Modern Control Theory

| First Edition - 2008 i Atco stello-lM ableolCo-leton ce atiitee| Table of Contents : Chapter -4 State Variable Analysis & Design (1-1) to (1-14) | Chagter-2 Stato Space Representation Chapter- 3 Matrix Algebra & Derivation of Transter Function Solution of State Equations Pole Placament Technique Chaptor-6 ——— Continlors Chapter-7 Nonlinear Systems Chapler-@ Phase Plane Method haptor-9 —_Liagunav’s Stabilty Analyst's Chapterwise University Questions with Answer Features of Book | # Excellent theory well supported with the practical examples and illustrations. # Important concepts are highlighted using key points throughout the book. #® Use of informative, self explanatory diagrams, plots and graphs. # Book provides detailed insight into the subject. % Stepwise explanation to mathematical derivations for easier | I I 1 \ ! i # Large number of solved examples. 1 | * Approach of the book resembles classroom teaching. \ I I I I I ' | L understanding. h Best of Technical Publications ae ee As per Revised Syllabus of VTU - 2006 Course 2] Semester - V [EEE] <& Modern Control Theory Bakshi 4 © Signals and Systems Chitode || |= Transmission and Distribution Bakshi | Linear ICs and Applications Bakshi, Godse i & D.C. Machines and Synchronous Machines Bakshi i (4S Bei Modern Control Theory ISBN 9788184315066 All rights reserved with Technical Publicotions. No port of this book should be reproduced in any form, Electronic, Mechanical, Photocopy or ony information storage ond retrieval system without prior permission in writing, from Technical Publications, Pune. Published by : Technical Pubtications Pune” . ‘#1, Amit Residency, 412, Shaniwas Peth, Pune - 411 030, Indiv. Printer : Abert DTPhintess: Seno. 10/3,Sinhagad Road, Pune - 411 047 ‘ESPreface The importonce of Control Theory is well known in vorious engineering fields. Overwhelming response to our books on various subjects inspired us to write this book. The book is structured to cover the key aspects of the subject Modern Control Theory. The book uses plain, lucid language to explain fundamentals of this subject. The book provides logical method of explaining various complicated concepts and stepwise methods to explein the importon! topics. Each chapter is well supported with necessary illustrations, prectical examples and solved problems. All the chopters in the book are arranged in a proper sequence that permits each topic to build upon earlier studies. All care has been taken to make students comfortable in understanding the basic concepts of the subject. The book not only covers the entire scope of the subject but explains the philosophy of the subject. This makes the understanding of this subject more clear and makes it more interesting. The book will be very useful not anly to the students but also to the subject teachers. The students have to omit nothing and possibly have to cover nothing more. We wish to express our profound thanks to all those who helped in making this book a reality, Much needed moral support ond encouragement is provided on numerous occasions by our whole family. We wish to thank the Publisher and the entire team of Technical Publications who have token immense pain to get this book in time with quolity printing. Any suggestion for the improvement of the book will be acknowledged and well appreciated. Authors U, A, Baksh A.D. Baksht Dedicated to Varsha, Pradnya and GururajTable of Contents Chapter? 1” Stats Variable Analysis & Design’ 1.1 BACKGIOUNG ......cccessssssestesesescessaseneetacaneeese 1.1.1 Limitations of Conventional Approach ...... 02... .0.s.s+sseseeeseeeee eeesees 1.1.2 Advantages of State Variable Analysis. 4.2 Concept of State ccs 1.2.1 Important Definitions... eee 1.3 State Model of Linear Systems........... . 1.34 State Model of Single Input Single Outout System... 1.4 State Diagram Representation ae2.4 State Model by Cascade Programming........ Examples with Solutions. BA Background 2... ce cece sees ceeeeecee sec seseeceseseeseesetesesseneesees .2 Definition of K 3.2.1 Types of Matrices... . 3.2.2 Important Terminologie: 3.3 Elementary Matrix Operations ............sssoessesesesseteeesssscneesesneessessnssnersnne 1B SAINVETSE Of Mn eatnnnannnnnstnannnnanannnnan d= E 3.4.1 Properties of Inverse of a Matrix. 3.5.1 Characteristic Equation .. 3.5.2 MIMO System... . 3.6 Eigen Values ...... 3.7 Eigen Vectors ....scscssncesssscesssscssnceressncesesssccscessessuensessuetiassareuetsascenesaeceses 3.9 Diagonalisation.... sone 3.10 Generalised Eigen V Vectors... 3.10.1 Vander Monde Matrix for Generalised Eigen Ve Vectors: Examples with Solutions.............00.... sesso Review Cc westions. 3 - 465 4.1 Background 4.1.1 Homogeneous Equation 4.1.2 Nonhomogeneous Equation 4.2 Review of Classical Method of Solution 0 4.24 Zero Input Response. ...ecceesssesesseeeesccesescesseeeeseens eens eeenes 4d 4.3 Solution of Nonhomogeneous Equation...... RE aa nae4.4 Properties of State Transition Matrix....... 4.5 Solution of State Equation by Laplace Transform Method 4.6 Computation of State Transition Matrix... esse 4.10 Similarity Transformation Method.......... 4.11 Controllability and Observability .. 4.12 Controllability... 4.12.1 Kalman's Test for Controllability... oes. sce ees eens seen se seeeee eee serees 4.12.2 Condition for State Controllability in s-Plane- 4.12.3 Gilbert's Test for Controllability. .......... o 4.42.4 Output Gontrollbllity ss. ccscseeescesceeceecseseesseeeesseseeesees 4.13 Observability 4.13.1 Kalman’s Test for Observability . 4.13.2 Condition for Complete Observability in s-Plane ..........-....-....--. 0s 4-33 4.13.3 Gilbert's Test for Observability Examples with Solutions.......... $.4.3 Ackermann’s Formula . 5.5 Selection of Location of Desired Closed Loop Poles. 5.6 State Observers a 5.7 Concept of State ODS@MNWEF un. eecesseseseesaseesesceseseneereesessesoesersees 9 2D.5.8 Dual Problem... 5.9 Necessary and Sufficient Condition for state ‘Observation 5.10 Evaluation of State Observer Gain Matrix Kg... 5.11 Selection of Suitable Value of Observer Gain Matrix Ke..........1.eese Examples with Solutions..........c0::0::cecuseseeuseeessesesesersceesceeersteesseceres 9 SB ter 60 | 6.1 Background ... seeecetoeseseeenes 6.2 Properties of Controller... 6.2.41 Error. 6.2.2 Variable Range... ce eee eee cee eee eee eee e cesses eeees senses 6.2.3 Controller Output Range - 6.2.4 Control Lag GES Applications... sees cseeesceseccecesseseesecsssseseeeessseeeessenesees 8.6 Integral Control Mode. 6.6.1 Step Response of Integral Mode. . 6.6.2 Characteristics of Integral Mode . . 6.6.3 Applications... cs seccecceececscesseceesseseceeeseeseeeeesvteseess 6.7.2 Asohatens set eees eee se ese accesses se esueses se sse sees eseeseesssseesene 6.8 Composite Control Modes... 6.9 Proportional + Integral Mode ro Control ol Mode). 6.9.1 Characteristics of P| Made6.9.2 Applications... 2.6... eee cece c cece ence e eee de reeset eeeetetenseeneeees 6-15 6.10 Proportional + Derivative Mode (PD Control Mode) uu... = 15 6.10.1 Characteristics of PD Mode. . 6-16 GAO.2APPHCAHONS 0... ee eee eee see ese sees see een nator renee eeetenceeseeneets 6-17 6.11 Three Mode Controller (PID Control Mode)..........:+1:ssssssssseseseuessessensen = 17 6.12. Response of Controllers to Various In| ane 6.13 Effect of Composite Controllers on 2" Order “System 6.14 PD Type of Controller...............+ 6.15 Pi Type of Controller... 6.16 PID Type of Controller... 6.17 Rate Feedback Controller + (Outout Derivative ‘Controller 6.18 Review of Second Order System Specifications ......... 6.19 Steady State Error.............00 Examples with Solutions...............000 7.1 Introduction to Nonlinear Systems......... 7.2 Properties of Nonlinear Systems........... 7.4.7 Nonlinear Spring 7.5 Absolute Value NonlinearityChapier'-'8: Phase Plane Method Se = 8.1 Introduction 8.2 Limit Cycle... 8.3 Jump Resonance 8.5 Basic Concept of Phase-Plane Methed ...........:s:00: 8.6 |socline Method for Construction of Phase Trajectory .. 8.7 Application of Phase-Plane Method to Linear Control System 8.8 Second Order Nonlinear System on Phase Plane ..............ssssessscsseseeses 8.9 Different Types of Phase Portrait 8.9.1 Phase Portraits for Type 0 System . 8.9.1.4 Stable System with Complex Roots. . 8.9.1.4 Stable System with RealRoots . . . . . os 8.9.1.5 Unstable System with Positive RealRools.. |. | |. 8.9.1.6 Unstable System with One Positive and One Negative Real Root 8.9.2 Forced Second Order Type 0 System. 8.9.3 Phase Portraits for Type 1 System . .. 8.9.4 Phase Portraits for Type 2 System ..... oo... sss eee eee eee tees 8.10 Singular Points of a Nonlinear System......... Examples with Solutions. .........scsssssssssenissessners Review Questions... “nied” 9.2 Stability in the Sense of LiapUMov ..................ccscceseesesseressseseeeissssnssonseneers 9.3 Asymptotic Stability .........cccssssesssssessnressnceserssaennsssenscecerssenrenesenontereereserss 9.4 Asymptotic Stability in the Large9.7 Som Important Definitions .......seee 9.7.2 Negative Definiteness... oo... ee cesses cece sees eee eeeteeeee eee eens 97.3 Positive Samide 9-5 9.7.4. Negative Semidefinite, 0... esse cece ees eee n er eee sees e tee eee eens eens 9-5 " 9°75 Indofinitences . 9-5 9.10 Liapunov's Second Method........... 9.11 Liapunov's Stability Theorem........ “9.42 Stabili 9.13 Construction of Liapunov's Functions for Nonlinear Systems by” of Linear and Nonlinear Systems .. 9.14 The Direct Method of Liapunov and the Linear System Examples with Solutions.Matrix Algebra & Derivation Modern Control Theory 3-7 of Transfer Function Properties af Multiplication 1. Commutative law is not valid for the matrix multiplication. AB + BA 2. The transpose of a product is the product of transposes of individual matrices in the reverse order. (aby’ = Bat ia Example 3.3: Obtain the matrix multiplication of, 7 auf ot “|2 10 Solution : A has order 3 x 2 and B has order 2 x 3 hence C = A x B has order 3 * 3. 3x1-1x2 3x0-1x1 3x-1-1x0] C = Ax B=!0x141x2 Ox04+1«1 Ox-14+1x0 2K14+0x2 2x0+0x1 2x-14+0«0 1-1 -3 =|]2 1 0 2 0 -2 3.4 Inverse of a Matrix In algebraic calculations we write ax = y ¥ ie. = Land x= a7 ie, x = 5 an ay Similarly in matrix algebra we can write, AX = B ie. x= A TB where Av! = Inverse of matrix A The inverse of a matrix exists only under the following conditions, 1. The matrix is a square matrix. 2. The matrix is a nonsingular matrix,‘Matrix Algebra & Derivation Modern Control Theory 3-8 of Transfer Function Mathematically inverse of a matrix can be calculated by the expression, 3.4.1 Properties of Inverse of a Matrix 1, When a matrix is multiplied by its inverse, the result is the identity matrix, ATA s aa? 2. Inverse of inverse of a matrix is the matrix itself. (wtytsa 3. Inverse of a product of two matrices is the product of their individual inverses taken in the reverse order. (aby! = Bo ATT 23 ima Example 3.4: Find the inverse of A = ? 1] U Solution : Check that A is nonsingular. 23 JA] = li j[=2-12=-10 1 _. [Adjoint of A] Ad. “ene Al Adj A = [Cofactor of A]? ETL 1-3 act, LA 2) [01 +08 ~~ =10 "140.4 =0,2 Cross check: AAT= [2 3][~O-1 0-3) [TO] | 4 1) [+04 -0.2 al Important Observation ; For 2 x 2 matrix, the adjoint of the matrix can be directly obtained by changing the positions of main diagonal elements and by changing the signs of the remaining elements.Matrix Algebra & Derivation Modern Control Theory 3-9 of Transfer Function 14 | 8 -4 For example, A= al: agjas| ‘| This result is directly used hereafter while solving the problems, when matrix has an order 2 x 2. Note that for any other order of matrix the adjoint must be obtained as transpose of cofactor matrix. With this study of matrix algebra, let us discuss the method of obtaining transfer function from the state model. 3.5 Derivation of Transfer Function from State Model Consider a standard state model derived for linear time invariant system as, X(t) = A X(t) +B U(t) sus (Ia) and YW) = CX) + DUM) ws (Ib) ‘Taking Laplace transform of both sides, [s X(s) ~ X(Q)] = A X(s) + BUG) ou (2a) and Y(s) = CX(s) + D Ute) ves (2b) Note that as the system is time invariant, the coefficient of matrices A, B, C and D are constants. While the definition of transfer function is based on the assumption of zero initial conditions ive. X(0) = 0. s X(s) = A X(s) + B Us) s X(s)— A X{(s) = B Us) Now s is an operator while A is matrix of order n x n hence to match the orders of two terms on left hand side, multiply 's' by identity matrix ] of the order n x n. sl X(s) = A X(s) = B Uts) [sl - A] X(s) = B U(s) Premultiplying both sides by [sl - AI’, [sl - AJ 1 [sl - A] X(s) = [sf - A}! BU(s) Now [sl - AP fst- A] = 1 Xs) = [sl - Al”? BU(s) -f3) Substituting in the equation (2b), ¥(s) = C [sl- AP ' B Ute) + D Uls) ¥(s) = (C [st- A! B+ D} Us)Matrix Algebra & Derivation Modern Control Theory 3-10 of Transfer Function Hence the transfer function is, +» (4a) -1 _ Adj[sl-A] Now [sl- A" = Ta-AT~ q¢) = CASISINAIB Lp vb) ~~ [st=aT Key Point: The state model of a system is not unique, but the transfer fusction of obtained from any stale model is unique. It is independent of the method used to express the system in state model form. 3.5.1 Characteristic Equation Tt is seen from the expression of transfer function that the denominator is |sI - A]. Now the equation obtained by equating denominator of transfer function to zero is called characteristic equation. The roots of this equation are the closed loop poles of the system. Thus the characteristic equation of the system is, a ..» Characteristic equation 4 isl { The stability of the system depends on the roots of the characteristic equation. In matrix algebra, the roots of the equation |sI - A| = 0 are called eigen values of matrix A and these are generally denoted by i i> =Example 3.5: Consider a system having state model -3] 1X, 3 _ X,] | [| . [¥ mayett it I with D = 0. Obtain its T.F. (VTU ; JanJ/Feb.-2008) Solution : TR. = C[st-ap! B Is-ay’ {sI- A] = 5Matrix Algebra & Derivation Modern Control Theory 3-11 of Transfer Function fs+2 3 [sl A] = 4 1] ag < [oR Ce " fs-2 4 jf [s-2 -3 1 ley Cy 3 s+2] “] 4 42 |sl-A] = (g +2) (8-2) 4 12=s? -4412=87 48 s-2 -3 _ Lt s+2 Ist- Ay" st 48 ny? 3) P 38-21 TR = 4 os+2 Jiri ay 5s + 22 “ 8748 33 +8 _ f8s+1} © shes 3.5.2 MIMO System For multiple input multiple output systems, a single transfer function does not exist. There exists a mathematical relationship between each output and all the inputs. Hence for such systems there exists a transfer matrix rather than the transfer function. But the method of obtaining transfer matrix remains same as before. imp Example 3.6: Determine the transfer matrix for MIMO system given by, x if? 3167 2 fn) Myf apa Xo} Le? ybe} ue ales bel Lt osx, Solution : From the given state model, 21 EK at D=(0] as[2a} eal T.M. CIsi- a}! B+ D s -3 Cy Cy] fats -2]" fst 3 Cy Ce} “L3 s] “[-2 5 37 458+ 6 = (s + 2) (s +3) Adj [sl - A] | sI-A]Matrix Algebra & Derivation Modern Control Theory 3-12 of Transfer Function Adj [sl- Al _ [ - 1 paliciek OP sake. [sla] JsI-A] (+2) (843) " f2 1)fs45 sft 1 o| -2 sit 1 TM. = C[sl-ay! B=* f2 alt: 3 ee 3s+14] 0 ua s-2 s-2 +8 s+8 | © 4se2)ise3) (s+ 2){s+3) 3s+14 3s+14 (s#2) #3) (+2) (S+3) s+8 s+8 (s#2) (8+3) (s+2) (8+3) ie. Y(s) = TM. U(s) 3s+14 3e+14 YG)] _ | F2GF3) G2) +S) [Ui Yo(s)} s+8 s+8 Us(s) (s+#2) (s+3) (+2) (s+3) ‘The above relation indicates that cach output depends on both the inputs. 3.6 Eigen Values Consider an equation AX = Y which indicates the transformation of 'n x 1' vector matrix X into 'n x 1' vector matrix Y by 'n x n' matrix operator A. If there exists such a vector X such that A transforms it to a vector 2X then X is called the solution of the equation, AX = 2X Le. AK-AX = 0 ie. [AI- AX = 0 (1) The set of homogencous equations (1) have a nontrivial solution only under the condition, jJAT-A] =0 we (2) The determinant [21 - A] is called characteristic polynomial while the equation (2) is called the characteristic equation.Matrix Algebra & Derivation Modern Control Theory 3-13 of Transfer Function After expanding, we get the characteristic equation as, JAIT-A] = +a, 7's. 4a, =0 QB) The ‘n’ roots of the equation (3) ie. the values of 4 satisfying the above equation (3) are called eigen values of the matrix A. The equation (2) is similar to |sI - Aj = 0, which is the characteristic equation of the system. Hence values of 4 satisfying characteristic equation are the closed loop poles of the system. Thus eigen values are the closed loop poles of the system. 3.7 Eigen Vectors Any nonzero vector X, such that AX, = 4 X; is said to be eigen vector associated with eigen value 2. Thus let 4 = A, satisfies the equation, Q,I-A)X = 0 Then solution of this equation is called eigen vector of A associated with cigen value 4j and is denoted as Mj. If the rank of the matrix [A; | - A] is 1, then there are (n - r) independent cigen vectors. Similarly another important point is that if the eigen values of matrix A are all distinct, then the rank r of matrix A is (n - 1) where n is order of the system. Mathematically, the eigen vector can be calculated by taking cofactors of matrix (A; 1 - A) along any row. Cy Ce Thus = Eigen vector for 4; = where k = 1,2, 0. Cha where Cy is cofactor of matrix (A, I - A) of k" row. Key Point: if the cofactors along a particular row gives null solution ie. all elements of corresponding eigen vectors are zero then cofactors along any other row must be obtained. Otherwise inverse of modal matrix M cannot exist. 3.8 Modal Matrix M Let Ay, Aa .- Ay are the n eigen values of the matrix A while M,, Mz, ... M, are the eigen vectors corresponding to the eigen values 4;, 22 ... 4, respectively. Each cigen vector is of the order n x 1 and placing all the eigen vectors one after another as the columns of another matrix, n x n matrix can be obtained. Such a matrix obtained by placing all the eigen vectors together is called a modal matrix or diagonalising matrix of matrix A.Matrix Algebra & Derivation Modern Control Theory 3-44 of Transfer Function Modal Matrix The important property of the modal matrix is that, > M4] [AM, : AM; : ... AM,] [ig My thy My se Ay Mal AM = A[M,:M,:.. " AM = MA () 0 0 where A. = Diagonal matrix : R Thus premultiplying equation (1) by M~?. M"lAM = M™'MA M7'AM = A (Diagonal matrix) It can be noted that, 1, Both A and A have same characteristic equation. 2. Both A and A have same eigen values. Hence due to transformation M“ 'AM, eigen values of matrix A remain unchanged and matrix A gets converted to diagonal form. 3.8.1 Vander Monde Matrix If the state model is obtained using the phase variables then the matrix A is in Bush. form or phase variable form as, 0 1 0 oO 0 0 1 0 Ags 0 0 0 - od vAn “nd “Ane =] And the characteristic equation i.e. denominator of T(s) is, F(s) = s"+ays"7 4. +ay_ysta,=0Matrix Algebra & Derivation Modern Control Theory 3-15 of Transfer Function In such a case, modal matrix takes a form of a special matrix as, Aya dey Ve qi 5 ts aplog? Such a modal matrix for matrix A which is in phase variable form, is called Vander Monde matrix . a> Example 3.7: Consider a state model with matrix A as, o 20 A= 4 oo” -48 -3M 79 Determine, a) Characteristic equation, b) Eigen values c) Eigen vectors and d) Modal matrix. Also prove that the transformation M~ 7AM results a diagonal matrix. Solution : a) The characteristic equation is |AT- A] = 0 10 0/}/0 2 0 ajo 1 o- 4 0 1 = 0 00 1) |-48 -3 -9 A -2 0 -@ A -al=0 ol) 484249 a (+9) 42% 4840+0-80+9)+ 9450 +98 + 264+ = 0 This is required characteristic equation. b) To find eigen values, test 4 = — 2 for its rootModern Control Theory 3-16 Matrix Algebra & Derivation of Transfer Function (+2) (24+ 74+ 12) = ie. (+2 A+3 G4) = ie. = These are the cigen values of matrix A. ¢) To find eigen vectors, obtain matrix {).; value of 4 in equation (1). -2 -2 0 For};=-2 [Aj1-A] = |-4 -2 -1 48 4 7 Cy M, =|C,.