Y1s2 s 3 1Jm r Mr 2 2s Bm 4 ® e 1s2 1RaJT KKsKin BmLa 2s RaBm KiKb 4 Ans-1 . These answers are given for checking purposes. .5t 2e t t 0 t 0 CHAPTER 3 Block Diagrams and Signal-Flow Graphs 3-1. (c) Characteristic equation: s3 21. (b) Force equations: M dt dt2 d2u1 K 4-3. (d) Steady-state speed 6.66 ft/sec V1s2 0. (c) Characteristic equation: s3 5s2 6s 10 0 Y1s2 1 (d) Transfer function: R1s2 s3 5s2 6s 10 3-24.

5-32. Bv1t2 KiKf i2 1t2 ia(t) as input. k2 1 3k1 3k2 0 . 5-43.325. Two poles on imaginary axis. 6-3.1 0 20 1 x3 0 0 TD # x4 0 0 0 5 x4 30 0 (b) Characteristic equation: s3 22s2 170s 600 0 The system is controllable. e3t 0 (c) State transition matrix: f1t2 d c 0 e 3t (g) Characteristic equation: ¢1s2 s3 15s2 75s 125 0 (a) Eigenvalues of A: 2. (c) Forward-path transfer function: G1s2 4-18. 5-28. 5-16.323 j1.6667 106 s 4. (a) f1t2 c d d . (c) Characteristic equation: 4-21.5623 Y1s2 s 2 (b) (3) Output transfer function: U1s2 1s 12 2 1 1 0 9 6 1 (c) S 3B AB AB2 4 £1 2 4 § . 0. (b) State equations: 4-26. (b) Poles are at s 5.3376 j0.5 j1. 1 j.42s 2. Stability requirement: Kt 7 0.7 106 K s1s 423.5 5-6. Stable. (f) Two roots in RHP. P SM £ 6 5 1§ 1 6 23 3 1 1 0 1 1 0 (e) S c d . (b) Condition for stability: K 7 4.2342 108 2 2 1JKL B2s K2B K3K4e tDs 0 Js dx1t2 dv1t2 v1t2. (e) Conditions for stability: K 7 2 and K 6 2. 0.5623. Marginally stable. S is singular. ® v 1s2 KIH (a) Transfer function: 2 2 R1s2 Jv s 1JG s KP s KI KN 2 11 (b) For controllability. (d) Poles are at s 5. c 12 2 12 sin t cos t Y1s2 51K1 K2s2 (a) Forward-path transfer function: G1s2 E1s2 s3s1s 421s 52 104 51K1 K2s2 Closed-loop transfer function: M1s2 s4 9s3 20s2 110 5 K2 2s 5 K1 te t e t (b) State transition matrix: f1t2 d c 0 e t # x1 2 20 1 0 x1 0 1 # x2 0 10 1 0 x2 0 0 u (b) State equations: D # T D TD T D Tc d x3 0. .Ans-2 Answers to Selected Problems 608. 5-45. for observability. j12. 6-2. 5-11. 1 j. 5-8. 5-42. 0V 0 2 CHAPTER 6 Stability of Linear Control Systems 6-1. (a) There is one root in the region to the right of s 1 in the s-plane. All poles in the left-half s-plane. a dt dt TD s Y1s2 KKinGc 1s2e 3 2 4-16.323. 0. j12. P SM c d 3 B AB 4 1 3 1 1 2 2 2 12 cos t sin t (c) S 5-13. 5-19.. (a) Eigenvalues: s 0. (b) No roots in RHP.3376 j0.9055 6-5.6667 6-9.081 6-12. (b) Forward-path transfer function: E1s2 s51Ra Las2 3 1Jm JL 2s Bm 4 KbKi 6 CHAPTER 5 State Variable Analysis 5-3.