| where C3 20) f4 M, = |-20]=|-1 40] [-2 -3 -2 0 For 4,=-3, [AyI-A] =]-4 -3 -1 6 36 [Cy 16 M, =|Cj)|=|-24 C3 8 4-2 0 ForA,=-4, [AjI-A] =|4 4 71 48 34 5 Cy “ My =|Cy |=|-28 Cis 56 a 0 -2Ag=-3 and dg=-4 1 = AJ for each eigen value by substituting Cry Ca, Chg cofactors of row 1. «.. The common factor can be taken out. -3 My, M, and Mg are the eigen vectors corresponding to the eigen values 2, 42 and 23.Matrix Algebra & Derivation Modern Control Theory 3-17 of Transfer Function d) The modal matrix is, 1201 M = [Mj:M):MJ=|-1 -3 -2 214 Let us prove M~ 1AM is a diagonal matrix. mo! = AdiIMI _ {Cofactor of My T M. 10 +8 -7]" [-0 -7 =A Adj(M] = |-7 6 -5] =| 8 6 1 -1 i -1 7-5-1 1 21 |M| = j-1 -3 -2/=-1 -2 14 07 1 - Adj [MJ] Mls =|-8 6 -1 M 7 5 1 0 2 0 1 2 1 2-6 -4 AM =] 4 0 I||-1 3 -2}/=)/2 9 8 -48 -34 -9][-2 1 4 4-3 -16 M~'AM = |-8 -6 -1]]2 9 8 Thus M~'AM is a diagonal matrix. > Example 3.8: Obtain the Eigen values, Eigen vectors and Modal matrix for the matrix o 10 A=|0 0 1 -6 -ll +6Matrix Algebra & Derivation Modern Control Theory 3-18 of Transfer Function Solution : Eigen values are roots of |AI- A|=0 aA -1 «(0 |AI-A] = ]O 2 -1 J=0 6 1 A+6 Le. 0346.2 +1146 = 0 C+ 1A+ 2) (A+ 3) = 0 Ay =-1, Ay =-2, 44 =-3 are eigen values. To find Eigen vector, Let Ais kp=-t -1 -1 0 fAyI-A] = Jo -1 a 61 5 Cu M, = [Cy where C = Cofactor isdiez [6 1 M, = |-6| ie, M,=[-1 6 1 For 4, =-2 f-2 -1 0 [Ag I-A] = | 0 -2 -1 [6 lb 4 [C4 3 1 M; = |Cy2 =|-6| =|-2 C1 forK=3 2 4 3 -1 0 [asi-A] = fo 3 2 [6 ii 3Matrix Algebra & Derivation Modern Contro! Theory 3-19 of Transfer Function Cy 1 M3 = [Cay =|3 Cra Jor ke3 ° Key Polnt: The cofactors about any row can be obtained. +. Modal matrix M = [M, :M,:M,] 111 M =|-1 -2 -3 149 Tt can be easily checked that, 10 6 MTAM = A=/|0 -2 0 oo 3 3.9 Diagonalisation The diagonal matrix plays an important role in the matrix algebra. The eigen values and inverse of a diagonal matrix can be very easily obtained just by an inspection. Thus in state variable analysis, if the matrix A is diagonalised then it is very easy to handle mathematically. When matrix A is diagonalised, then the elements along its principle diagonal are the eigen values. The eigen values are the closed loop poles of the system, from which stability of the system can be analysed. The diagonal matrix A indicates noninteraction of various state variables. The derivative of one state variable is dependent on the corresponding state variable alone and other state variables do not interact in it, when A is diagonal matrix. This is practically important from controlling point of view. Due to all these reasons, the diagonalisation of matrix A is always motivated. The techniques used to transform the general state model into diagonal form i.e. canonical form are called diagonalisation techniques. Consider n™" order state model in which matrix A is not diagonal. AX() + BU) () cx) +DUe) @ Xn ¥(t) Let Z(t) is new state ~rector such that the transformation is, x) = M Z(t) vs» (3)Matrix Algebra & Derivation Modern Control! Theory 3-20 of Transfer Function Here M = Madal matrix of A Xt = M ZW = @ Using equations (3) and (4) into equations (1) and (2), MZ) = AM Zit) + BUCH we 5) and ¥() = CM Z(t) + DU) : we (6) Premultiplying equation (5) by M~' on both sides, Mo 'a zy = MAM 2() + M7! BUG) 2m = M7? AMZ()+M7 "BUH 2 and Y(t) = CM Z(t) + D UW) 8) The equations (7) and (8) gives the canonical state model in which M~ 1AM is a diagonal matrix denoted as A. The matrix M which transforms A into diagonal form is called diagonalising matrix or modal matrix. The new canonical state model is represented as, 2) = AZ) +BUH oe yw) = Ez) + DUM) ees (10) where A = M7lAM= Diagonal matrix B= M's t= cM The method of obtaining modal matrix M is already discussed in the last section. The transformation M~ 1AM is called diagonalisation of matrix A. imm> Example 3.9: Reduce the given state model into its canonical form by diagonalising matrix A, . 0 1-1 0 X(t) = |-6 11 6} X(t) +10) Ut) -6 -1l1 5 1 and Y(t) = [1 0 O} X(t)Matrix Algebra & Derivation Modern Control Theory 3-21 of Transfer Function Solution : From the given state model, 0 1-1 0 A= |-6 -1l 6], B=|0/, C=[1 0 0] -6 -11 5 1 Let us find eigen values, eigen vectors and modal matrix of A. 1oo}fo 1-1 |AT-A] = [4/0 1 O]-|-6 -11 6|)=0 001) [6 -1 5 [ho- 1 6 Atll -6] = 0 6 Mo 3-8 AQ + 11) (h~ 5) + 36 + 66 - 6 (A+ 11) + 62-5) + 662=0 B+6P+1+6 = 0 (A+1)Q+2 +3) = 0 0 Thus eigen values are, Ay =~ 1,42 =-2, 43=-3 To find eigen vectors, Fora, =-1 1-1 1 A, I-A] = | 6 10 -6 6 11 -6 For Ay = -2, 2-1 1 [Ag I-A] = r 9 6Matrix Algebra & Derivation Modern Control Theory 3-22 of Transfer Function For 43 = - 3, 3-101 [AsI-A) = 16 8 -6 6 8 Cy) f2) fi M; = |C,)|=[12]=16 Ci} {18} [8 111 M = [M, : M,: Mj]=|0 2 6] = Modal matrix 149 -1 0 0 M"'AM =] 0 -2 Olea 00-3 While B= M7B ~6 6-2)" [43 4 5 8 3 6 8 -6 met. AGM) La -5 2) 12 3 2 © TMT -14 * 4 0.4285 0.3571 ~0.2857 = |-0.4285 -0,5714 +0.4285 0.1428 0.2142 -0.1428 -0.2857 -2 B= M7'B=|+04285}=| 3 vs» Multiplying by 7 -0,1428] [-1 and Ct = CM=[i 1 1] Thus canonical state model is, 2) = Az) + Bu and Y(t) = © Z(t) where X(t) = M Z(t)Matrix Algebra & Derivation Modern Control Theory 3-23 of Transfer Function 3.10 Generalised Eigen Vectors The generalised cigen vectors are the eigen vectors corresponding to repeated eigen values. Uptill now, it is assumed that the eigen valucs are distinct but if an eigen value i, is repeated for r times then the corresponding eigen vectors are called generalised eigen vectors. If a particular eigen value 2, is repeated r times then the rank of n x n matrix [A,1- A] is n - 1 and there is only one independent cigen vector associated with 4, given by, Cu Cc M, = |-2 .. Cky are cofactors about k* row Cha 7 important Note: If cofactors along a particular row gives all values zero i.e. eigen vector as null matrix, then cofactors along other row must be obtained. For remaining (r - 1) eigen vectors for r - 1 times repeated value 44, generalised eigen vectors must be used as, 1 ay i PCy Tl Gay 2! an 2, 1 ac} oo [1 2 oe My = T ay M; =/2) an 1 ac, ae Th on 4, ‘ 1 2! an i arly) (Dt gar? ale aan gn () oar The modal matrix M then can be obtained by placing all the eigen vectors one after the other. Note that in such a case, M~ 1AM matrix is not diagonal but consists of a r x r Jordan block in it where r is the order of repetition of a particular eigen value,Matrix Algebra & Derivation Modern Control Theory 3-24 of Transfer Function 3.10.1 Vander Monde Matrix for Generalised:Eigen Vectors If matrix A is in phase variable form andvan eigen value 4, is repeated for r times, then n cigen values are Ly, 21 --- Aq Bp ay oly bye The corresponding modal matrix is Vander Monde matrix in modified form as, 1 0 0 Do one Lowe 4 ay i a Oe Ay oan x 2ay 1 Oa By oe M= * a yp-} aay! 1 cee el ye! a di, 2! az rel ” tmp Example 3.10: Find the eigen values, eigen vectors and modal matrix for, oo 1) Az=/20 0 82-5 Solution : Find eigen values which are roots of |21-A| =0 1oo0]fo o sll alo 1 0|--2 0 o|}=0 001] Ls -2 -5]f a0 - 22 O0}]-0 4802 045 2 +5) 4448 = 0 +50 +844 = 0 Try A=-1 (41) 02 + 444 4) (+1) A+ 27° uo ooMatrix Algebra & Derivation Modern Control Theory 3-25 of Transfer Function Ay = -LAg=-2Ag=-2 Thus '- 2' is repeated twice hence it is necessary to use generalised eigen vectors. -1 0 -1 For hy =-1,[4yl-Al=|-2 -1 0 24 Cy] [4] f- M, = {C,. f=} 8=| 2 G3} ta) Lt Now 2. = — 2 is repeated twice. My 0 Al Dgl- A] = ]-2 2, 0 8 2 Ay t5 Cn 13 +52, M; = ICy =| 2A, +10 Cashier | “4-8 hye-2 ~6] f-1 =| 6/=]1 12 2 245 +5 1 ore 2 and Ms = 5 a, |" 2 =| 2 8 je LS. 1 dCy; 2 Tl Gy Thus My, Mj and My are the required eigen vectors. a-101 . Modal matrix M = [M):M,:Mjl=| 2 1 2 1 2 -8Matrix Algebra & Derivation Modern Control Theory 3-26 of Transfer Function Examples with Solutions in Example 3.11: Find the TF. of the system having state model, s ool 1 X= [! a}e{ale and Y= [1 0] X Solution : From the given model, A= [; al: a=[j|-c-0 Oo) D=0 Ys) = “ese _ 1 TR = Cfsl-ay'B+D 107 f0 1) fs -1 {sl- A] = (i tH alk al s+] +1 . Adj [sl - A] = [" | . Change diagonal elements and change signs of other elements. |sl-Al = 5 (643) +2es?4 35425 (8+ (s+ 2) s+3 17fl “TR 2 Cadifst-ayp_ © af s}fo| “OS —TSI=AT st ist3) (s+3)} ~ (s+) (s+2) mm Example 3.12: Find the T.F. of the systent having state model, s [-3 1 0 xf op iy and Y=[1 0] X Solution : From the given model, A= [:2 ol: a-[}}. C={1 0, D=0 . IO _ art TR = ay = C lst AP B+D 10 —3 1 s+3 -1 Ist - A] = ‘(i No a7 2 |Matrix Algebra & Derivation Modern Control Theory 3-27 of Transfer Function . 3 1 : Adj [sI- A] = [5 wa |sI-A] = ss +3)+2=s? +3s+2=(+1)(s+2) role sali TR = CAdj[sl-A]B _ 72 s+3](1 [sI-A| (s+1) (+2) _ 1 ~ (s+i)(s+2) im Example 3.13: Obtain the transfer function matrix for the MIMO system having state model, . 2-10 -1 0 X=] 1 1 2)X+}1 O]U -1 O01 0 2 110 mi vf Sx Solution : From the given state model, 2-10 -1 0 1 1 2), Bali of, -1 01 02 11 0 c= fl pee A Adj [sI-A] B Transfer matrix = C [sl - Ay B+ D = SAdII-AI 10 2-1 [st-A] = sfo1 o/-[1 1 O00 1) [-1 0 (s-1)? s-3 s-1 7° AdjfsI-A] = | 541 (s-2(6-1) 1 -2 2(s-2) 8? -38+3Matrix Algebra & Derivation Modern Control Theory 3-28 of Transfer Function st-2s+1 -s41 -2 =| s-3 s*-3s+2 2s-4 s-1 1 s?-35+3 |sl-A| = (s-2)(s-1)-2+s-1 = s- 4s? 4 65-5 s*-2s+1 541 2 -1 0 Adj[sI- AJB = | s-3 9 s?-3s+2 0 28-4 10 s-1 1 s?-35+3}| 0 2 5745 -4 C Adj (st - A] B Transfer matrix s*-4s+5 4s-8 “s+2 2s? -6546 ~s? ~ 11 oy | 3 45 4 s* —-ds45 ds—8 101 2 —5+2 2s*-68+6 —3st5 0 4s-12 -s?+2 2s? -65+2 [-3s+5 ds-12 js? +2 2s? -65+2 33 4s? +6s-5 imu Example 3.14: Obtain the eigen values, eigen vectors and modal matrix for, o 1 0 3 0 2 -12 ~-7 -6 and prove that M~"AM = A = Diagonal matrix, — (WTU: July/Aug.-2007,Jan/Feb.-2008) Solution : For eigen values, [AISA] = Aa 0 3 2 2/2 12,07 A+6 0Modern Control Theory 3-29 Matrix Algebra & Derivation 20+ 6) + 243k 18 + 14 + 624 11k+6 Test K=-1, -1 1 (a+ 1) (2 + 52+ 6) ie. (+1) +2) (A+ 3) t Hence eigen values are, 2, = - 1, 42 = - 2, For the eigen vectors, dype-L fay 1-Al M Ag=-2 [gt-Al M, A3=-3, [A31- A) My M " of Transfer Function 0 0 u 6 0 0 age-3 f1-1 0 3 -1 -2 lz 7.5 Cy) P93] fa Cy) |=]-9] =]-1 Cat (9) [a 2-1 0 32 2 207 4 om 6 2 Cyy | =} -12] =| + Ci5 3} la 3-1 0 -3 +3 -2 207 3 Cy 5 1 Cyz | =|-15]=|-3 Ca} Las] £3 12 1 Modal matrix =|-1 +4 -3Matrix Algebra & Derivation Modem Control Theory 3-30 of Transfer Function i EEIAILintaleei En 9 6 5)" fo 5 2) -5 4-3 6 4 2 wot. Adi) [22 2) [2 3 2 SMP ae 45 25 1 =|-3 -2 -1 2515 1 AM. in uo o we 1 4 de 4 oo re 45°23 1) f-1 4-3] f-21 of o M lam = |-3 -2 -1/]/1 8 9]= 2515 1) [1 2 -9 Thus, M~'AM = A= Diagonal matrix itm Example 3.15: Find the eigen values, eigen vectors and modal matrix for, 401 2 Az=]1 0 2 i -i 3 (VTU; JanJFeb.-2005) Solution : For eigen values, JAI-A] = 0 A441 2 -1 2 2]20 -1 1 A+3 | AGA) (A-3)- 2-24 2-2434M-BH0 2-72 + 154-9 = 0 (4-1) (FP - 6449) = 0 A-)a-3% = 0 Ay = 122 =/3, 23 =3 Now 4, = 3 is repeated twice hence for 42 = 3, the generalised eigen vectors must be used. ‘Matrix Algebra & Derivation Modern Control Theory 3-31 of Transfer Function 3-1 2 For 2y fay I-A} = [-1 1-2 -1 1 -2 €,} fo M, = |[C,,/=/0 Cy 0 M, = |Cy}=|8 Now for hy =A = 3, use Oo a a ~ Ss 1 ! nny For i = 3, My = Hence the modal matrix M is,Matrix Algebra & Derivation Modern Control Theory 3-32 of Transfer Function ium Example 3.16: The Fig. 3.1 shows the block diagram of a control system using state variable feedback and integral control. The state madel of plant is, B)- (2 JRE: Y xy fo WJ Ix 2 i) Derive the state model of the entire system. ii) Derive the transfer function Y(s)/LI{s). yt) Fig. 3.1 Solution : i) The input of integrator shown is X. Ry = -1* [U()- Yn] = YO - UH w ) and rt) = -35X;—2X,- 15% Q From the plant model given, X= -3X, + D+ rl) vs @a) % = 4X, -5% afb) and YW = % 2 Be) Using equation (2) in equation (3a) we get, Xp = — 3X, + 2X35 Xy- 2% - 15% X= - 5X, + 05X; - 3.5%; vs @a) Xy = AX, - 5% ws (8b) Substituting equation (3c) in equation (1), Xy = %- UI a (de)Matrix Algebra & Derivation Modern Control Theory 3-33 of Transfer Function and YW = X wn (4) Thus equations (4a) to (4d) give state model of the entire system. Xi -5 0.5 -35] [x,] [ 0 Xe} =] 4 -5 0] |X, ]+) o] Ug Xa 0 1 of {xs} [4 Xx) and Y(t) = [0 1 O} |X, X3 ie. X = AX4+BU and Y=CX withD=0 ii) The transfer function Y(s)/U(s) is, Ys) _ 1 ue = Cfsi-AP'B+D 1 0 0} f-5 05 -3.5] fs+5 -0.5 35 [l-A] =s]0 1 Of-) 4 -5 0 |=|-4 s45 0 001 a 1 O 0 -1 s Oss i ods to o4 7 Adj [sI- A] =|0.5s-3.5 $ s?+5s i s45 “3.5(8+5) i -1d i (845)? -2 s?+5s 0.5s-3.5 -3.5s-17.5 =| 4s s? +5s -14 4 st5 s* +10s+23 |sI- A] = 5(545)'+ 14-28 =5°+ 10s? + 2384+ 14 Ys) _ CAdj[sl-A] B U@ > T1si-AT and D=0 s?45s 0,583.5 -3.5s-17.5 |[ 0 C Adj (si- A] B = [0 1 0] |4s s'+5s 0 =id 0 4 s45 s? +10s+23 ||-1 =14 ¥(s) 14 4 Uls) ~ [sl-Al 53 420s? 4.235414Matrix Algebra & Derivation Modern Control Theory 3-34 of Transfer Function | _ 1d ~ (s#ljis+2)(s+7) This is the required transfer function. 0 21 0 tm Example 3.17: For the matrixA=|0 0 1 24 -3 Find the eigen values, eigen vectors and modal matrix M. Solution : For eigen values, JAI-A] = 0 a -1 0 oa -1/=0 2 4 443 BG+3) +244 = 0 #4344042 = 0 Test 4=-1, (+1) 02+2142) = 0 +1) Q+1+ jl) A+1-j1) = 0 Aye -L Calculate eigen vectors. For 4, =-1, -1 -1 0 Pyl-aAj =] 0 -1 -1Matrix Algebra & Derivation Modern Control Theory 3-35 of Transfer Function 1-jl -1 0 For dg = -A-jl, Pgl-Al=| 0 jl 1 2 4 2p Cy 1 Mz = [Cy |=|-1-j Cy j2 .-. Choosing 3 row to calculate cofactors. 14jl 0-1 0 ForAg=-1+jl,[Ag3I-Al=| 0 -1+jl1 1 2 4 24j1 Cy i My = |Cy) |=|-14j C33] [| -i2 a. M = |-1 -1-j -1+j 1 j2 -j2 nm ~Example 3.18 : For a system with state model matrices. “1 0 1 0 ay A=|1 -2 0/;B=|0];C=}1 0 0 3 1 0 Obtain the system transfer function. (VTU: March-2001) Solution : The TF. is given by, TR. = C[sl-Ay's s+1 0 -1 [sl- A] =| -1 s+2 0 Q 0 s-3 Adj[sI-A] = [Cofactor st—A]™ (s+2)(s-3) — (s-3) 0 * 0 (s+1) (8-3) 0 Hs4+2) +1 (s+1) (s+2)Matrix Algebra & Derivation Modern Control Theory 3-36 of Transfer Function (s+2)(s-3) 0 (s+2) = (s-3) (s+1) (6-3) 1 0 0 (s+1) (s+2) |sI-A| = (+1) (6 +2) (6-3) - _ Adjfsl—A] [sI-AF = Iza] 1 1 s+1 9 (s#1)(s—3) 1 1 1 G+HG+2) (+2) G+) G+26-3) a 0 1 3-3 1 0 1 0 st1 (s+1) (8-3) 1 1 1 TR =f 1 Ol) per Gr GED e-a]|" 0 a i 3-3 1 1 G+l) 6-3) fo. 9 1 (s+) (8 #2) (8-3) 1 (s-3) a * (41 G-3) | GtlG+D -3) (s+3) (s+1) (s+2)(s—3) map Example 3.19: Find the transformation matrix P which will convert the following matrix A to diagonal form. The eigen values are dy, hp and.d3. 0 1 0 A=| 0 0 1 1 Ay AgMatrix Algebra & Derivation Modern Control Theory 3-37 of Transfer Function Solution : The matrix AI - A is, aA A 0 (I-Ap=/0 A = a, a, Atay To find P, means to find cigen vectors and modal matrix. a, 1 0 Fork=Ay [Ay I-A]=|0 a, -1 a, a) Aytay To find eigen vector, find cofactors about 3 row. Cy 1 My = [Ca] =]Ay Cy x hy 1 For A= Az[kg I-A] =| 0 2 Cy 1 My = [Cay |=] Ay Cy 4 en) Forh=Ay fAjI-A} = |0 a; -1 a; a; Asta;Matrix Algebra & Derivation Modern Control Theory 3-38 of Transfer Function Hence transformation matrix P is, to. Pla, 2p Ay Bow Thus if matrix A is in phase variable form, fa 1 QO 0 0 0 To 0 Asli: a 0 0 ae oT Hy yp yg Oy And the eigen values are distinct as 44, 42, ... 4, then it can be proved as above that the diagonalising matrix is, 1 1 1 AL OA, ey grok yal at Key Point: Note that while finding the eigen vectors find the cofactors about last row. me} Example 3.20: Determine the transfer matrix for the system. ees S] Ce) RR ES 2 (VTU: July/Aug.-2005) Salution : From the given state model, 3 1 (* 6 1-1] As B= .C= ,D= [2 al , cs a ' 1) patel [sl- A]? = Adj [sl- A] _fse3 +] [sI=A] ‘BI-AI= , s . s 2 jf fs 4 Adj[st-A] = I <3| = (: wa 4Matrix Algebra & Derivation Modern Control Theory 3-39 of Transfer Function 1 ] fs 1 s+3 elo +3 jsi-ay! = s? 43542) (9#1)(5+2) 1 -1)fs 1 4 6 TR = Cfsl-ay'B=|8 1][2 s+3}[5 0 (stl) (s+2) 1-1) [as-5 65] — [8 1} j-Ss-23 -12} ~ (st+i)(s+2) 9s+18 6s+12 9 6 LTR = (s+1) (s+2) (s+1)(s+2) _ s+1 s+ TONS) 278-63 48512 *) 27-63 48-12 (s+1)(s+2) (s+1) (+2) (s+1)(s+2) (s+1} (s+2) This is the necessary transfer matrix. hep Example 3.21: Convert the following state model into canonical forin. (VTU: July/Aug.-2005) » fl -4 ° _ X= is -6 Xe fu veo a] x Solution : From the given model, Bde fipere Let us find the eigen values, eigen vectors and model matrix of A. 10) fl-4]|_ pe. 4 IM Al = li HL “di-b adn? (A-1)(X+6)412 = Die 27 +524+6 = 0 (+2) A043) = Obie Ay =2 4, =3 To find eigen vectors, 34 c For A, =-2, [2I-A] = 5 ‘| ie. m= [C5]Matrix Algebra & Derivation Modern Control Theory 3-40 of Transfer Function mesons [3 oem SU 41 M = [M):M,]= [4 ,]= Modal matrix 2 6] MAM = _3|=4 While B= MB Mo = i f i [574] Be [ “dl (.] -[)] € =cM=[1 0] [3 ‘}-8 1] The canonical state model is, At) = azit) + Bu Y(t) C Z(t) where X(t) = M Z(t) tus Example 3.22: Convert the following square matrix A into Jordan canonical form using asuitable non-singular transformation matrix P. (VTU: July/Aug.- 2005) Ke 1 0 A= 1/0 0 i |- 4-9 -6 Solution : Find the eigen values of A. aA o | |AI-A| = jo a -1 |=0 4 9 A+6 2(A46)44492.20 ie. 2 +602 4924420 . (AA)(RA1)(A41) =0 ie. Ay sd, Ay =-1. Ay =-1Modern Control Theory ForA,=-4 -4 [ai-a] = |o 0 For A, =-1, he [A I-a] = jo 4 fc M, = |c Cc As 1, =-1is repeated twice, mie Adj [M] -4 12 Matrix Algebra & Derivation 3-41 of Transfer Function -1 0 Cy] fa —4 -1] 2M, =]C))|=|-4 9 2 Cy] [16 -1 0 hy nl 9 4, +6 n 1B +61, +9 4 1 RD = -4 =|=-4)]=/-1 —4D. 4] [1 33h 2y 2 Soe dc, dh ac, 2h, +6 4] fi —2)=| 0 =| O}=| 0 ae 4 4} [-1 dC,, dey -42yo of 2 1] -1715) =|-4 -17-4 -43 2 15° 3 201 -17-4 15 3 9 201 o 1 o]fi a1 -17-4 0 0 1] f-4-1 0 15 3) [+4 -9 -6| [6 1 -1Matrix Algebra & Derivation Modern Control Theory 3-42 of Transfer Function ..-Diagonal matrix with Jordan block => Example 3.23 : Consider the matrix (VTU: Jan./Feb.-2006) 2-2 3 Azejl 1 1 1 3-1 i) Find the eigen values arid eigen vectors of A. it) Write the modal matrix. ifi) Show that the modal matrix indeed diagonalizes A. Solution : For eigen values | M-A|= 0 A-2 20-3 Hl 4-1-1) = 0 Hl -3 Al] ie. (R-L)(AF1)( R= 2) + 2-9 HA- 14 YAFT)= A= 2)=0 ie, -20? -5A+6=0 ie (A-1)(At+2)(A~3)=0 Apetl, Ag =-2, hy =3 For 2, =1, f-1 2 -3 Cy] f-3 -1 (AI-A] = |-1 0 -1 2 My={C, {=| 3) =) 1 1-32 yf L3l [a For dg =-2, “4 462-3 Cy] fo (AI-A] = |-a -3 -1 2 Mg=|Cy, |=] 0 -1 -3 -1 C5 0 As Cy, = Cy) = Cys = ©, calculate cofactors about other row. Cy} fu Mz = |Cy,|=|1 Cy -14Matrix Algebra & Derivation Modern Control Theory 3-43 of Transfer Function For 4, =3, 12-3 foul §) A [Al-A] = [-1 2 -1 Ne=|im =|5[=/1 “1-3 4 3} if lt “1 Mit M = [M,:M, :M;]=]1 101 1 -M1 wet = Adi M__ [Cofactor matrix of Mj! rs] iM , fis -25 = aylo -2 2 -15 -3 -B , [5-25 to [2-23] pau t MTAM = -4,/0 -2 2 [1 11 14. “ [-15 -3 -12][1. 3-1] [9 -14 1 15-25 10 -}Mo4 1 30 0 #0 1400 30 0 60 O}=]0 -2 O)=A G 0 -90 o 0 3 Thus modal matrix indeed diagonalizes A. y Example 3.24: Given the R=AX—Bu,y = am Examp Given the state modet X = AX—Bu, y =CX ce Oo ol 0 0 where A=/0 O 1/,B=|OjandC=[l 0 0] -1 -2 -3 1 — Ys). bees i) Simulate and find the transfer function Te using Mason's gain formula. ii) Determine the transfer function from the state model formulation.Matrix Algebra & Derivation Modern Control Theory 3-44 of Transfer Function Solution : The state equations are, Xp sXy. X} Hence the signal flow graph is, From Mason's gain formula, Ys) _ Tay + Tag ht TA Tis) ~ a Here K=1, — T, = Forward path gain = The various loop gains are, Jol Fig. 3.3 All loops are touching to each other. 201 As1-[L,+L,+L3] 214344 s s? Ay = Eliminating all loop gains from A which are touching to K' forward path Ay =iMatrix Algebra & Derivation Modern Control Theory of Transfer Function Ys) _ UB) 14+=+ si 43s? +2541 3 Now let us use the state model method of finding the transfer function. Y(s) I Ug * Cist-Al'B+D D=0 s -1 0 [sI-A] = fo s -l 1 2 s+3 2 T fez s*+3s+2 0 =I 38 Ss" +3542 s+3 1 Adj [sI-A] = |s4#3 s(s4+3) -2s—1) =/-1 s(8+3) 8 1 s s? -s -%s-1 0s? 32 435+2 s+3 1 -1 s(s +3) s i[st- -2s-1 fi-ap! = Adj[st=A} _ |-s 8 |sl- A] $3 4357 42541 s?43s¢2 543 i|fo Ys) _ fl 0 Off-1 s(s +3} silo Uis} © sy —a-1 0st fit s>+35s* +2541 1 ..Same as obtained above $3438? +2541 1 2 2 -3 mm Example 3.25 : The vector| 2| is an eigen vector of A=| 2 1 -6| Find the eigen -1 -1 -2 oO value of A corresponding ta the vector given. (VTU: July/Aug.-2006) Solution : The eigen vector is that vector which is non-zero such that AX, = 4,X, 1 where X, = Eigen vector =| 2 {4Matrix Algebra & Derivation Modern Control Theory 3-46 of Transfer Function 2, = Eigen value = To be obtained 2 2 -3)f1 1 2 1 -6/] | 2)=a) 2 -1-2 0 [-1 -1 5 om io] = 12a, 5 -hy eS ..Corresponding eigen value. Review Questions . Derive the transfer function from state model. What is characteristic equation of a system matrix A ? Define eigen values and eigen vectors of a matrix. When the generalised eigen vectors are used ? How 1o wse them ? What is modal matrix ? State its importance. State the advantages of diagonalisation of a matrix. How to achieve diagonatisation of matrix A? Rediice the gioen state model in diagonal form, SNP ew ND : 01 1 x= [° t)x[ LQ) and Y(t) = [1 0} Xi) 9. What is Vander Monde matrix ? When it exists ? What is its impertance ? 10. Find the T.F. of the systems with following state models, oe o kK [.: je [ile yen ox . 0-1 1 6b) X= = ) [{ a )e+(de y=10 UX . 0 0 2 1 Q X= |-1 0 O|X+)0)U, y=f1 0 OX 4-2-5 0 1 - Obtain the TF. of the system having state model, . 5-1 2 X(N) = [; =x «[5| ut) YW) = (1 2) x) [ans 128+59 } (s+ 2)(s+ 4) Q00Solution of State Equations 4.1 Background Uptill now the methods of obtaining state model in various forms and obtaining transfer function from the state model are discussed. It is seen that the output response. depends on the state variables and their initial values. Hence it is necessary to obtain the state vector X(t) which satisfies the state equation X(t) = A X(t) + B U(t) at any time t This is called solution of state equations, which helps to obtain the output response of a system. Consider the state equatio n of linear time invariant system as, XW = AX) + BUW The matrices A and B are constant matrices. This state equation can be of two types, 1. Homogencous and 2. Nonhomogencous 4.1.1 Homogeneous Equation If A is a constant matrix and inpvt control forces are zero then the equation takes the form, x) = AX) ea(1) Such an equation is called homogeneous equation. The obvious equation is if input is zero,how output can exist ? In such systems, the driving force is provided by the initial conditions of the system to produce the output. For example, consider a series RC circuit in which capacitor is initially charged to V volts. The current is the output. Now there is no input control force i.e.. external voltage applied to the system. But the initial voltage on the capacitor drives the current through the system and capacitor starts discharging through the resistance R. Such a system which works on the initial conditions without any input applied to it is called homogeneous system. (4-1)Modern Contro! Theory 4-2 Solution of State Equations 4.1.2 Nonhomogeneous Equation If A is a constant matrix and matrix U(t) is non zero vector i.e. the input control forces are applied to the system then the equation takes normal form as, Xt) = AX(Q+ BUC) a Q Such an equation is called nonhomogeneous equation. Most of the practical systems require inputs te drive them. Such systems are nonhomogeneous linear systems. The solution of the state equation is obtained by considering basic method of finding the solution of homogeneous equation. 4.2 Review of Classical Method of Solution Consider a scalar differential equation as, dx SR = ax where (0) = x9 @) This is a homogeneous equation without the input vector. Assume the solution of this equation as, x(t) = bo + byt + bo +... + by + Q Att=0, x0) = x9 = bg -- QB) ‘The solution has to satisfy the original differential equation hence using (2) in (1), a [bp + Byte + BT = a fby + byt +... + byt) by + byt +... + kbyth") = aby + abjt +... + abt For validity of this equation, the coefficients of various powers of ‘t' on both sides, must be equal. by = aby, 2by = aby, ... kby = aby} by = aby by = Sab) = Sates = 2" by b, = Lakby and x(0) =byState Variable Analysis and Design 1.1 Background The conventional approach used to study the behaviour of linear time invariant control systems, uses the time domain or frequency domain methods. In all these methods, the systems are modelled using transfer function approach, which is the ratio of Laplace transform of output to input, neglecting all the initial conditions. Thus this conventional analysis faces all the limitations associated with the transfer function approach. 1.1.4 Limitations of Conventional Approach Some of its limitations can be stated as = 1) Naturally, significant initial conditions in obtaining precise solution of any system, Joose their impartance in conventional approach. 2) The method is insufficient and troublesome to give complete time domain solution of higher order systems. 3) It is not very much convenient for the analysis of Multiple Input Multiple Output systems. 4) It gives analysis of system for some specific types of inputs like Step, Ramp etc. 5) it is only applicable to Linear Time Invariant Systems. 6) The classical methods like Root locus, Bode plot etc. are basically trial and error procedures which fail to give the optimal solution required. Hence it is absolutely necessary to use a method of analysing systems which evercomes most of the above said difficulties. The modem method discussed in this chapter uses the concept of total internal state of the system considcring all. initial conditions. This technique which uses the concept of state is called State Variable Analysis or State Space Analysis. Key Point: State variable analysis és essentially a time domain approach but it has number of advantages compared to conventional methods of analysis. (1-1)Moder Control Theory 1-2 State Variable Analysis & Design 1.1.2 Advantages of State Variable Analysis The various advantages of state variable analysis are, 1) The method takes into account the effect of all initial conditions. 2) It can be applicd to nonlinear as well as time varying systems. 3) It can be conveniently applied to Multiple Input Multiple Output systems. 4) The system can be designed for the optimal conditions precisely by using this modern method. 5) Any type of the input can be considered for designing the system. 6) As the methed involves matrix algebra, can be conveniently adopted for the digital computers. 7) The state variables selected need not necessarily be the physical quantities of the system. 8) The vector matrix notation greatly simplifies the mathematical representation of the system. 1.2 Concept of State Consider a football match. The score in the football match must be updated at every instant from the knowledge and information of the total score before that instant. This procedure of updating the score continues till the end of the match when we get exact and precise score of the entire match. This updating procedure has main importance in the understating of the concept of state. Consider the network as shown in the Fig. 1.1 Vn & c Zero fricon Fig. 1.2 To find V4, the knowledge of the initial capacitor voltage must be known. Only information about V,, will not be sufficient to obtain precisely the V,., at any time t20. Such systems in which the output is not only dependent on the input but also on the initial conditions are called the systems with memory or dynamic systems. While if in the above network capacitor “C' is replaced by another resistance “R,' then output will be dependent only on the input applied V,,,. Such systems in which the output of the system depends only on the input applied at t = 0, are called systems with zero memory or static systems.Modern Control Theory 1-3 State Variable Analysis & Design Consider another example of simple mechanical system as shown in the Fig. 1.2 Now according to Newton's law of motion, F = Ma a = Acceleration of mass M d F = M4 (vi) Le. v() = as f() dt Now velocity at any time “t' is the result of the force F applied to the particle in the entire past, t t t Wey fmas § ProareZ pana om ce is where t,, = Initial time. t, Now u pay dt = vit) t vit) = vty) + Jf) dt to From the above equation it is clear that for the same input f(t), we get the different values of the velocities v(t) depending upon our choice of parameter t, and the value of v(t). If v(ty) is known and the input vector from “ty! to “fis known then we can obtain a unique value of the output v(t) at any time > to. ; Key Point: Thus initial conditions ie. memory affects the system characterisation and subsequent behaviour. Thus initial conditions describe the status or state of the system at t = t,. The state can be regarded as a compact and concise representation of the past history of the system. So the state of the system in brief separates the future from the past so that the state contains all the information concerning the past history of the system necessarily required to determine the response of the system for any given type of input. The state of the system at any time “t’ is actually the combined effect of the values of all the different elements of the system which are associated with the initial conditions of the system. Thus the complete state of the system can be considered to be a vector having components which are the variables of system which are.tlosely associated withModern Control Theory 1-4 State Variable Analysis & Design initial conditions. So state can be defined as vector X(t) called state vector. This X(t) i.c. state at any time “t' is ‘n’ dimensional vector i.e. column matrix n x 1 as indicated below. X(t) Xp(t) X(t) = Xa lt Now these variables X,(t) , X(t), .... called the state variables of the system. If state at t = ty is to be decided then we must know X(ty) and knowledge of the input applied between ty —t,. This new state will be X(t,) which will act as initial state to find out the state at any time t > t,. This is called the updating of the state. The output of the system at t = t, will be the function of X(t,) and the instantaneous value of the input at t = t,, if any. The number of the state variables for a system is generally equal to the order of the system. The number of independent state variables is normally equal to the number of energy storing elements (eg.: capacitor voltage, current in inductor) contained in the system. . X,(t) which constitute the state vector X(t) are 1.2.1 Important Definitions 1) State ; The state of a dynamic system is defined as a minimal set of variables such that the knowledge of these variables at t = ty together with the knowledge of the inputs for t 2tg, completely determines the behaviour of the system for t >t). 2) State Variables : The variables involved in determining the state of a dynamic system X(t), are called the state variables. X)(t), X(t) ....X,,(t) are nothing but the state variables. These are normally the energy storing elements contained in the system. 3) State Vector : The ‘n’ state variables necessary to describe the complete behaviour of the system can be considered as ‘n' components of a vector X(t) called the state vector at time ‘t’. The state vector X(t) is the vector sum of all the state variables. 4) State Space : The space whose co-ordinate axes are nothing but the ‘n’ state variables with time as the implicit variable is called the state space. 5) State Trajectory : It is the locus of the tips of the state vectors, with time as the implicit variable.Control Theory 1-5 State Variable Analysis & Design Definitions can be explained by considering second order system : ‘Order is 2 so number of state variables required is 2 say X,(t) and X(t). State vector will be the matrix of order 2 x 1. x0 x) = {07 i i ~ > 3 ie. X() = X\()+XZK) (vector addition) The state space will be a plane in this case as the number of variables are two. Thus state space can be shown as in the Fig. 1.3. x x0 X (ty) % (h), Xt Xl) (9.0) 2 Katy) Fig. 1.3 Fig. 1.4 Now consider t = t, > 4 ~ ity) = Xy(ty) + Xplty) x (0 X (la) Xs (toh, nans: 0 Xq (ty) Fig. 1.5 Fig. 1.6 Now consider t = t, > > > X(ty) = Xp(ty)+ Xp (ty) The state trajectory can be shown by joining the tips of the two state vectors ie. X(t,} and X(t),Modern Control Theory 1-6 State Variable Analysis & Design 1.3 State Model of Linear Systems Consider Multiple Input Multiple Output, nth order system as shown in the Fig. 1.7. Number of inputs = m Number of outputs = p Uy U2 Inputs Urs Hy Xp X3 Xa State variables Fig. 1.7 UAH x0 ¥0) U,(t) X(t) ¥,) um =]: bx®=) : |= Un tt) Xn (0) Yp(0) All are column vectors having orders mx 1, nx 1 and p x 1 respectively. For such a system, the state variable representation can be arranged in the form of *n first order differential equations. aX) s . =n = X(t) = fy Ope Xp prereeXy p Uy (Ug vorren Un) dX,() + Gp = XR = fy Kp Xp ee Ny Uy Uy creer Und dX, (0 Kant) = fy Xp Xp pee Xq /Uy Uy oe dtModern Control Theory 1-7 State Variable Analysis & Design f h Where f =| ; | is the functional operator. fh, Integrating the above equation, t Xi = Xilgh + PO, Xq see My Uy , Uy vor Un) dt 'o where i = 1, 2, net Thus ‘n' state variables and hence state vector at any time “t' can be determined uniquely. Any “n' dimensional time invariant system has state equations in the functional form as, Xt) = £(X, U) While outputs of such system are dependent on the state of system and instantancous inputs. .. Functional output equation can be wrilten as, Y(t) = g(X, U) where “g' is the functional operator. For time variant system, the same equations can be written as, . X(t) = £(X, U, #) ... State equation 0) g(X, U, t).... Output equation Diagramatically this can be represented as in the Fig. 1.8. Ui) Instantaneous: Input Input Uti) vit) f Output X(t) Initial state: Fig. 1.8 Input-output state description of a systemModern Control Theory 1-8 State Variable Analysis & Design The functional equations can be expressed interms of linear combination of system states and the input as, XL = ayy Xy + ayQXq +. + ayX, + b + byU, +... + bin Un Ko = agg Xp + ayyXp oa. + aggXq + byyUy + bay +... + Bam Urn * Xn = ayy Xp + aygXg t+ -. + ALK, + byyUy + bygUy + ... + Bag Uy For the linear time invariant systems, the coefficients ag and bj are constants. Thus all the equations can be written in vector matrix form as, Xa) = AX) + BUM where X(t) = State vector matrix of order n x 1 U(t} = Input vector matrix of order m x 1 A = System matrix or Evolution matrix of order n x n B = Input matrix or control matrix of order n x m Similarly the output variables at time t can be expressed as the linear combinations of the input variables and state variables at time t as, YV y(t) = cy, X(t) + oe + cay X(t) + dyy Uy(t) +... + dy, UL tt) Ypll) = cpr NaCl) + ee + cpr Null) + dy Uylt) + oo. dogg Unpl t For the linear time invariant systems, the coefficients cj and dj are constants. Thus all the output equations can be written in vector matrix form as, YW) = CX() + DU where Y(t) = Output vector matrix of order p x 1 C = Output matrix or observation matrix of order p x n D = Direct transmission matrix of order p x m The two vector equations together is called the state model of the linear system. x) =AX() + BUY - State equation Y(t) = C X(t) + D U(t) ...Output equation ‘This is state model of a systern.Modern Control Theory 1-9 State Variable Analysis & Design For linear time-variant systems, the matrices A, B, C and D are also time dependent. Thus, Xi) = AC) X0)+80 UO | For tnear time variant system Y(t) = C(t) X(t) + D(t) Ui) 1.3.1 State Model of Single Input Single Output System Consider a single input single output system ie. m = 1 and p = 1. But its order is ‘n’ hence 1 state variables are required to define state of the system. In such a case, the state model is Xt) = AX() +B UW YW = CX +d UW) where A = nx nmairix, B=n x1 matrix C = 1% n matrix, d = constant and U(t) = single scalar input variable In general remember the orders of the various matrices. A= Evolution matrix =nxn B = Control matrix =nxm C = Observation matrix = p xn D = Transmission matrix => p x m 1.4 State Diagram Representation It is the pictorial representation of the state model derived for the given system. It forms a close relationship amongst the state model, differential equations of the system and its solution, It is basically a block diagram type approach which is designed from the view of programming of a computer. The basic advantage of state diagram is when it is impossible to select the state variables as physical variables. When transfer function of system is given then state diagram may be obtained first. And then by assigning mathematical state variables there in, standard state model can be obtained. State diagram of a linear time invariant continuous system is discussed here for the sake of simplicity. It is a proper interconnection of three basic units. i) Scalars ii) Adders _ iii) IntegratorsModern Control Theory 1-10 State Variable Analysis & Design Scalars are nothing but like amplifiers having required gain. . xt. —[» /}—~ Xtdeaxgy — Xpltd * X(t) = (1) + Xs) + x4) (a)Scalar (b)Adder Fig. 1.9 Adders are nothing but summing points. Integrators are the elements which actually integrate the differentiation of state variable to obtain required state variable. Now integrators are denoted as “1/s' in Laplace transform. So transfer function of any io —--[ fF }-— X(t) = 1K at integrator is always 1/s. With these three basic units we can draw the state diagrams of any order system. Fig. 1.9 (c) Key Point; The output of each integrator is the state variable, 1.4.1 State Diagram of Standard State Model Consider standard state model XW) = AX() +B U() and Y(t) = C X(t) + D U(t) So its state diagram will be as in the Fig, 1.10. ‘uth Fig. 1.10 State diagram of MIMO system The thick arrows indicate that there are multiple number of input, output and state variables, There must be n parallel integrators for n state variables. The output of cach integrator is a seperate state variable. Jf such a state diagram for the system is obtained then the state mode! from the diagram can be easily obtained.Modern Control Theory 4-11 State Variable Analysis & Design Remarks 1. To obtain the state model from state diagram, always choose output of each integrator as a state variable. Number of integrators always equals the order of the system i.e. ‘n' 2. Differentiators are not used in the state diagram as they amplify the inevitable noise. ‘o> Example 1.1: Obtain the state diagram of SISO system represented by equations, Xu) = a, X4b, UY, Ket =a, X,() +a, X,(t) 4b, UW anid VY) = cy Xy(t) + cy XQ) Solution : As there are 2 state variables X,(t) and X(t), the two integrators are required. Fig. 1.14 1.5 Non Uniqueness of State Model Consider a standard state model for a system " xy yi} Let X = M Z where M is a non singular constant matrix. AX +BUQ (1) CX) + DUH ~@ " Key Point: A matrix whose determinant is nonzero is called nonsingular matrix. This means Z is new set of state variables which is obtained by linear combinations of the original state variables.Modern Control Theory 1-12 State Variable Analysis & Design X = MZ hence X=MZ Substituting in a state model M2) = AM Zit) + BU 8) Y(t) = CM Z(t) + DUH ww (4) +. Premultiplying equation (3) by M7! 2) = MAM Z(t)+ M71 B UG) where M?!M = I ic. Identity matrix 2) = AZH+ BUH w= (3) where A = M1tAM B= MB and Yu) = 62) +D UH (6) Where ¢= cM Equations (5) and (6) forms a new state model of the system. This shows that state model is not a unique property. Key Point: Ary finear combinations of the original set of state variables results into a valid new set of state variables. 1.6 Linearization of State Equation Any gencral time invariant system is said to be in equilibrium, at a point (Xp, U,) when, * X = £ (Xp Uy) = The derivatives of all the state variables are zcro at a point of equilibrium. The system has a tendency to lie at the equilibrium point unless and untill it is forcefully disturbed. The state equation xt) = f (X, U) can be linearized for small deviations about an equilibrium point (Xg, Uy). This is possible by expanding the state equation using Taylor series and considering only first order terms, neglecting second and higher order terms.Modern Control Theory 1-13 State Variable Analysis & Design So expanding 1 state equation, : Baf,(X,U) 9 f,.0.U) X= fy Mo, Uy) + PAS KX LAG] Wj- Up) isp Or [X=Xg jel j |xeX9 UsUg U=Up Now A(Xq Up) = 0 at the equilibrium point. And let the variation about the operating point is, 2 . X, = Xp-Xq hence X =X} = Uj-Up a and ; Hence the k'* state equation can be linearized as, 2 RRO U) «BAL OX U ~ = xX, 1 x=} 2x4 +h OU; |xexy Uj Usuo UsUy In this equation X,, X, and u, are the vector matrices and the remaining terms are the matrices of order n x n and n x m respectively. Hence the linearized equation can be expressed in the vector matrix form as, af, OR, © Ox, af, where A= Ox OX on isn x n matrix af, af, af, Ox, Oxy Oy, of Of of, dU; aU; aU, at, af aty p= [U1 Ue 0m | ig nm x mt matrix of, fy fy ou, Uy WUModern Contro! Theory 1-14 State Variable Analysis & Design All the partial derivatives of the matrices A and B are to be obtained at an equilibrium state (Xp, Up). Such matrices A and B defined above interms of partial derivatives are called the Jacobian matrices. Review Questions 1. Explain the concept of stale. 2. Define and explain the following terms, a. State variables b. State vector €. Slate trajectory di, State &. State space 3. Explain adoantages of state variable method over conventional ane. 4. Write a short note on advantages and limitations of state variable approach. 5. Write a note on linearization of stale equations. 6. Prove the nonuniqueness of the state model. Qo0gState Space Representation 2.1 State Variable Representation using Physical Variables The state variables are minimum number of variables which are associated with all the initial conditions of the system. As their sequence is not important, the state model of the system is not unique. But for all the state models it is necessary that the number of state variables is equal and minimal. This number ‘n’ indicates the order of the system. For second order system minimum two state variables are necessary and so on. To obtain the state model for a given system, it is necessary to select the state variables. Many a times, the various physical quantities of system itself are selected as the state variables. For the electrical systems, the currents through various inductors and the voltage across the various capacitors are selected to be the state variables. Then by any method of network analysis, the equations must be written ihterms of the selected state variables, their derivatives and the inputs. The equations must be rearranged in the standard form so as to obtain the required state model. Key Point: It is important that the equation for differentiation of one state variable should not involve the differentiation of any other state variable, In the mechanical systems the displacements and velucities of energy storing elements such as spring and friction are selected as the state variables. In general, the physical variables associated with energy storing elements, which are responsible for initial conditions, are selected as the state variables of the given system. m= Example 2.1: Obtain the state model of the given electrical system. vi) ~ I. Volt) itt} . Fig. 2.4 (VTU : July/Aug.