7 in. (a) The gain-crossover frequency is 10 rad/sec.68.223 rad/sec r (e) Mr 1.07. 9-12.27 (h) GM infinite.82 rad/sec.0206 n Y1s2 802.333 6 a 6 3. System transfer function: R1s2 s2 25. (a) GL 1s2 s1s 0. The characteristic equation has two roots in the right-half s-plane.82 rad/sec. Rise time tr 0.47 8-7.707.78882 CHAPTER 8 Root-Locus Technique 8-2.65 . 0. (a) Unit-step input: ess a0 KH 3 (d) Unit-step input: ess 0. (h) Intersect of asymptotes: s1 4.84s 802. 2. (e) GM 6. Maximum Mr 1.Answers to Selected Problems CHAPTER 7 Time-Domain Analysis of Control Systems 7-3. 9-33.8333. breakaway points: 0. except at K 1. (a) For stability.14 7-27.2222 7-35. 9.57.7272.47 seconds. (b) PM 2. PM infinite .34. Thus the closed-loop system is stable.8952 s1s 0. unit-ramp and unit-parabolic inputs: ess 7-10. (a) Breakaway-point equation: 2s2 311 a2s 6a 0 For no breakaway point other than at s 0.02 and K 7 0 7-13. BW 1. unit-ramp input: ess 0.56 2 (c) Stability requirement: 0 6 K 6 1400 Y1s2 100s1s 22 7-28. Kt 0. (b) Nyquist plot intersects the negative real axis at 0.5. (f) a 5 ` D1s2 r 0 s3 100s2 699s 1000 0. BW 6.59 7-24. The closed-loop system is unstable.618. 9-14. K 8.22252 0. (b) k2 59 10k1 (c) k2 13. (a) For stability. 9-18. The system is stable for 0 6 K 6 . (c) Kp . unit-parabolic input: ess 7-6. 10.851 8-21. j1. breakaway points: (K 7 0) 0. (c) K can be increased by 28.4 8-16.72 dB.995 4. 3. (d) 0.3887 8-5. K 4.4 9-26. 0. (b) Td 0. (d) £M 34.71 dB.4.12 rad/sec r 9-5. (a) The system is stable for 0 6 K 6 240.496.59 . PM 13. 5. 9-21. 9-27. unit-ramp input: ess 7-5.327. Mr 1. (b) Maximum D 15. (a) Stability requirement: 0 6 K 6 3000. 9-11. 9-10. N 6 3 (N has to be an integer). (a) GL 1s2 (e) GL 1s2 s1s 0.75 rad/sec. (a) The system is stable for 25 6 K 6 .435 8-4.51 dB. (b) Mr 15. Ka 0 b0KH 1 2 a1 b .975 7-32. (c) Unit-step input: ess 2. (a) K 0: 1 135 K 0: 1 45 (e) 1 108. (a) s1 1.2 sec.82 dB.53 rad/sec r 9-35.05. BW 9.45 rad/sec n r 9-2.1244 second 9-40. Mr 17. Td 1. 9-25. 4. (c) For K 100. (a) The Nyquist plot encloses the 1 point.72. j1. 9-15. 8 (l) Breakaway points: 2. 0. (b) GM 30.5° (g) BW 30 rad/sec 9-36.47. PM 50. (d) K must be decreased by 2. minimum BW 1.2 (20% overshoot) 7-15.07. (a) Stability requirement: Kt 7 0. 10 rad/sec.4106 rad/sec 9-9. (a) Breakaway-point equation: 2s5 20s4 74s3 110s2 48s 0 Breakaway points: 0.92 dB. Ans-3 CHAPTER 9 Frequency-Domain Analysis 9-1. 2. (a) The system is stable for 0K 0 6 1200. 4 rad/sec. GM 10. Kv K. ymax 1.

(b) KP 2. (a) R2 2. PM 93. (a) Set 1 aT 10 so that the pole of G(s) at s 10 is cancelled. (a) a 1000. (a) KP 1. BW 1.Ans-4 Answers to Selected Problems CHAPTER 10 Design of Control Systems 10-3. (b) KD 0.851 sec) when KP 0.9718 for stability. K 5.8 . (b) KD 399. (a) KD 0. 10-16.1.41 rad/sec 10-9.00348s 1 15. The new gain-crossover frequency 70 rad/sec. (a) Gc 1s2 10-43. (a) ts minimum (0. KI 200 . PM 31.51 10-13.222. (a) K1 11 1 117. T 0.00175 10-7. KI 100 10-22. n 45 rad/sec 1 0. KP 4.6.85.0487s 10-34. KP 50. Mr 1.65. 10-25.63s 10-40. (a) KP 10. KD 6. (c) Gc 1s2 1 0.09.0004 (c) a 0.93 10-20.13s 10-45. Kv 10. k3 20.074 10-30. (b) p 1. GM .65 105 gives a relative damping ratio of 0. KP 348. (b) Set the additional phase lead at 67 . KD 0.93 10-38.34 10-46.09 10-5.84. 10-27. (a) k1 64. KI 10.

Sign up to vote on this title
UsefulNot useful