-2006) (2-1)Modern Control Theory 2-2 State Space Representation Solution : There are two energy storing clements L and C. So the two state variables are current through inductor i(t) and voltage across capacitor i.c. v,(t). X) = it) and X,(0) = v,(0) And U(t) = v(t) = Input variable Applying KVL to the loop, vi) = i R +L a0 + va(t) Arrange it for di(t)/dt, au = tyw-Fig-tv oer but a = Xt) ic. Xt) = = Fx,o-E xy} uw @ While vo(t) = Voltage across capacitor = z Jiat Sol 2 Lien par Kol -% Le Xt) = Lxyo os (2) ‘The equations (1) and (2) give required state equation. * R 1 1 ‘so = fF TE] POO] Tug x Tog} at 20} é 0 ie. Xt) = AX()+BUQ) While the output variable Y(t) = v,(t) = X(t) x Y(t) = (0 1] [xi] + Ui) ie. Y(t) = C X(t} and D = [0] This is the required state model. As n = 2, it is second order system. Note : The order of the state variables is not important. X;(t) can be v,(t) and X(t) can be i(t) duc to which state model matrices get changed. Hence state model is not the unique property of the system.Modern Control Theory 2-3 State Space Representation ‘im Example 2.2: Obtain the state model of the giver electrical network in the standard form, Fig. 2.2 Solution : U(h) = input = e,(t) Y(t) = output = e, (0) State variables : X(t) = i(t) Xp(t) =i, X3(0 = vel) Writing the equations : di,(t e(0 = Ly a8 + velt) d ie. qe xy ~ () ‘Then, vet) dint) R “Fr = rp vel - iat ‘ 1 R, Xa(t) = Ty Xa TE Mal 2) and che = i,(t)~ip(t) = Current through capacitor dv) 1, 1, a = eiO- Gin kat) = 2xyn-2x,0 ~ )Modern Control Theory 2-4 State Space Representation y o 0 -t 1 x — : R nm x,] | Xo} =fo - 2 X, [+] 0 | UC) ‘ BP lx 0 x 2 il 3 3 c € 0 ie. XW) = AX(t) + BUG) and eo(t) = ig) Ry Y(t) = Xp(t) Rp x, Yt) = [0 R, a] |X, X3 ie. Y(t) = CX(t) + DU(t) where D = 0. This is the required state model. 2.1.1 Advantages: The advantages of using available physical variables as the state variables are, 1. The physical variables which are selected as the state variables are the physical quantities and can be measured. 2. As state variables can be physically measured, the fecdback may consists the information about state variables in addition to the output variables. Thus design with state feedback is possible. 3. Once the state equations are solved and solution is obtained, directly the behaviour of variaus physical variables with time is available. But the important limitation of this method is that obtaining solution of such state equation with state variables as physical variables is very difficult and time consuming. 2.2 State Space Representation using Phase Variables Let us study how to obtain state space model using phase variables. The phase variables are those state variables which are obtained by assuming one of the system variable as a state variable and other state variables are the derivatives of the selected system variable. Most of the time, the system variable used is the output variable which is used to select the state variable. Such set of phase variables is easily obtained if the differential equation of the system is known or the system transfer function is available.Modem Control Theory 2-5 State Space Representation 2.2.1 State Model from Differential Equation Consider a linear continuous time system represented by n'” order differential equation as, Yay g¥™ tay g¥P ht - . () . - mal mm a, Yeayvtt = boU+b, Ut.tb, {U"! +b,,U dy") =n" derivative of Y(t). at" In the equation, Y"(t) = For time invariant system, the coefficients a,_j, a,_y -. ag, by by ... by are constants. For the system, Y(t) = Output variable U(t) = Input variable (0), 10), ....¥(0)°"! represent the initial conditions of the system. Consider the simple case of the system in which derivatives of the control force U(t) are absent. Thus ty = Uwe. =u™ <0 2YP ayy YT) ay ¥ + ag YO = BUY v= Q) ‘Choice of state variable is generally output variable Y(t) itself. And other variables are derivatives of the selected state variable Y(t). Xi) = YD xt) = YO =X X30) = Yu = Xo = Xt) Thus the various state equations are, XW) = x0 Xw = x0) Kel) = XA . XOModern Control Theory 2-6 State Space Representation Note that only n variables are to be defined to keep their number minimum. Thus * * X10) gives n'™ state variable X,(t). But to complete state model X,(t) is necessary. Important : x, is to be obtained by substituting the selected state variables in the original differential equation (2). We have Y(t) = Xj, Y(t) = Xz, Yen = Xp oe YON) = XA, YP) = HAD PX) # ag ep Ml + apg Xp a gl) + oes + ay NE + ag M(t) = by UD ce qt) == 09 Xp ay Xp ee Op 2 Xp ~My Xp + by UD ~@ ‘Hence all the equations now can be expressed in vector matrix form as, . Xi 1 0- 0 x Raff 2 8 Pm FY MO lu ke mag Ap Ane} Pal [bo ie. X = Axi) + BU Such set of state variables is called set of phase variables. The matrix A is called matrix in phase variable form and it has following features, 1) Upper off diagonal i.e. upper parallel row to the main principle diagonal contains all elements as 1. 2) All other clements except last row are zeros. 3) Last row consists of the negatives of the coefficients contained by the original differential equation. Such a form of matrix A is also called Bush form or Companion form. Hence the method is also called companion form realization. ‘The output equation is, YQ) = X(t) x0 X4(b) Ye) = [1 O.o}] : Xn). ie. ¥() = CX() where D = 0Modern Control Theory 2-7 State Space Representation This model in the Bush form can be shown in the state diagram as in the Fig. 23. ‘Output of each integrator is a state variable. "3, un If the differential equation consists of the derivatives of the input control force U(t) then this method is not useful. In such a case, the state model is to be obtained from the transfer function. im Example 2.3: Construct the state model using phase variables if the system is described by the differential equation, 2. 2 d Yen | afl ayit) 7a) + 2Y(t) = SLU) ae det Draw the state diagram. Solution : Choose output Y(t) as the state variable X,(l) and successive derivatives of it give us remaining state variables. As order of the equation is 3, only 3 state variables are allowed. x) = YO xt) = X40 = = S0 and ‘yin = Sty = Yen = LO Thus RM = X21) = )Modern Control Theory 2-8 State Space Representation XA) = X3(t) + Q) To obtain Xi. substitute state variables obtained in the differential cquation. GY od ae YO= "ax, [yO] s 32 = X50) ce X(t) + AXG( + 7 Xt) + 2X,(0) = SUC) RW) = = 2X — 7 YC) — 4X) + 5 UC 8) The equations (1), (2) and (3) give us required state equation. XW) = AX() + BUCH) . 01 0 0 where A=1]0 0 1] and B=/0 27 4 5 The output is, Y(t) = X(t) ¥() = CX() + DU(Y where Cc = [100,D=0 This is the required state model using phase variables. The state diagram is shown in the Fig. 2.4. Fig. 2.4 2.2.2 State Model from Transfer Function Consider a system characterized by the differential equation containing derivatives of the input variable U(t) as, Yt agg Yb + + ay V+ ag Y(t) = bg +b, U+ + by UT +b U™ we ()Mader Control! Theory 2-9 State Space Representation In such a case, it is advantageous to obtain the transfer function, assuming zero initial conditions. Taking Laplace transform of both sides of equation (1) and neglecting initial conditions we get, Ys) Is" + a, ys" +. + ay 8 + ag) = [by + sb +. Hb 7 87 + by, Ss] Ul) Ys) _ 4) = Pa tsb $Dyy..8 1 +b, 8™ n=l gt + (2) Ay $9 yt tay 7S Practically in most of the contre! systems m Uw Lns uo) Suepsor 40) wesBerp ayes 117 “Bia Squsuaja JoURSIP BujuEWoYModern Control Theory 2-25 State Space Representation For series integrators, the state equations are, = -ajX, + Xy x * Xp = = apX, + Xs Near = aX iat X a " - aX, + Utt) While for parallel integrators, the state equations are, Sear = Xa + UW) * 0 = ay Xq + UU) While Yt) Xp + egXn te ROX Hae Xe a + + Ca Xq Key Point: Y(t) has additional cg L(t) term if m =n. Hence the state model has matrices in the form, Element way forrtimes f for r—1 times “At Oe 8 | Ordo N ~a, : sesseeeateeecs 0 Jordan block ‘only diagonal elements Fig. 2.18Modern Control! Theory 2-26 State Space Representation 0 of r-1 : | times B=|? CH [eye ay ve 1 1 1 The matrix D is zero if m Example 2.10: Obtain the state model of the given nekwork in the standard form. Assume R,=12 C,=1F Ry = 292 C,=1F Ry = 32Madern Control! Theary 2-32 State Space Representation Ry +e I Ry Input Uft ; Re mt 4 Cop Y(t} Output -¢- La} Fig. 2.23 Solution : Selecting state variables as voltages across capacitors C, and, ive. ¢, and ¢, ey) = Xt ey = X(t) Applying Kirchhoff's laws, Ult) =i, Ry - ey Ry iy - i) = 0 Fig. 2.23 (a) Uh = i, Ry +e, +R, (i, - i) oO) for second loop, ey - ig Ry -e2 “ig =i) R, = 0 ey = igRy Hey (ig —i,) Ry ~~ Q) Solving simultancously equations (1) and (2) ut) = i, (R, +Ryy-i, Rg +e, e, = -i, Ry +i, (RL +R,)+e, Substituting the values, Ut) = i, (1+2)-2i, +e, ie. Ut) = 34, -2i, +e a 8) e, = -2i, +51, +e, a AY Multiply equation (3) by 2. and equation (4) by 3 2 Ut) = ij - iy +20, 3e, = + 15i, + 3e, and adding BUlt) + Be, = Ii, + 3ey + 2c,Modern Control Theory 2-33 State Space Representation . 1 3 in UN + Fei Fe » B) Now from equation (3) UW) = 3i, -2i, +e, Substituting i, from equation (5) h , 6 2 Bip = UN+ Ut ere ay ee 6 2 6 Ul = 3 - UW - Fert qete 17 9 6 ly = 4g UD 35 er - aye + (6) dey . Now ot = i Capacitor current is C [dve/dt] as C, = 1, de, 17 9 6 3 1 3 ee UO mage eM taye de 8 12 3 1. 3 -—=— = ar * ag UD age + gg ee ‘ 8 yy _ 12 1 Xi # ag UO x) 7 X w= 7) de, . and CGP = ig ...Capacitor current is C [dve/dt] as c=1 dey 3 1 3 ae TE UO TTS Gee 3 1 3 Xo = gq, UW + ap X17 Xe ~» (8) and Yit) = ey = Xp “. State model is,X = AX + BU and Y = CX + DU BL & 3 Ti] p.|3 = = yo {Bef S| C=al D= fo] TN i. i where A=Modern Control Theory 2-34 State Space Representation hm Example 2.11: Consider the mechanical system shown in figure. For shown displacements and velocities obtain the state model in standard form. Assume velocity of M2 as output. Fig. 2.24 Solution : Select the state variables as energy storing elements ie. displacements and velocities related to spring and friction. X(t) = ¥,(t) X34 = VQ) XW = Yo), X= V0 Ut) = F(t), Y(t) = V2 (0) Draw the equivalent mechanical system. Due to F,(t), M, will displace by Y,. Due to spring K, and friction B, which are between the two masses, the displacement change from Y, to Y;. While mass M3, spring K and friction By are under the influence of Y alone as K, and B, are with reference to fixed support and not between two moving points. Represent each displacement by separate node, Connect the elements in parallel 0 Fig. 2.25Modern Control Theory 2-35 State Space Representation which are under the influence of same displacement, thus’K,, B, parallel between Y and Yg, My, Ky and By parallel between Y; and reference i.e. fixed support and so on. The spring force is proportional to net displacement in spring while frictional force is proportional to the velocitics, At node Y,, ay, B, d(Y, ~ 5) FL = M, art K,(%-¥)+# « () At node Yq, d? y. dy. B,d(Y, -Y,) 0 = M, ae +K,Y) +B, E+ K, 0%) -¥) + w 2) ‘Substituting all values in terms of state variables we get, . a?y, . Y= Xp. a 7 Xt=V =X5, aD = Xs adYy 8 dy, s Yy = Xz, qe TX eV, 2X, re = X4 -. Substituting in equation (1) and equation (2) Ul) = My X3+K, (X, -Xg]+ By [X5 -Xy] = Q) 0 = M,X4+K,X, +B,Xq+K, IX, -X]+B,1Xy-X3] @ From equation (3) and equation (4) we can write * 1a, X3 = wy, [UW -K, (X, - X3)- By(Xy - X4)] ~ @) ‘ 1 : ta and Xa= My [-K, X, -B, Xy- Ky % -X)-B, X,-X5)} + ©) and Xi = Xz, 2 =X. Y= Vy = XO +. State model can be constructed in the standard form AX + BU CX + DU YState Space Representation Modern Control Theory 2-36 a Qa 1 0 it} QO it 1 = | K, By Where As ~My +My -M, Ky 24K) By Ma M, M; 0 O B= lm, c=[00 0 1] 0 i Example 2.12 : For the given TF. of a system obtain the state model by i) Direct decomposition ii) Gullemin's Form iii) Foster's Form (s+2)(s+3) T(s) = 5(s+1)(s* +95 +20) Solution : i) Direct decomposition 1) = s?+5s+6 _ s(s° +10s? +2984 20) {{[s+ 10]s+ 29) s+ 20 }s s2 +5546 State diagram is as follows Us) X, =X, X= Xp, Xp =Xy, X, = U-20X,-29X,-10X,, Y = 6X, +5X, 4X,Modern Control Theory 2-37 ‘State Space Representation ~.State model is X = AX + BU and ¥ = CX + DU o 1 0 oO 0 o o 1 90 0 where A=], 9 9 ifBelo 0 -20 -29 -10 1 C=[651 0) D=[0] fi) Gullemin's Form Ts) = (s+ 2}(s+ 3) _ +2) (+3) 1 1 S) = Sist+ipis+ dts) (41) (+4) 45) Ss State diagram is, . . * xX, = X X= -5X,+3X,+ X; . . + X, = -4X,+2X,+X,, X, = Uls)-X, Substituting X, back in X, Xy = —4X, +2X, + Uls)-X, =— 4X, +X, + Uls) Substituting X, back in X, X, = - SX) + 3X, -4X5 +X, +Uls) = -5X,-Xy +X, + Us) and Y¥ 0 %Modern Control Theory 2-38 State Space Representation -. State model becomes X = AX+BU and Y = CX Oo 1 o 60 fo o 5 -1 1 I where A= 0 0-41 Bei c=[1000} oo 0 al uj iii) For Foster's form : Find out partial fraction expansion of T(s) Ts) = ‘ut X, = Ul)-X,, y=-1x+bx,-dxy+hx, X, = Ul) - 4x, X, = U)-5X, X, = UbModern Control Theory 2-39 ‘State Space Representation State model is, X = AX+BU and Y= -1 oo 1 o -4 0 0 1 303 where A= lo co 5 0) BR -% il 0 0 0 0] 1 ma Example 2.13 Derive the state model in Jordan's canonical form for a system having T-F. 1 Ts) = ——-—- tas? $5s42 Solution: qy.) - _ i 1 --A_,_B ® so 44s7 45s (s+1P (s+2) (s+)? S+D (+2) ie. A(s + 2) + Bs +1) (8 +.2)+Gs+1)? =1 As + 2A + Bs? + 3Bs+2B+ Cs? +2Cs+C = 1 B+C = 0,A+3B+2C=0, 2A + 2B+C 1 W Now C=1,B=-1,A=1 1 1 1 Tis) = Gen: “Be * GD State diagram is , un Fig. 2.29Modern Control Theory 2-40 State Space Representation : X, = =X4X, X, = UW-X, . X3 = Ult)- 2%, Ys) = X,-X, +X, «State model. is, “1100 0] where A=]0 -1 Qf, Bell, 0 0 -2 L c=f1-1 4 i= Example 2.14 : Obtain the state model of system whose T.F. is 3 2 Tis) = — S38 #5 _ iy Foster's form. s? +12s* +47s +60 3 2 . Solution : Ts) = 23s 3? 412s? + 475460 As numerator and denominator are of same order we cannot directly find out partial fractions. For partial fraction, numerator degree must be less than denominator. So directly divide N(s) by D(s) and find partial fractions of the remainder. g3 412s? +475+60) s3+3s7+2s (1 33+ 12s? + 475+ 60 — 957 ~ 455 - 60 Y(s) 9s? + 45s + 60 Ts) = oe =1-| ot 8 = OB [ne a =1 9s? + 453+ 60 =1_4 B Cc ~|6¥3)GFHEF5) | SSS 7 sea sss 3, 24 30 s+3 s+4 s+5Modern Control Theory 2-41 State Space Representation: ‘State diagram : Fig, 2.30 Ul) - 3X, . Xp = Ul) —4X, x u » j = Ut) - 5X5 Vis) = - 3X, +24X, ~ 30X, + Uls) -. State model is, X= AX+BU and Y=CX+DU -3 0 0 1 where A= o-4 0 B=|1 0 0-5 1 C= [-3 2-30, D=fl] When N(s) and D(s) are having same order, or degree of N(s) > D(s) then, there is always direct transmission matrix D present in the model. a> Example 2.15 : Obtain a state space model of the system with transfer function ¥(s) 6 US) ~ 63 46s? +118 46 (VTUJan/Feb.-2007,July/Aug.-2007,Dec/Jan-2008)Modern Control Theory 2-42 State Space Representation Solution : Y¥(s) 6 The TFis —— = ———————— Us) o3 +65? 411846 Using factorisation of denominator, Ys) 6 Us) ~ (s+1) (+2) (+3) Taking partial fractions, YO) 1 3 6, 3 Us) ~ stl s#2> 543 Hence the state diagram is as shown in the Fig, 2.31 Uts) Fig. 2.34 This is Foster's form of representation, From above Fig, 2.31 we get, X, = Uls)-X, X= Ul) -2X, X= Uls)-3X, and Ys) = 3X, - 6Xq + 3X5 Hence the state space model is, . X = AX + BUModern Control Theory and yY=cx -1 0 @ where A=] 0-2 0 Oo 0 -3 1 Be=/1 C= [3 -6 3] i? Example 2.16 : dty dy dy M2 4 6 Sf 4 11 4+ 10y = 3U ap a + la + 10y = 3U(t) Solution : System is 3rd order, n = 3 State Space Representation Obtain the state model for system represented by (VTU : May-99, Dee.2007/Jan.-2008) 3 integrators and variables are required. Select y =X, and then successive differentiation of y as next variable. X, = Xy = dy/dt ° dy xX, =X, = ae - (1) Q) Now as 3 variables are defined, X, 4X, but X, must be obtained by substituting all selected variables in original differential equation. X3 + 6X, +11X, +10X, X3 y -. State mode] can be written as, x = AX + BU and y = CX+DU ao 1 6 Where A= oo 1 vy =3U as X,-"4 dt? = 3U-10X,-11X, - 6X, 3) =X which is output equation ,C={10 0}, D= {0}Medern Control Theory 2-44 State Space Representation | State diagram : Ys) Fig. 2.32 map Example 2.17: Obtain the state model of the differential equation Se 2a He sd el acc) = Setty ae dp dt (VTU = Aug-95) Solution : The equation can be written as, aCe deMrey+2ec) = Slt) The order of the equation is 3. Selecting state variables as, et) =X) X21) x0 =a) well) X(t) x0 = eit) 2) oy Xy(t) = c(t) Substituting in the original equation, 4X, (0+ 3X, (+X, (04 2X, (0 = Sxl) . 5 Xf) © 2X 0-4 X,-3X, 040.) Hence the state model is, x o 1 of [xo 0 Xz}=] oo 0 1) |x m+] o| . 201 3 5 x: 2 1 _3) | x,@ 5 ° a9 9 0 qModern Control Theory 2-45 State Space Representation While the output equation is, et) = Xu) x, ie. yt) = [100]] X,« xX, ‘om Example 2.18: Obtain the state model of the system whose closed loop transfer function is Cis) 2(s+3) Re ~ Gerd (VTU : Feb-97, Oct-98) Solution ; Let us use the parallel programming i.e. Foster's form. Finding partial fractions of Cis)/R(s). co) 4 Ris) s+ Hence the state diagram is as shown in the Fig. 2.33 Ris) Fig. 2.33 Hence the state equations from the Fig. 2.33 are X, = Ri) - xX} el) X = RG) - 2% wQ) and Cis) = 4X, -2X, Hence the state model is,Modern Control Theory c(t) and yp Example 2.19 : Choose i,, iy and v¢ ait) GS) Solution: Let e(t) Voltage across Ry i,(t) i,() vel) 2-46 State Space Representation walls] Write @ set of state equations for the netwcark shown in the Fig. 2.34 ~ as state variables. (VTU : April-98,July/Aug.-2007) Ry Ly ly TOOT H0T—-------5 | -- R, eit) Fig. 2.34 = Input = ey (i) = output = X(t) = X30 = X;(t) Let i,(t) and i,(t) are the loop currents. Applying KVL to the loops. e(t) and = Ryiy + ty Stave = Ry -v¢ +e()] = cee X,-X, +e(0] we (1) = Ly ou +iy Ry = iy Ry ve] = + Q) .- Capacitor current u knit)State Space Representation Modern Control Theory 2-47 . Xs = @M%-%) (3) While eg(t) = i,R,=R,X, «(4 Hence the state model is, x x 0 [ 1 L x R, X, uy z=]; 0 -72 “| + | 0} Ult) where U(t) = eft) . ti Xs ° Xy Cc Cc x and YW) = cg =[0 R, Of |X, Xs mime Example 2.20: Considering v- and J; as state variables and J, as the output variables in the circuit shown below, obtain the state model. 1g (0) Fig. 2.35 (TU : Aug. - 97) Solution : Convert the current source to voltage source as shown. Fig. 2.35 (a)Modern Control Theory 2-48 State Space Representation Two inputs, e(t) and L.() ie. U, = e(t) and U; = Ice) One output, 1At) Le. Y(t) = Lt) Variables, ve = X(t) and 1, = X,(t) Let 1,(t) and I, (t) be the loop currents. Applying KVL to the two loops, Loop 1, -1, Ry - ve tet) = 0 I, Ry = eft) -ve 1 1 l= 7 e-R velt) Y(t) = ke U,- im X40 .. Output equation Ry -Ip Rp +ve =0 1 R,, Ry = pel * R. Xy = LXiO-52 Xt .. State equation and current through capacitor, dve I,-l = Cc a dv 1 —S = Sfl,- dt cli Substituting I, from output equation, ° 1 1 1 x, = ci, Ui ce, MX -» State equation * ec at 9 x | c eal? CR, fu] i, Ry [X00 0 & Uy X00 1 iY, v= fay olbeol Le ele:Modem Control Theory 2-49 State Space Representation ump Example 2.21: Derive two state models for the system described by the differential equation D3y +4D7y +5Dy +2y = 274+ 6Du+5u where D = dit i) One in phase variable form. ii) Other in Jordan-Canonical form. (WTU: March-2001) Solution : Take the Laplace transform of both sides of the equation and neglect the initial conditions to obtain transfer function of the system as, s3Y(s)+ 487 Y(s) +5s¥(s) +2Y(s) = 2s? U(s) + 6s Uls) +5U(s) Ye) 2s? +6545 Uls) 53 44s? 45842 i) Phase variable form Use the direct decomposition. Y(s) Is? +6845 UG) f(s ¥4)s+5]s+2) The state diagram can be shown as in the Fig. 2.36 So X, = % Xp = Xq X5 = U(s)-2X, -5X, -4X, and Vis) = 5X, -6X, +2X5Modern Contro! Theory 2-50 State Space Representation So state model is having matrices, o 1 0 oO A=|0 01 ale and C = [5 6 2) 2-54 i} ii) Jordan canonical form : Factarise the denominator as, Ys} _ 2s? +6845 2 A | BC UGS) (sty? +2) (s+1)? FN (+2) * A(S+2)+ B+) (8+2)+C(941)? = 2s? +6545 AS+2A4Bs? +3Bs+2B4+Cs? +2Cs+C = 2s? +6845 ee B+C=#2, A+3B+2C=6, 2A+2B+C #5 A=1, B=1, C#1 Ys) 1 1 Ue) ~ Gin? GER The state model is shown in the Fig.2.37 (a) (8) Fig. 2.37 * : U)=-X, —-X,=Xy~=X, Xz = Uls)-2Xy eo x and ¥(s) = X,+X2+X5 So state model is having matrices, -1 0 O 1 A=/1 -1 0] Be{o] andC=f14) 0 0 2 1Moder Control Theory 2-51 State Space Representation nm Example 2.22: For the system shown in the Fig. 2.38, obtain the state model choosing u,(t) and v(t) as the state variables, ({VTU: July/Aug.-2005, Jan/Feb.-2007) 1MQ voi) 1M vit) ‘ I T Ua) T uF T MF Y(t) Fig. 2.38 Solution : Select the two currents as shown in the Fig. 2.38(a). And write the equations using KVL and KCL. . ut) ima v0 ima ¥() + rn en + + 0) 7) = we) sh ur i, ip Fig. 2.38 (a) it) = LO -vett) .Q) ix 10° vel) = Bf, Ty) at dv y(t) ie. yr = CTR 2 Q) . Y2-¥1 ‘) = ves 0 = i & And vy(t) = alia dt dv, (t) it) = C +r we 4) Eliminate i, and i, from above equations and C = 1 x 10° °F,Modern Control Theory 2-52 State Space Representation Substituting (1) and (3) in (2) we get, dv [poe] [aoa dt | ayqo® | Lb e0® 110° 1x10° dv, - 4 Ult) | ¥,(t) aber dt Chixio®} Cixio® Clixi0® d aa = v4(t) - 2v9(t) + UW) 8) Using (3) in (4), dt 1x106 dv “i = ~ v4 (t) + volt) - 6 Select X= v(t) and X,(t) = volt) Using selected state variables, X= Xt % and Xq = Xp 2% + UW) and Y= y(t) = X(t Hence the state model is, XW) = AX(t) + BUC) and Yi) = CX(t) + DUG) fh aef? *) sePlcen a v=o where Fla Fl =[1 0), D= i> Example 2.23: Derive two state models for the system with transfer function, s vs) 245 © as) i) For first model, matrix A must be in companion form. ii) For second model, matrix A must be in diagonal form.Modern Control Theory 2-53 State Space Representation Solution : Arrange the transfer function as, 1 50x =x(s+5) 1000 (s +5) Ts) = = 36+ D6 +50) sx}x(s+ 2) xd x(5450) i) The companion form means phase variable form for which use direct decomposition of denominator, 1000(s+5) 1000s +5000 ts) = —/000ls+5) —_ _ 1000 +5000 __ ® fs? +52s4+100]s__ {l{s+52)s+100]s) The state diagram is as shown in the Fig. 2.39. From state diagram, and Xq = - 100X, ~ 52X; + Ult) While ¥(t) = 5000X, + 1000X, 0 1 0 0 A =1]0 0 1 B=|0 0 -100 -52 1 C = [5000 1000 0. D=0 ii) For diagonal form use partial fractions, A_B c TG) = St5ya* 350 A = T(s)3| = 1000x5 _ 59 s=0 3x50Modern Cantral Theory B c T(s) The state diagram is, UO. 2-54 State Space Representation 1000(3} = Ths)(s+2)|,_ oan =-3125 1000 (-45) = Tes)(s+50)], CeTae =~ 1875 50 31.25 18.75 $+50 s s+2 Fig. 2.39 (a) Thus the state equations are, % and Yt) Thus, A c = Ul, y= ~ 2X + UW, % = — 50% + UH) = SOX, - 31.25X, ~ 18.75X, 0 0 0 1 =]o 2 0, B=lt 0 oO -50 1 = (50 -31.25 -1875],D=0Modern Control Theory 2-55 State Space Representation m=» Example 2.24: Derive the state model for twa input two output systent shown in the Fig. 2.40. Fig. 2.40 Select output of simple lags as state variables. Solution = a. suggested, the output of at is X,, 1 sag 5 Xo 1, 5 Xy and — is X,. —p= is s+0.5 $42 The state diagrams for simple lags are, Fig. 2.40 (a)Modern Control Theory 2-56 State Space Representation eal) - @) Fig. 2.40 (c) X, = 1 -05X, wo) Fig. 2.40 (a) Ky = my 2% tt) From the given block diagram, ry = K,[U,-¥%) and ry = Ky [U, - ¥y] -»Modern Control Theory 2-57 State Space Representation and Y¥, = X, + 5X voi a) Yo = AX; + 4X, --(6 b) Substituting in (5) we get, ry = —K, X,- 5K, X, + Ky Uy a ry = — OAK, X3 - 4K, Xy + Ky U; (8) ‘Using (7) and (8) in the equations (1) to (4), X= (1-K,) Xy- 5K, Xp + Ky Uy wala) Xp = — 5Xy — OAK, Xg - AKy Xy + Ky Uy (9B) X= = Ky Xy 5K, Xp - 5X5 + Ky Uy 29 o} Xj = - OAK, Xy + © 4K) - 2) X; + Ky Up wD) The equations (9a) to (9d) and (6a) to (6b) give the required state model as, X = AX+BU and ¥ = CX + DU where (-1-K)) 5K, 0 0 K, 0 0 3 44K, ~IK, 0 K, Ae B= -K, -3K, -05 o |" K, 0 0 0 -04K, (4K, -2) 0 K, 15 0 0 c= 0 04 it D= 1 ‘im Example 2.25: A series RLC circuit with R= 10, L = 1H and C = 1 F is excited by v= 10 V. Write the state equations in the matrix form. [Bangalore Univ., Dec.-95] Solution : The circuit is shown in the Fig. 2.41. R L ie ] : ve ~ D covet ft) Fig. 2.44 Applying KVL to the circuit, d v= i) R+ Lae + ve(t) AI)Modern Control Theory 2-58 State Space Representation and volt) = él i(t) dt dv c(t) dt = & i(t) s+ Q) Let i(t) = X, = Current through inductor vc(t) = X= Voltage across capacitor V = Ue WV Substituting all the values, 10 = Xy+ Xp +X hy = -X-Xy +10 6) and X = X oe A) Hence state equations in matrix form are, xa] ft -17 PX] po [&| fi alle: Lo [Yo um Example 2.26: Obliain the state space representation in phase variable form for the system represented by, D*y + 20D%y + 45D y+ 18 Dy +100 y =10D7u45Du+ 100 u with y as output and u as input. (VTU: Jan.JFeb.-2005) Solution : Phase varaible form Taking Laplace transform of both sides and neglecting initial conditions, s*Y(s)+ 2057 (s)+ 45s*Y(s)+ 18s ¥(s)+ 100 ¥(s) = 10s7U(s) + 5s U(s)+ 100 U(s) Ys) 10s? + 58+ 100 Us) 54+ 2053 + 45s? + 185+ 100 Using direct decomposition, Ys) 10s? + 55+ 100 Us) {[{[s+ 20]s+ 45) s-+ 18]s-+ 100} The state diagram is as shown in Fig. 2.42Modern Control Theory 2-59 State Space Representation Fig. 2.42 The state equations are, x Xp =Xq, X35 =Xy Xz =-100X, -18X, -4X,- 20K, +U and y = 100X,+5X,+10X, 0 1 00 0 0 0 1 0 0 = = = 10 AT) go 7 [BR] yf C= L005 10 0] ~100 -18 -4 -—20 1 ‘> Example 2.27 : Obtain the state model of the system responsed by the following differential equation : vs 6 y+ 5yt+yau. {VTU: July/Aug.- 2005) Solution : Take Laplace transform of both sides neglecting initial conditions, s*¥(s)+ 6s7¥(s)+ 5sY(s)+ Y(s) = U(s) Me) Us) 59 + 6s? +5541 Ys) 1 Us) © {[(s+ 6)s+5]s+ 7} The state diagram is shown in the Fig. 2.43 Ys)Modern Control Theory 2-60 State Space Representation The state equations are, Xj =Xq,X)=Xq,Xy X,-5X,-6X, +U and yYexX ol a 0 A= |0 0 1|,B=|0]/,C=[100] -1-5 -6 1 imp Example 2.28: Obtain two different state models for a system represented by the following transfer function. Write suitable state diagram in each case. (VTU: July/Aug.-2005) MGs) _ __ 8s? +178+8 Us) (s+1)(s? +88+ 15) Solution : 1) Direct decomposition : Ys) | __ Bs? 417848 88? 417548 Us) 349s? 4238415 {[(s+9)s+23]s +15} The state diagram is shown in the Fig.2.44 Fig. 2.44 2X2, X2=X3, Xs =-15X,-23X, -9X5 +U Y = 8X, 417K, +8X, 0 1 0 0 Az|0 0 1),B=/ol,C=[817 8] 15-23 -9 1Modern Control Theory 2-61 State Space Representation 2) Foster's term : MS) __Ss?+17s48 2 A BO Us) (s+1)(s+3)(s+5) s+] s+3 0845 A=-0125, B= -7.25, C = 15.375 . Ms) _ 0.125 7.25 , 15.375 ” Us) stl s#3° 845 The state diagram is shown in the Fig. 2.45 The state equations are, * X, =-X,+U, Xq =-3X, +U, . X4 =-5X, +U Y = = O.125X, ~7.25Xq +15.375Xy -1 1 0 1 As|0 -3 G],B=/1|,C = [+0125 - 7.25 15.375] 0 0 -5 1 . Choosing appropriate physical variables as slate variables, obtain the im Example 2.29 : (VTU: JanJFeb.-2006) state model for the electric circuit shown in Fig. 2.46Modern Control Theory 2-62 State Space Representation Fig. 2.46 Solution : The various currents are shown in the Fig. 2.46{a). Fig. 2.46 (a) The volatage across resistance and second inductor is y(t). t ig = Wey At) di and ix 2 = y(t) seul) The voltage across capacitor is, v(t) = u(t) - y(t) cite _ dve(t) tel) = Ca = ae 3) diy ~ y(t) = and 1 EL = uty y= ve © won) di From (4), yi) = uit) ou = uft) — velt) Using in (2), = ut) - velt) sous}Modem Control Theory 2-63 State Space Representation Applying KCL at nede A, ict ity = iat ip dve(t) dota = te + 0 el ut) vet =i sis (6) Select Xi =i, ,, X2=4), X3=Ve =X from. (4) -X3+U -from (5) = -X)+X,-X,+U --from (6) and 2 =-X3+U ~-from (2) 1 0 -1|,B=j1|,C= [0 0 -1],D=[1] -1 1 ¥(s)_s(s #2)(s +3) jim Example 2.30: For the transfer function ——~ = ——-——~ RG) (s41)7(s+4) (TU: JansFeb-2006) Obtain the state model m i) Phase variable canonical form ii) Jordan Canonical form ‘Solution : i) Phase variable form ¥(s) _ s(s+2)(s+3) 83 +65 RS) (s#1)2(s+4) 89 +687 +9844 Using direct decomposition, Ys) 83 +55? +65 Ris) {[(s+6)s+9]s+4}Modern Control Theory 2-64 State Space Representation The state diagram is shown in the Fig. 2.47, Fig. 2.47 Xy 5+ 4X,-9X,-6X,4+R Y = 6X, +5X5 +X3 = 6X, #5X5-4X,-9X, -6Xy +R Y = -4X,-3X,-X +R fo 1 0 f Asio o 1|,B=lo|,c=f4-3 -1),D=1) baa -9 -6 1 ii) Jordan Canonical form : As the degree of denominator and numerater is same, first divide and then obtain partial fractions 53465749544) 5945s? 46s (1 s3+6s? 49844 -s?-3s -4 wo 2 (4) fae R(s) (s+1)7(s-+4) (s+1)? Stl s+4 A(s+4)4B(s+1)(5+4)+C(s+1)?_ = 5243544Modern Control Theory 2-65 State Space Representation Ast+4A+Bs? +5sB44B+Cs?+2Cs+C = 5743544 BHC = 1, A+5B+2C=3, 4A+4B+C=4 A = 0.666, B= O.111, C= + 0.888 Ms) = -| 0.666 ees RG) (sei)? (s+1) (s+4) Fig. 2.48 2X, s-X) +R, Xp s-X,+Xy, Xgs-dXy +R Y = -OINX, -0.666X, - 0.888X, +R -1 0 0 1 Az| 1-1 0},B=]0|,C =[-0111 -0.666 -0.888], D = [1] 0 0 -4 1 imp Example 2.31: Fig. 249 shows the block diagram of a specd control system with slate variable feedback. The drive motor is an armature controlled d.c. motor with- armature resistance Ry, armature inductance Ly, motor torque constant Ky, inertia referred to constant K, and tachometer K,. The applied armature voltage is contrailed by a three phase futl-converter. e, is control voltage, e, is armature valtage, e, is the reference valtage corresponding to the desired speed. Taking X= w(speed) and Xq = i, (armature currenit) as the state variables, u = e, as the input, and y = @ as the output, derive a state variable mode for the feedback system. (VTU: July/Aug,-2006)Modern Control Theory 2-66 State Space Representation Fig. 2.49 Solution : For armature controlled d.c. motor, v() = (1) and ep(t) = san) Tm = Kyi, soon) The torque produced is used to drive the load against inertia J and friction B. dw Tm = Jat Bo Aa) Kyi, = 122. Bo do a seen (5) The state variables are X; =wand X, =i, 2 Ky B X, = Xa -F va (6) From the equation (1), di vA) = 1, S2+i,R, + Kot) wn (7)Modern Control Theory 2-67 State Space Representation But the armature voltage v, is controlled by three phase full converter. Koig Kee = K.IK,x-Kyi, | K_Kix-K-Kyig Kyle, -K, K.Kji, oy = K.Kye, = KK) K, @- KK i, Fig. 2.50 Using this in the equation (7) above, K.Kye, -K.K,K,@-K Kyi, = La fe i,R, +Kyo (8) * KK, +K, Se (kK as td x, Ke (9) Ma And yuo =X The equations (6), (9) and (10) give us the required state model with, K 4 + lop A= ,B=|K.K,|,C=[i 0] ~(KKiK, +K,) A&G +R) a a imp Example 2.32 : For a transfer function given by G(s) = — 2 s?435+2 , write the state model in the i) Phase variable form ii) Diagonal form Solution : i) Phase variable form using direct decomposition 2 2 CG) = 243542 [(s+3)s+2]Modern Control Theory 2-68 State Space Representation Dep pe Fig. 2.54 Xp=Xq, Xp =-2X)-3Xp+U, Y=2X, f. Jja-fpe-m ii) Diagonal form using Foster's form cs) = —2 2. — 2 A,B s Saxan Ger 1333 2 2 A= aay 7? P= 2.2 CO) = Ser 542 X= -X,+U X= = 2X, 4U Y = 2X, +2X, oft gel oC =[2 2] Fig, 2.52Modern Control Theory 2-69 State Space Representation mm> Example 2.33: For the network shown in Fig. 2.53, choosing ij(t) = X,() and ix(t) =X,(t) as state variables, obtain the state equation and output equation in vector matrix form. 1H 12 itt igtt) Fig, 2.53 (VTU: Dec.-2007/Jan-2008) Solution : Applying KVL to the two loops, fh i ae ip +u(t) = 0 Loop 1 s yae. y ai2 A(iy-in)+ig +P = 0 suLoap 2 2 = iy-2ig (1) Using (1) in equation for loop 1, a = iy -(i, 21, )-i, +0lt) = -2i, +i, +0) wf) using i,(t) =X, and ij(t) = X2 Xp = — 2X, +X, +ult) 3) X= 4) and y(t) = on) Xi] xX, x and yt) = fl -1] | A= hr a}e-[i}-e-8 -1], D = 0]Modern Control Theory 2-70 State Space Representation Review Questions 1. Discuss the'deripation of the state model using following methads of programming. i) Bush form ii) Gullemin's form iti) Foster's form ip) Jordan's form 2. Flaborate upon the basis of selecting suitable state variables for a system. + 3. Obtein the state model for the block diagram shown. 5. Consider the permanent magnet moving coil instrument as shown, Here, e = Voltage to be measured, R = Armature Resistance, L = Armature Inductance, — Ky= Back emf constant relating i= Current through armature coil , J = MAL. of rotating part, K, = Spring constant , Ky = Torque constant relating i and T , Derive its state model. Ans.: XSAX+ BU Y=CX+ DU o MB) fare whereA=| 0 0 1 }Be/ 0 | C=[010) KK 9 0 J JModern Control Theory 2-71 State Space Representation 6. For a mechanical system shoten below obtain its state model in standard form. Assume output as YAU). An =AX+BD, Y=CX wy 0 1 i 0 wheeAs| KB Be, Leetom mom! - 7. Obtain the state inodel in standard Bush form of a system skown in figure. Ans.:X=AX+BU, Y=CX Ys} o 1 o 0 | o oa 1 0 0 where A= Ba| || C= (72 96 24 O1 o o oO 1 0 -288 -176 -98 -19 1 8. State whether the following statement is True or False with reasons : ‘The state space approach of the system analysis is a frequency domain analysis. 9. State whether the following statement is True or False with reason. The state space approach of the system analysis is a time domain analysis.Modern Control Theory 2-72 State Space Representation 10. For the mechanical system shown in figure input is u(t) and output is y(t). Obtain state equation and output equation. Represent the system by block diagram. Q00Matrix Algebra and Derivation of Transfer Function 3.1 Background The methods of obtaining the state model in various forms, from the transfer function of a system are discussed in the last chapter. It is alsa seen that the state models can be different and not a unique property for the given system. But most important point about the system is its transfer function is always unique. Though number of state models are derived for the system, the transfer function of the system, obtained from all of them is always unique for single input single output system. In this chapter the method of obtaining transfer function of the system from its state model is discussed. The concepts of diagonalisation of system matrix A, cigen values and eigen vectors are also included in this chapter. Number of methods from matrix algebra are used to understand the relation between state model and transfer function hence revision of matrix algebra is included in the beginning of this chapter. 3.2 Definition of Matrix A matrix is an ordered rectangular array of elements which may be real numbers, complex numbers, functions or mathematical operators, The order of matrix is always defined as m x n. where, Number of rows Number of columns For example, consider matrix A of order 2 x 4 of real numbers. 43 2 1)- afi 0 slo 2 Rows a 4 Columns (3-1)Matrix Algebra & Derivation Modern Control Theory 3-2 of Transfer Function Any clement of a matrix is denoted as ay where i indicates position of row and j indicates position of column. Thus for matrix A above aj; = 4, ay; = - 2, ay = 1 and so on. Key Point: Note that matrix does not have a value but its determinant has a value. A matrix with number of columns as 1 i.e. order m x 1 is called column matrix while a matrix with number of rows as 1 ic. order 1 « n is called row matrix. It is seen that vector matrices X, X, Y are the column matrices. 3.2.1 Types of Matrices The various types of matrices are, 1, Square Matrix : If number of rows (m) is equal to number of columns (n) of a matrix, it is called a square matrix, The system matrix A in a state model is always a square matrix of order n x n. 2, Diagonal Matrix : It is a square matrix with all aj = 0 for all i + j. Only main diagonal elements are present which are nonzero while all other elements are zero. For example, 160 A=j]0 2 0 o 0 3 3. Unit Matrix : This is a diagonal matrix, with all the main diagonal elements equal to unity. It is denoted as I and also called identity matrix, ‘The unit matrix of order 3 x 3 is, 100 I=]0 10 oo 4, Null Matrix : This is a matrix having all its elements as zero. It need not be a Square matrix. 5, Transpose of a Matrix : If the rows of columns of m * n order matrix are interchanged then resulting n = m order matrix is called transpose of the original matrix. If given matrix is A then its transpose is denoted as A’. 012 « A=lo5 a4], ate [b 9 2x3Matrix Algebra & Derivation Modern Control Theory 3-3 of Transfer Function The properties of transpose are, 1. (A+B) salty Bt 2. (AB) = BT AT 6. Symmetric Matrix : If transpose of a matrix is matrix itself, it is called symmetric matrix. It is a square matrix. ATS A 1-4 yf .l-d aol tele 7. Conjugate Matrix : The conjugate of a matrix is the matrix obtained in which cach element is the complex conjugate of the corresponding clement of the original matrix. It is denoted as A‘. _ fl +i2 14 j1 we{) 2 tit “ [2 -j4 34j2]" “12 #j4 3-j2 8. Singular Matrix : A square matrix having its determinant value zero is called singular matrix. a-|i 4 ad jap=|) Ala ao “los 2} “jos 2,007" ™ Thus A is singular matrix 9, Nonsingular Matrix : A square matrix whose determinant is nonzero is called nonsingular matrix. 42 42 As [5 ‘| and [A [5 g[-28- = 140 10. Skew Symmetric Matrix : A square matrix which is equal to its negative transpose is called a skew symmetric matrix. T AU = -A 3.2.2 Important Terminologies Let us study the important terms and their meanings related to the matrix. 1. Minor of an element ; Consider a square (n x n) matrix A and the element aj. If now i row and j'" column are deleted then the remaining (n - 1) rows and columns form a determinant M,. The value of this determinant is called the minor of an element aj.Matrix Algebra & Derivation Modern Control Theory 3-4 of Transfer Function Consider a square matrix A as, 23 4 A-=1{1 0 3.2 uw Let us obtain minor of element aj; = 1. Delete second row and first column to get determinant My). Ms = |2 4 a 42 5 Thus minor of 'ay;‘\is 7. 2. Cofactor of an element ; If Mj is the minor of an element aj then cofactor Cj is defined as, Thus for the matrix A considered above, the cofactor of a, is, Cy = CIP *'My =I *7=-7 3. Adjoint of a matrix : The adjoint matrix is the transpose of a cofactor matrix. The cofactor matrix is the matrix obtained by replacing each clement of a matrix by the respective cofactor. Adj A = [Cofactor matrix of AJ” ump Example 3.1 = Find the adjoint matrix of the matrix A, jaa te2[2 1 Coser, [76 Cy=C) leo Cyeent? 2-8 Cy = tt ; =+6 1 Cy = eret?lt 2-7 Gy = 177?! 9\ 2 \3.2Matrix Algebra & Derivation Modern Control Theory 3-5 of Transfer Function eer? 3 —c_ pelt 3 Gy = (i i|--% Cy = C1Y 2 7*5 10 eee = Cg =F iy 2 ‘| 4 6 -1 =-8 ©. Cofactor matrix = | 6 -7 -2 -12+5 4 6 -1 -8]/" [6 6 -12 “ Adj A = | 6 -7 -2] =j[-1 -7 5 “12 +5 4 -8 -2 «4 4, Rank of a Matrix : To find the rank of a matrix means to search for a highest order determinant from a given matrix which is having nonzero value. Thus if r x r determinant has a nonzero value in a given matrix then the rank of a matrix is 'r’ and any determinant having order r + 1 or more than that has a zero value. For example, consider matrix A as, 123 Azl24i1 tii ji 23 Now 24 1) = -520 1 Thus 3 x 3 determinant is nonzera hence rank of matrix A is 3. 12 ‘mp Example 3.2: Find rank of A= (3 ‘| Solution : For the matrix A, 12 36 =0 hence rank is not 2. Any other determinant of order 1 x 1 is nonzero. Hence tank of matrix A is 1. 5, Equality of Matrices : The two matrices A and B are said to be equal if they have same number of rows and columns and the elements of the corresponding orientations are equal.Matrix Algebra & Derivation Modern Control Theary 3-6 of Transfer Function 3.3 Elementary Matrix Operations The various elementary matrix operations are, 1. Addition : The two matrices A and B can be added to get C = A + B if both A and B are of same order and, C; + by for all i and j 2. Subtraction ; The two matrices A and B can be subtracted to get C = A - B if both A and B are of same order and, Cy = ay — by for all i and j Associative law holds good for both addition and subtraction of the matrices. [A+B] +C = A+[BtC] where all A, B and C matrices have same order. 3. Multiplication by scalar 'a’ : A matrix is multiplied by a scalar 'o’ if all the elements of a matrix are multiplied by that scalar ‘a’. Gay, Gay Gay, has aa oe Gas axA = | 7 aad 2a Aq, HAD KAny 4. Multiplication of Matrices : The multiplication of two matrices is possible if number of columns of the first matrix is equal to the number of rows of the second matrix. Such matrices are called conformal matrices. ‘Thus if A is of order m « n and B is of order n x p. Then the multiplication C = Ax B is of order m * p. The clements of matrix C are obtained as, n & ag by fori=1,2...nandj=1,2,...p Thus if A is of order 2 x 2 and B is of order 2 x 3 then C = A x B has a order 2x 3. A = [*h An B by by by a2, An ba by, bay C= AxB - esate arbi: +8 pobx, et 21 FA 7by) Agybyy +9 ygDg a gyby3 +A gpb95Modern Control Theory 4-3 Solution of State Equations ‘Using all these values in the assumed solution, 1 2,2 1 ku & X(t) = by + abylt) + ra? byl? +... + Fak byt by[teatedatiteat ahh] 2 kt " fired atte dat] x(0) 2 ki But leate tates. baka ot 2! ki X¢t) = o* x(0) wu (4) This is the required solution of homogeneous equation in scalar form. Thus if the homogencous state equation is considered, xt) = A X(t) then its solution can be written as, x(t) = e*' X(0) In this case, e“! is not a scalar but a matrix of order n x n as that of matrix A. Observation : It can be observed that without input, initial state X(0) drives the state X(t) at any time t. Thus there is transition of the initial state X(0) from initial time t = 0 to any time t through the matrix e“!, As it is an exponential term, the matrix e“' is called matrix exponential. It is also responsible for the transition of the state X(t) at any time t from initial time hence also called state transition matrix. It is denoted as 6 (t). o(t) = e“* = State transition matrix The each term of the above equation is a matrix of order n x n. If instead of initial time t = 0, it is selected as t = ty then the state transition matrix is, oy = eb fto)Modern Control Theory 4-4 Solution of State Equations 4.2.1 Zero Input Response The solution of the homogeneous state equation is under the condition of zero input. Such a response is called zero input response and for convenience denoted as ZIR. ‘Thus the behaviour of X(t) under the initial conditions without any input U(t) is called the zero input response of the system, It is also called free, natural or unforced response of the system. 4.3 Solution of Nonhomogeneous Equation Consider a nonhomogeneous state equation as, X(t) = AX()) +B UC) Xt) - AX) = BUH Premultiplying both sides by e” “', e “hx — A X(Q] = eo BUC) But oa) eA xy = S [eA XC] Substituting in above equation, Stem x01 = e“tB UD Assuming initial time as t = 0 and integrating both sides from t = 0 to t, \ oxi | = feo BUG) dx ° a t fen" B U@) dt 0 eo x0) - X(0) Premultiplying both sides by e“', t wet At X(t) - ot X(0) = et fe™*® BUC dt 0 t xt) = eM x00) + fe) B UW de @ 0 This is the complete solution of the nonhomogeneous equation.Modern Control Theory 4-5 Solution of State Equations Observations : 1. The solution is divided into two parts. The first part is et X(0) which is nothing ‘but the homogeneous solution or zero input response (ZIR). t 2. The other part fe*'-") B U(2) dt is the part existing only due to application of a input U(t) from time 0 to t. It is called forced solution or zero state response (SR). Thus the solution is, t xit) = et x(a)+fAt9 BU(t) dt o ZIR ZSR If the initial time is considered as t = tg rather than t = 0, the solution is, t xi) = AM x(ty)+ fest BU(adt ‘a In general, the solution of nonhomogencous equation consists of both the parts, ZIR (zero input response} and ZSR (zero state response). 4.4 Properties of State Transition Matrix The various useful properties of the state transition matrix are, o(t) = = State transition matrix 1. 90) = e**0 <1 = Identity matrix 2. ° w = ott = (ety? = fo (- or? ie HW = OH 3. O(t +h) = efit) 2 At Ata = Ot) Cb) = oC) Ott) 4 eAltts) 5 Ate As 5. eATB EC eAt eB only if AB = BA 6 few = fe eet = (ny 7 lta 4) Ot - te) = (bh - ty)Modern Control Theory 4-6 Solution of State Equations This property states that the process of transition of state can be divided into number of sequential transition. Thus ty to ly can be divided as ty to t, and t, to ty, as stated in the Property. Interms of $(t), the solution is expressed as, \ O(t ~ by) XUtg) + folt-9 BUD dr x(t) where — o(t-t) = eto) and o(t-) = Aten) 8. 0 (t) is a nonsingular matrix for all finite values of t. 4.5 Solution of State Equation by Laplace Transform Method The Laplace transform method converts integro differential equations to simple algebraic equations. Due this important property, it is very convenient to use Laplace transform method to obtain the solution of state equation. Consider the nonhomagencous slate equation as, XW) = AX() +B UW) ve A) Taking Laplace transform of both sides, s X(s) - X(0) A X(s) + B UG) $ X(s) = AX(s) = X(0) + BU(s) As s is operator, multiplying it by Identify matrix of order n * n, [sl - A] X(s) = X(0) + BUG) Premultiplying both sides by [sI - AJ"', vfst - AD? [sl - A] X(s) = [st - AP! XO) + BUG) x(s) = [st - AJ! x@) + [sl- ap} B UG) 2) ZIR + ZSR " Comparing zero input response obtained earlier, [t-Al? = 06) and + 3) Al (t The matrix 6 (s) = [sl - AT | is called resolvant matrix of A. All the clements of this matrix are rational functions of s.Modern Control Theory 4-7 Solution of State Equations Taking inverse Laplace transform of (2), x) = LU UX(sy) =U! fst - AP} xq) + U1 fst - AP BUG)) Using [sI-AT* = $6), X(t) = Lo (9) XO) +17? (6s) BUG) A The term L~' [6 (s) B U(s)] is called zero state response. From the convolution theorem in Laplace transform it is known that, 17! (y(6) a(S) = (0) * (0) = convolution t = JRt-OGdt oO Thus if F,(s) = $(s) and F,(s) = BU(s) then, t L716) B UGH = fout—9 BU(AQds 8) 0 Thus equation (5) shows that ul {6 (s} B Uls)} is the zero state response, as obtained earlier by classical approach. X(t) = L7? [9 (3)] X(0) + L” "fo (s) B UGS)] o(s) = [sl - Ay [sI-A| L-? fo) =o) =e Key Point : The advantage of this method is without carrying out actual integration which is time consuming, the zero state response can be easily obtained. 4.6 Computation of State Transition Matrix While obtaining the solution of state equation, the computation of state transition matrix e' plays an important role. There are various methods of obtaining state transition matrix from the state model. These methods are, 1. Laplace transform method 2. Power series method 3, Cayley Hamilton method 4. Similarity transformation method Let us discuss these methods of obtaining e“' in detail.Modern Control Theory 4-8 Solution of State Equations 4.7 Laplace Transform Method While studying the method of obtaining: the solution using Laplace transform method, it is seen that the Laplace transform of e“ i.e, 9 (t) and given by, Adj [sI-A] [s12A] 6(s) = (st- ar! Thus knowing A, once [sl - A] is obtained then (s) can be easily obtained. This ¢ (s) is again a matrix of order n x n ic. same'as that Of A. It is called resolvant matrix of A. The e*' ie. $(t) then can be obtained by: finding: inverse Laplace transform of 6 (s) ie. inverse Laplace transform of each clement of the resolvant matrix 6 (s). At This is most simple method of obtaining state transition matrix e"’ and hence popularly used. i> Example 4.1: Find the state transition matrix for, Solution : Use Laplace transform method, 10) fo ifs 4 (st A} = oF Ho ac wo : s+3 2] fs+3 -1 Adj{st-Al = > o| = 2 8 JsI-Al = s43s+2=(5+1)(6+2) 1 2 Ia-ap! = -i* 81 (s+1) (8+2) s+3 -1 1 __|StDG+2) G+l(+2) o(s) = [st- AP = > 2 s (s+1(s+2) (F1is+2)Modern Control Theory 4-9 Solution of State Equations s+3 =i wot eteaph ei? (+1)G+2) (FI)+2) e = isl — = 2 s NEFA GFF) Using partial fraction expansion for all the elements. 2210 A, stl s+2 sel s42 202 -1,2 stl s+2 S41 542 eter -2 At _ | 20% -c Zeb 20° eh 4 20° This is the required state transition matrix. 4.8 Power Series Method The state transition matrix is matrix exponential and hence can be expressed in a power series as, ete re ate tattanabattke. 2! kt .() Thus by calculating various powers of A and then few terms of power series, e“' can ‘be evaluated. As the power series is infinite, it is necessary to stop after calculating first few terms of the series. Calculating high powers of A is practically time consuming and hence this method is not practically used to obtain state transition matrix. ump Example 4.2: Find e of A = [: 3 ] using power series method, Solution : Let us obtain various powers of A. 2 . fo -1}fo -1] f-2 3 Avs = 2 -3)[2 -3. 6 7 ae f° 7 3776 7 2 -3)|-6 7 14 -14Modern Control Theory 4-10 Solution of State Equations s At a Tr ates ates aes 1 0) fo al 2 31, 1f6 7] 3 = ( +f a oot ttl al + rt? 4194... ate1s 21a... = 6 2t-3t Be. 1-3+3.5t? -2 +. Consider the first element of et which is 1-7 +4... -t_ Le yza dy Now e' = 1+[- H+ Shh eri tae : 2et = 2-24 P24. -» (1) ait and . J = 14C2)4 aia +t (- 28 +... 1-2428-8P4.. -. Q) From (1) and (2) we can write, dete 2 140-4 e.. Paha. Hence the series present as the first element is 267! - e774 But practically such a task is difficult to identify the combinations of series. Hence this method is rarely used. 4.9 Cayley Hamilton Method This method is based on the Cayley Hamilton theorem. The theorem states that, every square matrix A satisfies its own characteristics equation. Let f(4) be the characteristic equation which is given by, iQ) = |AI-A] =0 W ie. fQ) = Peay Pls ea para, =0 . () According to Cayley Hamilton theorem, the matrix A has to satisfy the characteristic equation. Hence, HA) = AT 4a, A™ 14 +a, pA+a,I=0 @Modern Control Theory 4-1 Solution of State Equations where I = Identity matrix The theorem is used for evaluating the function of a matrix. Consider a matrix polynomial as, P(A) = agh + aA + aA? +... + aA" +aq,, A" * +... eal) This polynomial has degree more than the degree of A which is n. The above polynomial can be calculated considering the scalar polynomial, PO) = apt ata + tay eagy ete... + Divide this by the characteristic polynomial f(A), PA _ RO 5 fay” 28 * FG “®) where RQ) = Remainder polynomial pA) = £2) QM + RQ «+ (6) Now the remainder polynomial R(A) has order less than the characteristic polynomial and can be expressed as, RO = cy + aAt +... 4+0,.,87! wo Let Ay, ag, -.. a, are the n distinct eigen values of matrix A and f(A) = 0 fori = 1, 2, ... Tas Ay, Ag, -.. Xy are the roots of f(A) = 0. Evaluate p(A) at all the eigen values, pA) = £0) Q0) + RO) fori=1,2,..0 But £0) = 0 hence, PO) = RG) = oy + yh +a Ay to oy ATT oy 8) ‘Substituting Ay, 2o, ... 4) in equation (8) we get the set of simultaneous equations which is to be solved for the values of dg 0%... Oy — Now replace }. in scalar polynomial by matrix A, P(A) = f(A) Q(A) + R(A) but f(A) = 0 «. By Cayley Hamilton theorem ptA) = R(A) = Gl + A+ a AP +... yy Ant v= (9) ‘Thus knowing Gp, @ ... 04, _ y any matrix polynomial p(A) can be evaluated.Modern Control Theory 4-12 Solution of State Equations 4.9.1 Procedure to Calculate eo“ The procedure to calculate e“* using Cayley Hamilton theorem can be generalised as, 1. Find the eigen values of matrix A from [41-A| = 0. 2. Form the remainder polynomial R()) of order (n - 1). RQ) = og + ay A+... + 0y_, A -) = pa 3. If all the cigen values are distinct then substitute Ay, 2, ... 4, in remainder polynomial R(4) to obtain simultaneous equations interms of Gg, 04 ... &_ 1. important Note : If the eigen value 4, is repeated r times, any one independent equation can be obtained by substituting 4 in R(A). Then remaining (r - 1) equations are to obtained by differentiating both sides of the equation, PQ) = RQ) Alp) = TRO co tr-1 ves (10) aR thea aN haa 4. Solve the simultaneous equations 0, Oy, ... % 4. 5, Then f(A) = R(A) = gf + aA +... #0, AN o 1 mm Example 4.3; Find f(A) = A” for A= [= 4 using Cayley Hamilton theorem. Solution : The function to be evaluated is, f(A) = A™ hence p(t) = 2? Step 1: Find eigen values. Rh -1 [AT~ Al 2 A+3 P4942 = 0 ie (+1) @+2)=0 hy = -Ldg=-2 Step 2: RA = oy + % A= pl)... asn = 2. Step 3: Substitute 4 and d in RQ). ty + Oy My = Oy)? and y+ Gy Ig = (2g) as p= A? ie. My -a = 1)? =1 )Modern Control Theory 4-13 Solution of State Equations and Gj - 20, = (- 2)" = 4036 (2) Step 4: Subtracting (2) from (1), * a, = —4085 and oy = — 4034 Step §: A) = A” = RA) 0 o1 = gl + ay - «34 t] =| | 4] _ [-4034 -4035 ~ [| 8070 8071 aia [7403-4035 ~ | 8070 8071 Similarly e“t which is a power series interms of A can be evaluated using Cayley Hamilton theorem. imp Example 4.4: Find the state transition matrix for A 0 -1 [: 31 Using Cayley Hamilton Theorem, Solution : The function to be evaluated is, A) = ec" hence p(t) = ett Step 1: Find the eigen values, Al [AT Al -2 243 Oie, =0 P43h42 = 0 ie (+1 A+2=0 a Step 2: Construct R(A) RQ = Og + Okt oe + Gy BO! but = 2 RQ) = 09 + a, 4= pQ .. By Cayley Hamilton method where pa) = e& ... replacing A by 4 in {(A). Og +h = etModern Control Theory 4-14 Solution of State Equations Step 3: Substituting values of 4, and Jy, Ug-u = 0! ofl) and Up - 2 = oo * woe (2) Step 4: Solving equations (1) and (2), mm = ee sand Gy = 2% - oo ™ Step 5: f(A) = R(A) = og + aA ... replace A by A in RQ) At © ot nfl 9 -t_ af -1 ete *]) Ole eM» 3 of Qe et et get 2e7t 20° oh 4 207 ° " This is the required state transition matrix. timp Example 4.5: Fitd the state transition matrix of, 00 -2 A=|0 1 0) by Cayley Hamilton theorem. 10 3 Solution : The function to be evaluated is, f(A) = e** hence pi) = e” Step 1: Find the eigen values. AO 2 Jat-aj =] 0 aa 0 [eo -1 0 4-3 » AA=-1I)A-3)+2A-)p=0 Q-1) 2? - 3.42] = 0ie @-N@-)M-H=0 y= Mad, Iga? Thus 44 = 1 is repeated 2 times Step 2: RQ) = oy + OA + ar? wn =3 = p@=e*Modem Control Theory 4-15 Solution of State Equations. Step 3: For =1, My + +0, = of « f) For Ag = 1, as it is repeated use, Spm! = Sra dak lhedg ih Iedy doa d 2 ie. a ‘lo = quit tee at - te larg = a, +20, Whar + 2a, = tel - 2) Note that differentiation is with respect to A and not &. and for g = 2, Cg t+ 2a +4o, = 2. (3) Step 4: Solve equations (1), (2) and (3). From (2), a, = te'-2a, From (1), Oy = eh a - a = el - tel + 2 - = -p+a Using in (3), fl - +a). +2te-4uytda=e% a = ct ch tet ate = e%- cl tet a = te 20 + 20! + 2te! = ate! + 2e' - 207 and og = ef tel + et — ef ~ tel = — atel + Step 5: f(A) = RA) = ogl + A + 0 A? 0 0 -2}fo 0 -2] [20 + A’ =/01 O}f0 1 o]=}01 0 to 3)[1 0 3 3.0 7 100 00 ~2 20 6 fA) = eteag|0 1 Ol+a,jo 1 Ol+a,)0 1 0 001 10 3 3.0 7 ‘Substituting values of og, 0 and cy and rearranging,Modern Control Theory 4-16 Solution of State Equations 2et—e* 2c" —207 oft es 0 et 0 wel te™ 267 -e' ‘This is the required stale transition matrix. 4.10 Similarity Transformation Method Consider a matrix A of order n * n having eigen values Ay, dy, ... dy. There exists a diagonal matrix A such that its diagonal elements are the eigen values of matrix A. A Os 0 0 dy 0 | | A ot = Diagonal matrix ~- () 0 Ow A, n Both A and A have same eigen values and hence are called similar matrices. ‘The diagonal matrix A can be obtained by defining a similarity transformation of state variables as, X(t) = M Z(t) In such a case, M is the modal matrix of A. And the diagonal matrix is obtained as, A = M7'AM 2) Now ec = lear aes.. 1 07-0) 4, 0 0-0 2 0 0-0 22 PO F0 2 OO) alo Boo mole, . 2 | oo—1f [0 — —a, o— —2 1 42,24 Tea tto yt 0 1 42,2 = 0 THhgt+ sagt Q —~-- — 0 | | | - ( i 12,2 0 0 —— Itdgtt rant +Modern Control Theory 4-17 The transformation used is, x(t) = M Z(t) X(0) = M Z(0) M~!X@) = M7'M Z(0) = Z(0) and XW = MZ Thus X(t) = AX(i) = AMZ{(t) Mit) = AMZ() Mo vz) = MAM Z(H 2) = AZ) This is transformed state model. The solution of homogeneous state equation in X(t) is, x() = e“* x(a) While solution of equation (5) is, Zz) = eM 20) Using equation (4) in (7), Zi) = e M7?) xq) Premultiplying by M, Mz) = Me M7! xc) But MZz(t) = X(t) x) = Me*tmM7? x@y Comparing equation (6) and (9) it can be written as, et = Me“) where M = Modal matrix and e“ is given by equation (3). Solution of State Equations « (3) . 6) +» (6) + (8) +> Transformation used 9) w+ (10)Modern Control Theory 4-418 Solution of State Equations mm Example 4.6: Find state transition matrix of, oO -1 A= [ 3] by similarity transformation method. Solution ; Find the eigen values. AOI AI-Al = | | -2 A+3 =H 4324220 (A+ 1) (2+ 2) mech yen Calculate eigen vector. -11 For 4 = -1, [Ay b= A] = (2 2] = eal El-f] “21 For i = -2, yt al=[% il le )-G] oi M (My Mal =|, 2| elt eto |_fet 0 0 et *['o et 2 -1 Mole aan [A i]. 2-1 — Mp At ay,-1_f1 fet o ]f2 eo MeM [i alt ety a »». Required state transition matrix o tt M,Modern Control Theory 4-19 Solution of State Equations im Example 4.7: Obtain the complete time response of system given by, i 072 7 X(t) = [s al X(t) where X(0) = [i] and Y(t) = [1 — 1] X(t) (VTU: Jan/Feb,-2008) Solution : Given system is homogeneous whose solution is X(t) = e** X(0) Now As ot fo-ay th I-A] = 10 01 _/s 71 Ust-Al = |g a] > [2 of*|2 5 (st-ayt = Adj(sl- A) |sI- A] Adj (sI - A) |st- Al (sl- Ay s 2 At w1 {8° +2 e =e L- . 2 ut | = +} = (actga| vit st + st + afa| lala. 2 |.41, . (Fal “E {e =m Boer uf = I = “fa xl 2 sin J2t 8° + (v2)? At cos V2 t pin . -VEsinv2t og yaModem Control Theory 4-20 Solution of State Equations 1. x = x@0) 2M fl cos /2 t+ Fe sin V2 t cos /2 t-/2 sin W2 t -. Output response Y(t) = [1 - 1] X(t) 1. h =u errant cos 2 t- V2 sin V2 t 0 Fein Bt 2 sinJ2t — is the required response. YO = ium Example 4.8 : Find out the time response for unit step input of a system given by * ol 0 o 1 1 iw -[o 5] xw [3] Ut) and vo -(5 sl X(t) and xo =[j]. Solution = sO] fo 1]_fs -1 {st- A] = [0 Ho als [2 s+ 3] ; — [Cy Cy] _fs+a -2]" fea 1 Adj [ sI- A] {er c. 4 sl 7|2 5 |sl-A| = s?+3s+2=(s +1) (6 +2) et = LT st-ay] s+3 1 - erner2) ernie) = (6) (s+1)(s+2) (8+1)(s+2) Finding partial fractions, es Lt At Lol | Stl s+2 s+1 s+2 ee LY =| 2 a 2Modern Control Theory 4-21 Solution of State Equations | Qe bag 2 eta t ~ 2°! 207 FF oF 4 207 nt n2t ZIR = e™ x(0) =e“ [| roene ~2e' +2072! To find ZSR = L! { o(s)BU(s) F Ut) = Unit step ..U(s) = 1/s 3 zsR = 1 eeDee2) anaes (| [V/s] (s+1)@+2) res rresiy Se) (283 88 _ pg [S++ 2)) yf 8 sel s+2 wy PE ss {s+ 1) (s+ 2) s+1 s+2 _ [25-578 #250771] So"! sg 2t | 25-3et 415077! X(t) = ZIR +ZSR = Fiee 3e7t ~3e77! oi Yi) = [. al x(t) YO] [0 1) [2s-set+1se"?*] [get — 307 ¥3(0] [3 | ge'-se"2t | | 54 607% - 307! Y(t) = 3(e"t -e"™) Y,() = -54+3(2e* —e) These are the outputs for unit step input applied. wm Example 4.9: For a system X= AX, X(t) -| . =2e xo=| ° -| H| -e -l Determine the system matrix A. 2 1 . | when xo =| 3] andModern Control Theory 4-22 Solution of State Equations Solution : System is homogeneous so ZSR = 0 x(t) = et x@) Ay | Let Aes Now X(t) = andatt=0 Xx) i So x0). = Similarly X(t) = =1 +1 So x0) = [:3] when x0={")] Substituting in X() = A X(t) [Ja RE) Cyt 2) A,-2A, = -2 ~ Q) Ay-2A, = +4 ~ @ +A ,-A, = -1 » @) A,-A, = 1 = (4) Solving these 4 equations simultaneously, Ay=1, A,=0, A,=-3, Ay=-2 As 01 [2] 4.11 Controllability and Observability In a control system analysis, it is necessary to find an optimal control solution for a given control problem. The existence of such a solution depends on the answers of two basic questions which are, 1. For a given system, is it possible to transfer any initial state to any other desired state in a finite time under the effect of suitable control input force ?Modern Control Theory 4-23 Solution of State Equations. 2. If the output is measured for finite time then with the knowledge of the input, is it possible to determine initial state of the system ? The answer to the first question gives the concept of the controllability while the answer to the second question gives the concept af observability of the system. 4.12 Controllability The answer to the first question means the concept of controllability of a system which is related to the transfer of any initial state of the system to any other desired state, in a finite length of time by application of proper inputs. Hence controllability can be defined as, A system is said to be completely state controllable if it is possible to transfer the system state from any initial state X(t,) to any other desired state X(t,) in a specified finite time interval (t,) by a control vector U(t). The concept of controllability and observability were originally introduced by Kalman hence Kalman's tests are used to find out whether the system is controllable and observable or not. 4.12.1 Kalman's Test for Controllability Consider n'* equation as, order multiple input linear time invariant system represented by its state X = AX(t) + BU(t) wf) where A has order nxn matrix and U(t) is mx1_ vector i.e. there are m inputs. X(t) is nx 1. state vector, The necessary and sufficient condition for the controllable is that the rank of the composite matrix Q, ‘The composite matrix Q, is given by, Q. = [B: AB: A7B:... A" ~ 1B] Q) In this composite matrix Q,, B, AB, A°B ... are the various columns. Proof : Let us see the proof for this condition. Assume that the final state is the origin of the state space while the initial time is zero ie. tg = 0. As final state t = t;, is the origin of the state space then X(t) = 0. Xtp = 0 +B)Modern Control Theory 4-24 The solution of the state equation (1) is given by, xt) = eM XO) + fe &) BUC de a Substituting t = t; for the final state, At ' Xi) = © EXO)+ { MW pugs 0 Using (3) in (5), t 0 = eix@e freee BU (dt 0 At van e §X(0) = -fe f eA® BULa)dt o ' f X(Q) = -fe*" BUG) dr 0 Solution of State Equations ~@ wl5) + (6) This shows that any initial state X(0) must satisfy the equation (6) for complete state controllability. Thus the controllability depends on matrices A and B and is independent of matrix C. Now mathematically e** can be expressed as, = okt 2 Fae ak k=O Using (7) in (6), aad, ft x0) = - Plaka faye U(t)dr k=0 0 t Let fay Udi = By 0 azl x0) = - SA*BB, k=0 6, X(0) = -[B:AB:...:A"~" B] Baa ata) - (B) + 9)Modern Control Theory 4-25 Solution of State Equations Be B, x@) = ~Q, v-- (10) Bua The vector matrix of By, is of order n x 1 and if system has to be state controllable then any X(0) must satisfy equation (9) for which rank of Q, must be n which is a nxn matrix. For any rank of Q, less than 'n’ it indicates that some of the states are not controllable. This proves that the rank of composite matrix Q, must be 'n' for complete state controllability of the system. The advantage of Kalman's test is that for any form of matrix A whether A is canonical or otherwise, the test can be applied. But its limitation is that it does not give a physical feel of the problem. lf the answer to the Kalmans test is uncontrollable then it does not indicate which mode is uncontrollable. But due to mathematical simplicity of its application, Kalian’s test is used to test the controllability. Key Point: The matrix has a rank r means the determinant of order r x r of the matrix has nonzero value and any determinant having order r + 1 or more than that has zero value. ‘Thus rank of Q. = n for controllab: tam Example 4.10 Find the controllability of the system, ef} n=2 1 61 0] Solution : A= [: 4] [1] 1 ae = [2] 01 Q. = [B AB] = fF 4] ol a Now ( [#7 Ewhies is nonzero Rank of Q, = 2 =n Hence the given system is completely controllable.Modern Control Theory 4-26 Solution of State Equations imp Example 4.41: Evaluate the controllability of the system witht, X = Axs BU 1 4 fi] A= B and [0 “1 io Solution : n=2 Q. = [BAB] Determinant of 2x2 = 0 Now | Hence rank of Q.r = 1 and Rank of Q, 4 0 ©. The system is not state controllable. 4.12.2 Condition for State Controllability in s-Plane The system is represented by its transfer function in the s-plane. Such a system is completely state controllable only if the denominator and numerator polynomials of the transfer function do not have a common factor, except the constant. Such polynomials are called coprime. In other words, there should not be a pole-zero cancellation in the s-domain transfer function of the system for complete state controllability. If the pole-zero cancellation exists then all the information about the dynamic behaviour of the system is not carried all along the system. In such a case, the system cannot be controlled in the direction of the cancelled mode. 4.12.3 Gilbert's Test for Controllability For the Gilbert's test it is necessary that the matrix A must be in canonical form. Hence the given state model is required to be transformed to the canonical form first, to apply the Gilbert's test. Consider single input linear time invariant system represented by, X@ = AX +B UQ)Modern Control Theory 4-27 Solution of State Equations where A is not in the canonical form. Then it can be transformed to the canonical form by the transformation, X(t) = M Z(t) where M = Modal matrix The transformed state model, as derived earlier, takes the form, 2) = AZ) + BUN) where A B=M-'B 4 Diagonal matrix It is assumed that all the eigen values of A are distinct. In such a case the necessary and sufficient condition for the complete state controllability is that the vector matrix B should not have any zero elements. If it has zero elements then the corresponding state variables are not controllable. If the eigen values are repeated then matrix A cannot be transformed to Jordan canonical form. If A has eigen values y, hq, to, gr dye Age Aye - 4, then the transformation results Jordan canonical form as, ree2eo Socaooc In such a case, the condition for the complete state controllability is that the elements of any row of B that corresponds to the last row of each Jordan block are not all zero. hump Example 4.12: Consider the system with state equation. Xi 0 1 Ojfx,] fo Xz] =] 0 0 14}x, |+jo} ut Xa ~6 -11 -6}/x,] [1 3 Estimate the state controllability by i) Kalman’s fest and ii) Gilbert's testModern Control Theory 4-28 Solution of State Equations Solution : i) Kalman's test Q. = IB: AB: A*B)...n=3 0 1 o}fo 0 AB=] 0 QO 1/)/O}=} 1 6 -i1 -6{{1] [6 0 1 «o}fo 1 0 0 0 1 A =o 0 if{0 0 i=/-6 -11 -6 6 -11 6] [4 -11 -6} [36 60 25 0 0 1] {o 1 A%B = | -11 -6| |0]=|-6 36 60 5} jl] [25 00 1 Q=|o 1 -6 1-6 3B oo 1 0 1 -6; = 1, Thus |Q,| is nonsingular. 14 3 Hence the rank of Q, is 3 which is 'n’. Thus the system is completely state controllable. ii) Gilbert's test For this, it is necessary to express A in the canonical form. Find eigen values of A. A -1 0 JAI-A| =|0 & -1)=<0 6 ll Ate 3B +6R411k+6 = 0 2+ 1) (h+ 2) (+3) =0 Ay = 1, We-2% Ige-3 As matrix A is in phase variable form, the modal matrix M is Vander Monde Matrix. 111 11 M = 4 4 hy =|-1 -2 -3 WwW 14 9Modern Control Theory 4-29 Solution of State Equations T 6 6 =2 6-5 =1 38 3 6 8 2 3 25 0s 50. - Adj[M]_|[-1 2-1 2-3-1 Mo! = MS EES ela aA [M| ~ ~ 115 05 3.25 05]f0] fos B= Mi B=]-3 - -1|lo/=|-1 115 o5ffa} [0.5 As none of the elements of B are zero, the system is completely state controllable. Key Point; As Gilbert's test requires to transform matrix A into canonical form, it is time consuming and hence Kalman’s test is popularly used to test controllability. 4.12.4 Output Controllability In the design of some practical systems, it is necessary to control the output variables rather than the state variables. In such a case complete output controllability is defined. Consider the state model of a linear time invariant system as, * X(t) Yo A X(t) + B Ut) Cc X(t) + D U(t) ‘The system is said to be completely output controllable if it is possible to construct an unconstrained input vector U(t) which will transfer any given initial output Y(t) to any final output Y(t) in a finite time interval ty st s ty. In such a case, construct the test matrix Q. as, Q, = (CB: CAB: CA’B:... CA"-! BD] ‘Thus if the order of matrix C is p x nie. there are p outputs then for the compleie output controllability, the rank of test matrix Q, must be p. 4.13 Observability The observability is related to the problem of determining the system ‘state by measuring the output for finite length of time. Hence observability can be defined as, A system is said to be completely observable, if every state X(t.) can be completely identified by measurements of the outputs Y(t) over a finite time interval. If the system is not completely observable means that few of its state variables are not practically measurable and are shielded from the observation. Similar to the controllability, the ebservability of the system can be obtained by using Kalman's test.Modern Control Theory 4-30 Solution of State Equations 4.13.1 Kalman's Test for Observability th Consider n'" order multiple input multiple output linear time invariant system, represented by its state equation as, X= AX(t) +B U(t} +) and ¥) = C X(t) +2) where Y() = px1 output vector and C = 1xn matrix The system is completely observable if and only if the rank of the composite matrix Q, is tn’. The composite matrix Q,, is given by, where c! = Transpose of matrix C and Al = Transpose of matrix A Thus if, rank of Q, =n, then system is completely observable. Proof: For the system described by the equation (1) and (2), the solution is, t xt) = eM! xO) + fe? BUD dr we (3) t Yi) = Ce’ x@sc ferro B U(x) dt ve (4) oO Now as the matrices A, B and C are known and the input vector U(t) is known, the integral term in the equation (4) is known. This can be subtracted from observed value of Y(f, to modify the lefthand side. Hence the obserability can be investigated by considering the equation, Yt) = Ce™ xy . wa 6) as the left hand side is known due to measured output. Thus for observability, the unforced system can be considered i.e. homogeneous state equation can be considered. ‘Consider the state model of unforced system as, X= AX) and Y()=C X(t) The output is given by, Yi) = Ce™ xo) wa (6)Modern Control Theory 4-31 nol But ets Yaak k=0 —1 vo = Faw akxoy k=0 Y(t) = colt) CX(0) + cy lt) CA X(0) +... Solution of State Equations .@ + Oy) CAT? X(O) . 8) For the system to be completely observable, from the given output Y(t) over the interval 0 Example 4.21: Obtain the tinte response of the following system : x x, x1] ilk ‘|-( ue ko 1oaj}x, | by where U(t) is the unit step occurring at t=0 and X7(0)=[1 0]. Solution : From the given model, 1 0} 1 1 sofia} eh] xb] Now as matrix A is same as considered in Ex. 4.20 while X (0) is also same as consider in Ex, 4.20, the zero input response is same as obtained in Ex. 4.20. e ZIR = Xit)= i ‘The zero state response which depends on U(t) and matrix B is given by, = X(}=Lt {a(s) B U(s) } L {oy}=L {e*T} = [st-ay G nl where ols) y | as obtained in Ex. 4.20. (s-1)? (s-1) 1 U(s) -] as unit step 7 I = os 5 a " —— a ote ~o =a 6(s) B UG)Modern Control Theory 4-42 Solution of State Equations | oa 1 y-| ea aja s |(s=1)? (s-1) 11 = ss va using partial fraction (s-1)? 1a Z5R = 1 fo) BuGy=Et|s] [3] -1)? Total response = X(t) =ZIR+ZSR et e-1 2et 1 = + = te} [tet 2te! tm Example 4.22 : Find and sketch the response of the system with the following transfer function, input and initial conditions. ~4s+ 0) = Sor rt) = 10eft)... (input) 0) = 0 Solution: r(t)=input and y(t) = output Dividing numerator by denominator as are of same order, YO) |g, 1220 = 44+ Ris) 3+ 300Modern Control Theory 4-43 Solution of State Equations The corresponding state diagram is, Fig. 4.1 Xi r(t)-300.X, and ylt) = 1220 X, —4 r(t) A=[-300], Be[1], C=[1220], D=[4] [sI-A] = [s+ 300] bear » [55] AT [e- wry Now y(0) = 0 hence substituting in y(t) yO) = 1220X,(0)-4 r(0), 1(0) =10 as given to be 10 u(t) 0 = 1220 X,()-40 40 x, = — 10 1220 ZR = AT Xe = 30 1220 To find ZSR_ use, ZSR = L [o(s)BU(s)] =u {—1_.,.0 =Risy=22 zh ee A U)=Rie) = ec! 10 er ete] s(s+300) s s+300 al ootModern Control Theory 4-44 Solution of State Equations 40 eA ett 1220 X(t) = x@ = 254x107 «amt 30 yi) = 1220 X()—4 (0 = 1220 [5 “54x 1074 7 ont -4 (10) 30 = 0,667 -0,667 e~ 9" The response y(t) is, 0.667; Fig. 4.2 tum Example 4.23 : A linear dynamic time invariant system is represented by X= AX +B UM) o 1 0 o1 where A= /0 0 1] B=j00 0 -2 -3 10 Find if the system is completely controllable. Solution : For the system, n= 3 Q, = [B: AB: A7B] Oo 1 O) FO 1 00 AB =]0 0 1] /;0 O]/=| 1 0 0 -2 -3] [1 0 3 0Madern Contra! Theory 4-45 Solution of State Equations : 0 1 O}f oo} fia A*B = A[AB]=|0 0 1) [1 Of=|-3 0 0 2 3) [3 of [7 0 01 00 10 Q,=/00 1030 1030 70 Consider the determinant, ao 1 0 0 0 1] = 1=nonzero 10-3 Hence rank of Q, = 3 =n Thus the given system is completely controllable we Example 4.24: Consider the systent represented by, = Oy ew =} oon 02," * ha Yu) = [1 0] Xp " Find the complete observability of the systent. Solution : For the system, n = 2 ~0.2 O04 A -{ | c=f1 0] 0.1 -0.1 -02 O1 . 1 TL To ae [ od -oa c | Q, = 1cT ATCT) 02 oa) pt TeT L ac = [ oa -aal [P| 1-02 Q. = [i oa u i eo imModern Control Theory 4-46 Solution of State Equations Consider the determinant, 1 -02 | | = 04 = Nonzero 0 0. Rank of Q, = 2=n Hence the system is completely observable. ‘ws Example 4.26 : Using similarity transformation find e" for, a-{4 | -6 5 Solution : Find the eigen values. JAI-A] = us 2 2-2-2 50 (+1) (4-2) = 0 dy = -1, =2 Find the eigen vectors. For4=-1, [yl-A] = (: | m= [es)-E)-Ea For ly = —2, [il - A] ae | ee [-)-E] j IM] _ [2-1 IME oy [ntModern Contro! Theory eM [* Q Solution of State Equations _ [= ett 20"! -2e"" +207 | imp Example 4.26: Using Laplace transform method, find & for fo -3 0 1 aA=l, “i wa-[ 5 «| 0 -3 Solution : 0 A =[' “I s 3 (st— A] = [A oa s+4 -3 std 3 1 Adj(sI-A]_ [1 8 los ist- AT = [SIA] sist 4y+3 (s#1)(5+3) see s+} (s+3) (+1) (s+3) ols) = 1 s (s+1)(s+3) (s+1)(s+3) s+4 -3 1{SFDs+3) (8+) (+3) —1__s«__ 4 is+1)(6+3) (+1)(s+3) 15°05 -15 15 + s+] s+3 s+1 s+3 05 05 05,15 | s#1 s+3 s#1 s+3. Set —0.5e5! -15e4 41.50% Set 050% -05e7' +150Modern Control Theory 4-48 Solution of State Equations b)A= ° } [3 <4 s -1 I~ A} = (5 wal s+4 1 ‘s+4 1 stay? = Adjisl-a]_[-3 s}|_[-3 5 IsI-A| s2 44543 +0543) s+4 1 (s+1)(s+3) (s+1)(s+3) o(3) = — 3 (s+1)(s+3) (s+1) (s+3) s+4 1 At rl (s+1)(s+3) (8+1) (s+3) et = 3 8 (s+1)(s+3) (s+1) (s+3) 15°05 O58 05 s41 s4+3 0 s+ 543 15 15 -05, 15 aS, 15 05 15 stl s+3 stl 533 eit | 15e'-05e" O5et-0.5e% 1.50% 415e 9.50 +1503 ium Example 4.27 : Using Cayley Hamilton method, find <* for, f(A) = eo“ hence pt) =e" Step 1: Find cigen values A A 6 AS JAr-Al = 24546 = 0 ie, (+2 Q+3)=0Modern Control Theory 4-49 Solution of State Equations Step 2: Step 3: and Step 4: Step 5: b) Step 1: Step 2: hy = -2 and Wy=-3 n=2 RQ) = 9 + a= pO) Og + yA = et Substitute values of 1 Gg - 20, = o Gy ~ 3a = Solving equations (1) and (2), oy = tie 8 gga ge B23 f(A) = R(A) = cl + 0 A eM. get —20%[) rete 1 be" 60-20 + Be [oa] f(A) = 8 hence pid) = e# Find eigen values. B+ 4ne4 = 0 (A+ 27-2 0 4 = -2 Wg=-2 n=2 RA = 09 + & 4= pi) Oy + GA = ri ate Je72t 9.73 ett ed —3t | 6 5. ) 2)Modern Control Theory 4-50 Solution of State Equations Step 3: Substitute values of 1 Oy - 20, = ot -@) But as 4 is repeated, second equation must be obtained as, dR) _ dp da di a = te! at h=-2 wn Q) a, = te Step 4: Og = 6 + 20, = + ate w+ from (1) Step 5: f(A) = R(A) = dol + aA et 5 7 +a) ee" [s ‘] 24 ee 2te® | -2te% ea-2n im Example 4.28 : Find the response of the system, © 0 1 21 a ef Yool? tum am =] and Y(t) = [ i X to the following input, un = [n (l | u(t) P ‘ ust) 2 . | ser u(t) = unit step function. Sit) + Find ett [9 1 _fs -1 Solution : Find e’ for an|S 3 thus [sI- A] =}, 4 s#3 1 s+3 1 ta-ap? = Adibi-al_ 1-2 s]_ 1-2 s} |sI-A]_ 5? +3542 (St1)(s+2) s+3 1 (s+1) (s+2) (s+1)(s+2) $(s) = [st-ap'= -2 s (s+1)(s+2) (8+1) ($+2)Modern Control Theory At Now U(s) ZSR 4-51 Solution of State Equations ee ee ee stl s4+2 s41 s42 42. ay s+] s+2 s41 s+2. 4 toa -e et ne "(= [zn “t42e2t et Sea | et x(Q) = ot [etl L! {0 (8) B Us) 1 5 8 s+3 1 (s+1)(8+3) (stl)(s+2)|fo 4 2 s [ {(s+1} (842) (s+1) (s+2) +3, |~ win 2. —st3_ a] {#1@+3) (4142) [E7553 —_ is - (s#1)(s+2) (s+1) (s+2) 43 843 1 3(s+2)]] yf {StH G+3) (S41) (s+2)| | s(s+3) 2 1 (s#1)G+2) (+1642) | 543 I Ton *GaGD6+3 (s+2)(s+ 3) 5 anaes" (s#1) (s+2)(s+ 3) 3.3 {05 1 «08 s#1 s+2 s+3 1 O5 2 15 $43 s+] s+2 5+3Modern Control Theory 4-52 Solution of State Equations -2+2.5e' +207 -2 os | X(t) | 3-2.50' -e 40.5 ~et 49. 5e! ] et +2e7 2.50% ed . [ Heo] Yo(t) 1 1X YO = XQ = 3-25 '-e *+05e% and Yott) ' X() + XQ) er + eB 2% ” =p Example 4.29 : The Fig. 4.3 shows the block diagram of a process control system with state variable feedback and feed forward control. The stale modet of process is, ° 3 2 t x-[ }ee(e and ¥=[0 1X a) Derive state model for the entire system. b) Find the response if input r(t) is unit step assuming X(0) = (al Solution : a) From the block diagram, Ul) = +7 x(t) - 3X, - 15x, .() From the given state model of the system, X= — 3X, + 2X, + Ult) .. Q) and Ny = AX; - 5X> 1. Q)Modern Control Theory 4-53 Solution of State Equations Using, (1) in (2), xy H ~ 3X; + 2X, +7 r(t) 3X, - 1.5 X, Ky = - 6X) 405%) +7 x10) ve A) The equation (3) remain unchanged. Hence state model of the entire system is, X= [7 viel] x(t) 4-5) The output equation remains unchanged. yw) = [0 1) xt) -6 0.5 Tus, A=[ f ‘s]: v-[): c= 4 b) Find 6(s) = [st- ar? st+6 -0.5 Ist A] = ( ws (‘<° oa (‘3° 057 tt ar! _ Adj[sl-A] | 4 s+6. _ | s+6j {sI-A] (s+6) (s+5)-2 (5° 4118428) s+5 0.5 4 st+6 © (s#4)(8+7) $45 0.5 (s+4)(8+7) (+4) (5+7) o(s) = 4 _ sto +4) 647) 644647) 4 0 No need to calculate et as X(0) = 4 the ZIR is null matrix i.e. zero. ZSR = L” {g(s) B U(s)] but here input is r(t) and not U(t) to the entire system. ZSR = Lois) B RG} and Ris) = 2 sModern Control Theory 4-54 Solution of State Equations wool, JEJ}- EYewsils 7 0 28+) neo] 1.25 05833 _ 0.666 s(s+4(s+7) eu? i= ct sed 847 | 1 233 , 1.383 He | s std 847 125-05833e" -0.667 67" | [X,(0 ZSR = = 1-233e4 +1.33307! x30) xy YQ) = [0 al ‘|. XA(0) = 1-233 6° "+1333 This is the required response. ium Example 4.30 : Evaluate controllability and obscroability of the following state models, 201 1 aa-[t a). oli]: c=fl -1] -1 0 0 10 b A=} oO -2 oO], B=}1 2], cf t o 0 3 21 oo 0 40 e)A=}]1 0 -3], B={10!, C=[0 0 1] oi 0 Solution : a) For controllability, Q, = [B : AB] as n = 2 -2 «17/1 2 weft alll-[1] 1-2 af a] {3 1 = 1 hence rank of Q.=2=nModern Control Theory 4-55 Solution of State Equations Hence system is completely controllable. For observability, Q, = (C7: ATCT] as n = 2 ooer] "=f all)G a [3] we aps : «. Hence the rank of Q, =1<1n atcT Thus system is not completely observable. b) For controllability, Q. = [B: AB: A*B] asin = 3. 1 0 off o} fa oo AB=|0 -2 Oj{1 2/=|2 -4 0 a -3{]2 1] [6 3 -1 0 Ojf-1 oO -10 A’B = A[AB]=| 0 -2 o|/-2 -4]=| 4 8 o 0 -3\|-4 -3 18 9 1o-1 0-0 Q=|1 224 48 21-4 -3 18 9 10-1 1 2 -2) = -7 #0 hence rank of Q.=3=n 21 -6 Thus the system is completely controllable. 2 For observability, Q, = (CT: ATCT: AT’ CT} asn=3 1 0 oO] fi 3] fa 3 aATcT =] 0 -2 of {1 1] =[-2 -2 o o -3| [2 5s} [-6 -15 -1 0 Oj f-1 -3 103 atcl = avjatclj=| 0 -2 0] |-1 -2/" X1 = 2Xq - 2X5 - X od) but X; = X; hence X1 = % and using in (4), Xy = 2Xy - 2% - X3 = 2X_~3Xy aed)Modern Control Theory 4-60 Solution of State Equations The equations (2), (3), (5) give us required state model. . 001 0 X(t) = [-2 3 0] XQ) + | 2] UGE) 02-3 0 and YW) = ft 0 OxW For conrollability, Q. = [B : AB: A? B] as n = 3] 0 0 1)fo] fo AB = |-2 -3 0|}2/=|-6 QO 2 =3)|0 4 oo 1 0 4 A’B = A[AB]=|-2 -3 0)|-6/=| 18 0 2-3], 4] [+24 ooo 4 Q = |2 -6 8 0 4 -24 0 0 4] 2 ~6 18| = +3220 hence rank of Q.=3=n O 4 -24 Thus system is completely controllable : 2 For observability, Q, = (CT: ATCT: AT C]asn =3 0-2 OVf1] fo atcl = 0 -3 2/}0/= 1 0 -3}[o} [a Oo 2 O;70 0 a™cl 2 aTiaTcl slo 3 2//o/=] 2 1 0 -3}}1 -3 10 0 Q,=|0 0 2 o1 -3 10 0 :|0 0 2|=-280 hence rank of Q,=3=n 01-3 Thus the system is completely observable.Modern Control Theory 4-61 Solution of State Equations vamp Example 4.33: Use controllability and observbility matrices to determine whether the system represented by the flow graph shown in Fig. 1 is completely controllable and completely observable. (VTU: JanfFeb.-2005) Fig. 4.8 Solution : The value of the variable at the node is an algebraic sum of all the signals entering at that node. X3 = X)+X,+2u, Kae X; -2X, +X5+2u X, = -3X, +X, Y =X, +2K,-X, 3 10 a A=/]1 -2 1/,Bs/2|,c=ft 2-1] 1 810 2 For controllability, Q. = [B: AB:A7B] 3 1 0] fo 2 AB = [1 -2 1] |2| =|-2 1 1 of [2 2 3 1 07]/3 10 1-5 1 APM 2/1 -2 1] |1 -2 1J=|/-4 6 -2 1 of{2 1 of [2 -1 1 8 10-5 «1)foy f- Ag = |-4 6 -2/|2/= 2Modern Control Theory 4-62 Solution of State Equations |Q. = -3220 hence rank of Q.=3=n The system is completely controllable. 0 10 “4 1 -2 4 Q=];2 -4 8 <1 2 <4 2-4 1Q,) = D and 1 = ots 252 determinant i eco Thus rank of Q,= 1 hence system is not observable. imp =Example 4.34: Given the time invariant system : 0 IBIS me of and that u(t) = & and y(t) = 2- te", find XW and X,(t). Find also X,(0) and X,(0). What happens ifa = 0? (VTU : JanJFeb.-2005) 0 «@ a Solution : a-( i}e- [}e=k 0) ‘$ -o . s+l1 a [sI- A] = (0 wn , Adj (st- A] = (° | ha ‘| 1 « (slay! = Ad Weal lo ss 8 s(s+1) {sl-A| s?+s 0 1 s+]Modern Control Theory 4-63 Solution of State Equations 1 @(i-et eM! = Wys-ay! ‘| ( | + e 0 an. ayo. [) se )| Ne. frwwesent-«) 0 et 1X20) e*X,(0) i ZSR = L {o(s)BU(s)}=1"! | . (jo oe! oo ss41) sg? 8 stl Lo(BUGy) = EYP eet ... Using partial fractions £ 1 s(s+1) 5 _, fot-a+o0* =e 1-et X, (0) +X, (O)a(1-e7 Jee tat-e tate ae! X(t) = ZIR + ZSR = 10) +X etx, (0) 407 -etet yit) = X,(0) = X,(0)+ eX, (0)- 0X, (Wet +e Tate ate uet But given y(t) = 2-are™! At t= 0, y= 2=X,(Ojy_g =X) and X,(0) = 0 XH = 2+ateT+eTalet -1) and X,() = et(i-e) if @ = 0X, ()=2 imp Example 4.35: Given the state model of a system ; (VTU: July /Aug.- 2005) * {0 1 0 x= [P, peli ¥=[1 op xModern Control Theory Solution of State Equations with initial conditions X(0) = Hl Determine : i) The state transition matrix. ii) The state transition equation X(t) and output Y(t) for an unit step input. iti) Inverse state transition matrix. Solution : i) As 0 1 i -liycs - -l : s+5 1 [sI-A] = i was “Adj paar [3 rapt. Aa BIAl [‘ ‘J [8 ‘] 7 = FEL s|= - fA] jst-A] fos s745s+4 (+1) (s+4) a a +1s+4) (s#1\(s+4 ois) = fat-ayt = {© s+4) (o+1Xer4) —i_ s (s+1Xs+4) (s+1Xs+4) 430 Y3 V3 _ 3 #5) = t5 "3 an st .. Partial fractions wel” Ged Gal aed ti) +5 1 aul ns) = pet Ge) ts+1) (+4) | [0] 71 ZIR = L'{6(s) BU(s)} = L “4 . : (4 (s+1) (s+4) {s+1) (s+4)}Modern Control Theory 4-65 Solution of State Equations i Wao W312 apse Ge] als sal sed 1 3 (s+1) (S44) ZIR = x(t) = Y(t) = yy = Latent Zot 43 12 iii) The property of state transition matrix is, ot(t) = of-t) fetal ot Vit _d gat oye |ae yd Land gn 3 3 im Example 4.36 : Determine the controllability and observability of the following state nrodel. 0 1 0 1 X=}o 0 1|X+]olu ~6 -ll -6 1 Y = [10 5 1x (VTU ; July/Aug.-2005) Solution : From the given model, fe 1 0 1 Az=/0 0 1{,B=/0),C= [105 1] -6 -N -6 1 . For controllability, Qs [B: AB: Aa] sul n=3Modern Control Theory Solution of State Equations 4-66 o 1 olf 0 AB=|0O 0 1] {of =! 1 -6 -11 -6} [1 -12 0 10 0 1 A’g = A[AB]=|0 0 1 i| = faz ~6 -11 6; [+12 ot} 1 0 1 Q,=|o 1 -2 1-12 61 Now j Q.|= -84 #0 hence rank of Q, =n = 3. The system is completely controllable. 2 For observability, 9, .[c* satel sat c"] 0 0 -6) fio -6 Alc’ = {1 0 -11| | 5] =/4 9 1 -6} [1 -1 > 0 0 -6) f-s) fe Ach s atfaTe™]=]1 0 -11) |-als[s 5 10-6 6 Q={5 -1. 5 1-105 Now |Q,|=96 #0 hence rank of Q, = The system is completely observable. ‘um Example 4.37: A system represented by following state model is controllable but not observable. Show that the non-observability is due to a pole-zero cancellation in C(sI-AI'B. (VTU:; July/Aug.-2005) . Oo 1 0 0 x = 10 oO TT) X+/0) w -6 -li ~6 1 Y = fi 1 0) x Solution : From the given model, 0 1 0 0 Az={0 0 1/,B=/0]/,C=f11 0] -6-11-6 1Modern Control Theory 4-67 Solution of State Equations 2 T s -1 Qo s° +6s+11 -6 —6s [sI-A] = /0 ss -1 |, Adj[sI-A] =| 5+6 4+5(5+6) -(11s+6) 6 11 s+6 1 8 3? s?+6stll 54600 1 Adj [sl- A] = -6 s(s+6) 0s -6s -(1ls+6) 3? teap = AMUSICA] __Adifsto aj ___Adi [st- A} jsl-Al si+6s?411s¢6 (s+) (s+2) (s+3) cays = fi 1 oy ALA} (s+1) (s+2)(s+ 3) 1 1 faojfs [$=-——s+! (s+1)(s+2)(s+3) 1 er is+2)6+3) 5 Thus the factor (s+1) gets cancelled from the numerator and denominator. This shows that as there is pole-zero cancellation in C{sI-AJ"! B which is the transfer function of the system, the system is not observable. hm} Example 4.38 : Consider the homogeneous equation X = AX where A is a 3x 3 matrix. The following three solution for three different initial conditions are available, et en2t ‘ae ] me! || -2e# |, | be Qe! a 0 i) Identify the initial conditions ii) Find the state transition matrix iii) Hence or otherwise find the system matrix A, (VTU: JanJ/Feb.-2006) Solution : i) For initial conditions, put t = 0. et 1 Xi) = |-ot XO) =) -1 2et 25Modern Control Theory 4-68 Solution of State Equations 1 Xi) = ea X(0) =} -2 =" 203 2 XD = “et X(0) =|-6 0 0 ° t . -1 ii) xi) = fet xO) =] 1 26" -2 —2e7! _ iy = |+4e Xoy=| 4 0 Q best ‘ X(t) = | +180" Xo) =| 18 0 Q A, Az A; X@) = |A, A, A,| xO) [A As Ag <1} fA, A, Ax ]f A AyrAy+2A,4 ay alijelaAy ay A,|ia ie A,-A,+2A, (2) 2] JA, Ay AgiL2 A,-Ag+2A, wl) f2] [Ar As As |p A, (Ay | 4l=|ay Ay A, 1/2 ice. Ay-2A lB) lo IA, Ag A,[Le A, -2A (6)Modern Control Theory 4-69 Solution of State Equations -6 A, A, AS 2] 2A, -6A, =-6 wl?) Il=|A, AS Ag ~6! ie 2A4-A, =18 wf) O} JA, Ag Ay o| 2A, -6A, =0 9) Solving the equations (1) to (9), the system matrix A is given by, 010 As |-6 -5 1 0 0-1 iii) State transition matrix t so 07 (s41) (9+5) = 6(s-+1) o |] Ist-A]=]6 0 s45 1] Adj{stbA} =] (+1) s(s+1) 0 0 s#l| 1 3 245846 st 46545 (stl) 1 1 —o(s +1) s(s+1) s [sl-ap = Adi [sIT A] _ Lo +8 {sl- Aj s> +65? +1146 s346s7411s+6 = (s+1) (s+2) (s+3) —st5_ (s+2)s43) (s+2)(s+3) (s+1¥s+ 21543) fst-ay? = JS (s+2)s+3) (s+2Ns+3) (s+1}(s+2)(s+3) 0 0 — s+1 3.2 1 y2 (s+2) (s+3) (s+2 (s+1) (s+2) - | 7% 4 8 2, 3 y2 2 92 | ...Partial (s+2) (+3) (8+2) (s+3) Gi? (s+#2) (343) fractions 1 0 0 (s+1)Modern Control Theory 4-70 Solution of State Equations 2 Poot og git _ tt Tyt lg lit 2 2 - ~ any eon - -u 1-5. 305 eM 2 Lsl- Ay t= jeer" +608 = 207 4 Be Set £20 Set 10 0 et Imm Example 4.39: Obtain the time response y(t) of the system given below by first transforming the state model into a ‘Canonical model‘, ¢VTU: Jan. /Feb.-2006) a 1 0 0 k=]o 0 14|X+loluy=[r0 ox -6 -1l -6 2 u is @ unit step function and x0) = [0.02] Solution : First to transform given model into canonical model, find modal matrix M. aA-1 60 [A-AJ = |O 2% -1 6 Il A+é [M-A] = +60 411,4+6=0 ie. (Ab1)(AF2)(A4 3) = ap =A =a a3 f-1 -1 0 Cy 6 1 For Ay =-1, [Al-A]=]0 -1 -1],M,=/C),/=|-6)=|-1 6 5 Cy 6 1 2-10 Cy 3 1 Ford, =-2, [AI-AJ=|0 2 =a],M,=/C,, |=|-6]=|-2 eo nia C5 12 4 -1 0 Cy 2) ft Fort, =~-3, [M-A]=|0 -3 -1),M,=|C, | =|-6|=|-3 1 3 cq; 9 11 Ms 4 -2 -3|= Vander monde matrix 14 4Modern Control Theory 4-71 Solution of State Equations AdjM = |-5 8 3] = 1 -2}u 1 . -1 0 0] At) : 2 0a ns) YH = fl 1 1) a) X(t) = MZ(t) ie. Z() = MX) 3°25 05] fa 1 40) = M'X(O)=/-3 4-1 oO} s|-2 1°15 05) [2} [1 et 0 0 et ato et 0 0 o ee “9 0 ]fa et ZR = e270) =} 0 e% 0 }f-2} =|-20# ZSR = 1 Lois)BU(s)}Modern Control Theory 4-72 Solution of State Equations — 0 0 s+1 1 eu) o = o | |-2} tys} o o -}?? s+3. ; 11 ss+0) Ss s+1 s{8+2) s s+2 1 pe 3 s(s+ 3) s $43 1-et 7SR = |-14e7# lls 33 Zit) = ZIR+ ZSR =| ~t~° 1,250 3 Y(t)= [2 1 1)zy = 1-1-0778 weed = 1,2 ,-3 wet 33 map Example 4.40 : Given the system X 3 0 | U. Find the input vector U(t) to give the following time response : (VTU : July/Aug.- 2006) X() = 6(1-e*) Solution : The given time response must satisfy the state equation. xi = “ [60] = +607! Xo = SB [se 207 +6(0 01] = -9e7 4804 L607!Modern Control Theory 4-73 Solution of State Equations 60 _f3 0 6(1-e') | 73 9) 9 4 ge" e607 | l2 Ty | 308! -207 4 + 6th" tb 6et - -i8+18e"' af? ly ~9e +867 +607! 6=6e7§ -3e7 42¢°" | [3 2 18-12e" _ -6+12et —6e 8 +6074 3 oy" 18-12e7* op oy 32) [64120 -6e +607 | [3 2 3 2 [3] [3 sl anal? °) [2 ra sot “Ws 2] [3 3} _|5 ° 32) 7B a we fb 3 2| 22 u 9 18-12e"! 1 1) |-6412e"! -6e +60 | 2 6-4e . uw = w» Required input vector o [-sosizes ~3e3! vet | a P tmp = Example 4.41: The following is the state space representation of a linear system whose eigen values are —3, -2, —1 (VTU : July/Aug.-2006) Qo 1 0 0 x=lo 0 1) pxpslo} « -6 -11 6 2 Given that u = 0, X(0) = [00 1]". Find X(). Solution : -6 -li -6Modern Control Theory 4-74 Solution of State Equations. t s -1 0 s?+6s+11 -6 6s fsI-A] = [0 5 -L |, Adj{sI-A]=| s+6 — s{(s+6) {IIs+6) 6 Il s+6 } oa 5 s st+ostth s+6 1 Adj [sl-A} = ~6 s+6) 8 -6s -(11s+6) s* [sl-A] = $9 +687 +11s+6=(s+1) (+2) (s+3) __s? +6s+11 s+6 1 s+2)(s+3) (s+1)(st2}(s+3) (s+1)(s+2)(s+ 3) -6 3(s+6) s “| (st1)(st2)s+3) (s+1)(s+2)(84+3) (s41)(s42)(84 3) —6s —(11s+6) 3° (s+1)(s+2)(s+3) (8+1)(s+2)(s+3) (s+1)(s+2)(s+3) 0.5 s#3 28 s+2 843 O34 45 s+] s#2 s+3 s+] s+2 $43 s+1 s+2 543 Se! de 1seF! OS Bet ae gest 2 = [x +607 307 2 beh 4 Be 4 Se 0.507 Bet 120! 4.94 2! -j6e! 413.5e3! O.5et -4e?! 4.4.5 0} [ 0.5et-e" 40.50% | X(t) =e x(a) =e [0] =]-0.5e7t +207! -1.5e 1 0.5et det 4450 | amp Example 4.42 : Find the fransilion matrix t) for a system whose system matrix is 5-1] given by As 304 | by the following techniques : 3 Aj DLaplace transform ii) infinite series iti) Cayley-Hamilton (VTU; July/Aug.-2006)Modern Control Theory 4-75 Solution of State Equations Solution : . _ 3 4 . 3° -1 i) Laplace transform method ji-ap = 2} | Adjfst~A] ("S os] [3 sel 3 845 [sI-A] = (s+5)(s+1)+3=s? +65 +8=(s+2)(s+4) s+ =I Isteay = Adjfsl-A] _ |(5+2)(5+4) (5+2) (544) ° jsI-A] 3 (s+5) (s+2)(s+4) (s+2)(s+4) -05 15 -05 05 s+2 +— std st2 sd eft = i st-apls 2 0.5e7 41,5e4 0.5e 7 +0. 5e 1 15e% -1.5e4 1.5e7 -0.5e7 li) Power series method ae fea 5 -1) [22 6 ~ {3 -1) [3 aj [Hs -2 22 6) f-5 -1 -92 -28)] 3. = “ve [is Ab 4 [sr 20 | 2,2 3,3 Oo teate At AL resesan 234 teat? 28154... = 6 $5 i ator? +140? =. Ite, iii) Cayley - Hamilton method Step 1: Find cigen values. (Ath 1 [AI-Aj =0 ie, (f° =0 j-3 AeModern Control Theory 4-76 Solution of State Equations. ie +6h48=0 bed, =-2, 4, =-4 Step 2: Construct R(A) RA) = Oy +04 =p(2) Where p(i) =e" Oy +042 =0% Step 3: Substituting values of A; and}, t At se? = Gy -2a, se and a, —4a, = Step 4: Solving above equations, “1 1 =! - @ te ug = 202 ent 1 2 Step 5: fA) = R(A) = aol+a,A ea eat{? OF] (1 ap 1 eat 3-1 Ate 2 _ godt TV o-tr_} oat ete aottcenll (tea tows ~0.5e 7 41.5e%" 9.507 40.5e71 ets 15e7! -1.5e" — 1.5e -0.5e°# ‘im Example 4.43 : Obtain the state transition matrix using : Qan/Feb,-2007) i) Laplace transformation method and ii) Cayley-Hamilton method. For the system describe by, rr x) = fi 2480 Solution : As [". 4d X(0} i) Laplace tansform method reap ~-f® 7]. agpaeay-[et? 2 I-A] = |g yg} ADDI-AT=] pS [sI-A] = s? +4s+4=(s+2)?Modern Control Theory 4-77 Solution of State Equations std tay = Adj[st=A] _](s+2)° (s+2 - |sI-A} = | (s+2)? (s+2)? 1 2 1 v2 +2)? (s+2)* 3 2 = <4 12 .» Partial fractions (s+2)? s+2 (542)? -2t -2t 1 At =I -1 _ je" +2te te | ee = LL [sl-A {s1-A] | mate gt 2te 2 | li) Cayley - Hamilton method Step 1: Find eigen values A 1 PI-Al=0 ie” |-9 ie M4044 =0 4 A+d| G+2)? Die. Ay =-2, Ay =-2 Step 2: Construct R(A) RA) = Gy #0,4=p(A) where p(X) =e" a, +0,h =e" Step 3: Substitute 4, Gy -2a, =e! For A,, as it is repeated use, ‘ + n@| a) larg ah) ely | “(oy taal ay ded =H a, ste and ay =2te* +e!Modern Control Theory 4-78 Solution of State Equations Step 5: (A) = R(A) = ul+a,A 10 of O 2 Me 27 4 eH [ tee 17 4 _ [2te? +e°2# ~4te?! iim Example 4.44 > Compute e for the given matrix : (VTU ; July/Aug.-2007) [clr abe 6] Solution: If A= AytAy, then et 2 (Ata Alt y gia! wl) This is the property of state transition matrix if AyAy = AyAy. So verify this condition. 6 o)fo w [0 6a 0 6] |-o 0} ~ |-6o o pa, 2 [2 8) [6 OP _ 7 0 605 mw |-@ 0] [0 6] ~ [-6 0 As A,A, = A,A;, the above property holds good. So calculate c”!'and e*2' and then multiply to obtain e*', AA, Mt = fsa]! s-6 0 ' s-6 0 [st-A,] = [ ° a) 4 [sl-A,] | 0 wal jst-A,| = @-6? : +o fa-a,pt » AdilstoAa) |r ! I-A, | o 5-6, ~ -a_fe* 0 ts E'[st-a,] | a s -0 . 5 w| [si-a,] = (3 Y | aafst-aa)-[ ‘|Modern Control Theory 4-79 Solution of State Equations. jt-A,| = s? +07 5 @ [sI-A r = Adj[st—Ag] _ s? s*+0" 2 |st-A.,] 8 sta s'+o* cost sin wt Agt = “1 Pos = = . vifst-Ay] sin at cos ct Ate S 0 (Sr sin wt -| “ise e* sin «| a et -sin@t cos wt ~e®'sin ote cos ct amp Example 4.45 : Obtain the time response of the following vector matrix differential X | fo 17/X,7 fo x tion, |") -| i hhh dy=fl olf] he (£) it st mn fH IEE H eset of] arnt input and the initial conditions are X(0) = X,(0) = 0. (VTU: July/Aug.-2007) o 1) 0 Solution : As (", 1) 2-[i}¢-0 0) sl . st5 1 [sl-A] = [: wes} Aailst-al-[*S ‘| [sl-A] = 57 +45s+6=(s+2)(s+3) s+5 1 Adjfsl-A] _]S*363) GHG) ~ -6 s -ay! = [sl-A] I-A =0(s) (s+2)(8+3) (8+2)(s+3) As X(0) = ( , e is not required. 0 Thus ZIR = ( ZSR L-Ho(s)BU(s)} wts)=t for unit step s+5 1 -a}l (s+2)(s+3) (s+2)(s+3)] [0] 1 why m6 os 7] (s+2)(s+3) (s+2)(s+3)Modern Control Theory 4-80 Solution of State Equations 1 w2 YR] = pl SS*2)(8#3)] _ pfs “S42 s43t i 1 | | - i { L@+2)(e+3) | L Sea se3 | -se fll a] XQ} = ZIR+ Z5SR=|6 2 3 i ett et - X00) yt) = fl offic =X yoy = bbe ty lyst 6 2 3 ‘ne Example 4.46: Obtain the observable phase variable state model of TF. 3 Abs? +bys +b = LED De TOE EES Draw the signal flow graph of Ts). U(s) 89 says? +ay8+a, (VTU ; July/Aug.-2007) Solution : The given transfer function is st4ays*4agsta,] Y(s)obys? +b,s? +b,5+b,[U(s)] 1 s8[¥(s)-byU(s)] +8? [a,¥(s)-b, L(s)] +s[a, Y(s)-b, U(s)] +[a ,¥(s)-b, Uls)] = 0 Divide the equation by s°, Ys)-b,Uls)+4[a ¥45)-b US} +4 [24 ¥(s)-by GSI] + 1 [a,%(s)-b,U(s)] = 0 3" Ss” 1 [ s Y(s)=yUls) +E fa Vs) +b, U(s)] ag ¥(s)+b,Us)] +4, [-a,¥()+b,5)] (1) 3 Now define the state variables as, X,(s) = E[b,t4s)-a,%s)+X,(8)] 2) X36) = }[b,U(s)-a,¥%(8)+X,(s)] lB)Modern Control Theory 4-82 Solution of State Equations tm Example 4.47: Obtain the controllable phase variable form of the transfer function, ) 3h ray = YES) bas? tbs? sys U(s) s? +a Fags tay (VTU: July/Aug.-2007) Solution: The given transfer function can be written as, YW) Ly, fb =a by )}? +(by ~azby )8+(b; ~a 3b) t8) jee a a Us) s*+a,s* +a,s+a, Y¥(s)_ = bo l(s)+¥%(s) Where His) = (oi ~abg)#" #(by “aabo) + (b3 ~asbo) Us) s +a 8 +a,sta, . Ys) = Xs) =Qs) (b, -a,by )s* +(by-ayby }s+(by-ayby) 8° +a,s7*+a,s+a, UG) = s3Q(s)+a,s7Q(s)+a,sQ{s)+a Qs) s*Q(s) = -a,37Q(s)-a,8Q(s)-a ,Q(s)+U(s) wal) Ys) = (b, -a,by)s7Q(s)+(b, -a yb, )sQ(s)+(b, -a by )Q(s) 2) Define the state variables as, X,(s) = Qs} X4(s) = sQ(sh—X5(s) = s7Qs) sX,(s) = X4(s) 8X,(s) = X5(s) Taking inverse Laplace transform, x = Xe Xa = Xa Substituting selected state variables in the equation (1), 5X,(s) =-a y%3(8)-agX_(s)-a 5X, (8)+ U(s) ie. Xa = -a,X,-a,X,-a,X,+UModern Control Theory 4-83 Solution of State Equations x1 o 1 0 x] 0 X2]=] 0 0 1 1X, j+olU * X3 a, “4, 4, x | 1 From (2), Ys) = (b, -a bg )X5(8)+(by -a by )Xp(8)+(b5 ~a.gby )X,(s) Y(s)_ = byL{s) +(b, —ayby )X5(s)+(b, a pb )X2 (5) +(b, —a shy )Xi(8) Taking inverse Laplace transform, ¥ = [bya yby}X; +[by ~agby] Xp +[b, ~ayby] X3 thy © = [bs ~a5b9) (by ~azby) (b;-ayb)], D = [bg] This is controllable canonical form of the given transfer function. Review Questions 1. What is homogencous and nonhomogeneous state equation ? 2. Define state transition matrix using classical method of obtaining solution. 3. What is zero input response and zero state response ? 4. Obtain the complete solution of nonhomogeneous state equation using time domain method. 5. State the importance of state transition matrix. 6. State the various properties of state transition matrix. 7. Obtain the solution of nonhomogeneous state equation using Laplace transform method. 8. What is resolvant matrix ? 3. Explain Laplace transform method of obtaining ” 10, Explain power series method of obtaining 4. 11. Explain Cayley Hamilton method of obtaining e“*. 12. Explain similarity transformation method of obtaining At 13. Obtain e** for following matrices using all the methods, a [ o 31 2-3 rte -t_ ett Ans. { zee ee —te"teteM eh 20 1 @ Hf | ; Ane . oe te 4 o1 “+2 0 ( 1 1 | cosvit qytinvn Ans.: LN2sinv2t__ cosv2tModern Control Theory 4-84 Solution of State Equations 44. Obtain the homogeneous solution of the equation 3{#) = A X(t) where a r wel | and x0) =|" . - 0. 1 1 6 al 1-1 bas[s ale X(0) 0.561 364 | 0.56 73.560) Ans.:| P22 0.11e" 0.49¢ ~0.45e™ 15. Find the output of the system having state model, : o 1 1 X(t} = [ 5| X(t) + (1) u(t) and ¥(t)= [1 — i] X(t) 1 The input UG) is unit step and X(0)} = [ al (Ans.: YW = 3+ He ® qase ty 16, Show that the folowing system is completely state controllable and observable. . -1 2 | 2 Xit}=| 0 Ty X(t) + | 0] Ut) i io al 1 and Yit}={1 1 O} X(t) 17, Find the solittion is, xt) “(i 2| X(t) with X(0) = [3] 1 (Ans. : X(t) = fans. XO Be 418. Find the solution of, : 01 1 = X(t) u(t xt tS al «(3 ] Mt) and Y= [1 OF X¢t) 1 if X(0) = [3] and system is subjected to the unit step-input. s tans. :¥0 =F ste bet 19. Define and explain state controllability. 20, Explain Kalman’s test for determining state controllability. 21. Explain Gilbert's test for determining state controllability. 22, What is output controllability ? How to eoutuate it for the system ? 23, State the relation between transfer function in s domain and controllability, obesroability of a system.Solution of State Equations Modern Control Theory . Define and explain concept of abserenbitity. 25, Explain Katman’s tes! for determining complete abservabitity 26. Explain Gilbert's test for determining complete observability. 7, Evaluate the controllability of the following systems with, aazl Lo B 0 -1 0, B=|o 0 | L3 0] (ans.:a, Conteollable, b. Controllable ¢, Not controllable) 28, Fealuate the obseroability of the following systems with, 210 . 3 waco a tpeal? 24 i400) to 0 2} -1 . nals Shee Wy 0 1 oO eA=| 0 0 Ibe=f45 y [-6 =I -6. (Ans. 1a. Observable b. and c. Not observable) goo(4 - 86)Pole Placement Technique 5.1 Introduction This chapter deals with the problem of pole placement. The pole placement technique or pole assignment technique is a method for design of control system which is discussed in this chapter. If it is assumed that all the state variables for a given system are measurable and available for the feedback then it can be proved that for a completely state controllable such system, its closed loop system poles can be placed at any desired location through state feedback by means of an appropriate state feedback gain matrix. The desired closed loop poles are firstly evaluated based on transient respanse and/or frequency response requirements such as say damping ratio, speed, bandwidth as well as the steady state requirements. Let us consider that the desired closed loop poles are ta be located at s=Gy, §=0.2........8=0,. Then by properly selecting gain matrix for state feedback, it is possible to locate the closed loop poles at the desired locations provided that the original system is completely state controllable. In the following sections, we will consider the control signal to be scalar. It can then be proved that the necessary and sufficient condition that the closed loop poles can be placed at any arbitrary locations in the s-plane is that the system is completely state controllable i.e. it is possible to transfer the system state from any initial state X(t) to any other desired state X(t) in a specified finite time interval (t)) by a contro] vector U(t), The required state feedback gain matrix can then be evaluated using various methods. In this chapter, it is considered that the control signal is a scalar quantity and not a vector quantity. Under such case, the mathematical aspects of the pole placement technique is complicated as the state feedback gain matrix is not unique. 5-1)Modern Control Theory 5-2 Pole Placement Technique 5.2 Pole Placement Design The conventional method of design of single input single output control system consists of design of a suitable controller or compensator in such a way that the dominant closed loop poles will have a desired damping ratio § and undamped natural frequency ,, The order of the system in this case is increased by 1 or 2 if there are no pole zero cancellation taking place. It is assumed in this method that the effects on the responses of non dominant closed loop poles to be negligible. Instead of specifying only the dominant closed loop poles in the conventional method of design, the pole placement technique describes all the closed loop poles which requires measurements of all state variables or inclusion of a state observer in the system. The system closed loop poles can be placed at arbitrarily chosen locations with the condition that the system is completely state controllable. This condition can be proved and the proof is given below. Consider a control system described by following state equation x = Ax+Bu (1) Here x is a state vector, u is a control signal which is scalar, A is nxn state matrix. B is nx1 constant matrix. The system defined by above equation represents open loop system. The state x is not fed back to the control signal u Let us select the control signal to Fig. 5.1 Open loop control system be u = — Kx state, This indicates that the control signal is obtained from. instantaneous state. This is called state feedback. The k is a matrix of order 1xn called state feedback gain matrix. Let us consider the control signal to be unconstrained. Substituting value of u in equation 1. = Ax+B(-Kx) =(A -BK)x w (2) The system defined by above equation is shown in the Fig. 5.2. It is a closed loop control system as the system state x is fed back to the control system as the system state x is fed back to control signal u. Thus this a system with state feedback. Fig. 5.2Modern Control Theory 5-3 Pole Placement Technique The solution of equation 2 is say x(t} =e, (0) is the initial state ~~ (3) The stability and the transient response characteristics are determined by the eigen values of matrix A — BK. Depending on the selection of state feedback gain matrix K, the matrix A — BK can be made asymptotically stable and it is possible to make x(t) approaching to zero as time t approaches to infinity provided x(0) # 0. The eigen values of matrix A - BK are called regulator poles. These regulator poles when placed in left half of s plane then x(t) approaches zero as time t approaches infinity. The problem of placing the closed loop poles at the desired location is called a pole placement problem. 5.3 Necessary and Sufficient Condition for Arbitrary Pole Placement Consider a control system defined by following state equation. x = Ax+Bu ~ (1) Let the control signal selected be u = ~ Kx X = Ax - BKx = (A - BK) x (2) The solution of the above equation is given as, x(t) = eA POE (0) ~@ The eigen values of the matrix (A — BK) are nothing but desired closed loop poles. The necessary and sufficient condition for arbitrary pole placement is that the system is completely state controllable. If suppose the system is not completely state controllable then there are eigen values of matrix A — BK that can not be controlled by state feedback. According to Kalman's test for state controllability, the system is said to be completely state controllable if the rank of the composite matrix Q. is n. Q, = [B+ AB:A7B: so AT TB] Let us suppose that the system is not completely state controllable. The rank of the controllability matrix is less than n. rank [Bs AB:A2B: v.00 tAP IB] = s A-BK 1 " (A -BK)* =A? -2ABK+ B?K? 2 = A?-ABK-ABK+B?K? =A? -ABK-BK(A -BK) = A*-ADK-BKA A? = (A-BK)> =A? ~A?BK-ABKA -BKA? I Now consider A345,A?+8)A 45g = A? -A?BK-ABKA -BKA? +5,(A? - ABK-BKA) + 85(A -BK)+8,1 A} ~A?DK-ADBKA -BKA? +5,A? -8,ABK - 8, BKA +8,A -6,BK+ 851 = 5,146,A+8,A? +A} -8,BK-6,ABK ~ 5,BKA—A?BK-ABKA -BKA? Wehave, (A) = 8:145,A46,A7 +443 = 0 (for n= 3) and Hl+5,A+6,A74A% = O(A)*OModern Control Theory 5-18 Pole Placement Technique substituting the above equations, o(A) = 0(A)-8,BK-8, ABK—8, BKA —A*BK-ABKA —BKA? But (A) = 0 " OA) = 8)BK+5,BKA +BKA? 46, ADK+ABKA +A °BK = B(5,K+5,KA +KA*)+AB(5,K+KA) +A7BK O(A) = [Bap A?B] fs, +5, KA +KA 2] | 5, K+KA \ boK | As the system is completely state controllable, the inverse of the controllability matirx Q. = [B:AB:A7B] exists. Multiplying bothsides by inverse of controllability matrix. ee | [usapsa2e] ofa) = | 3, K+KA i K J Premultiplying both sides of abave equation by {0 0 1] we get 0.0.1] [B:aB:A 2B] '9(A) = [0 0 1] 5.45, KA Fi) 8, K+KA | 1 [ K K K = [0 0 1)[B:AB:AB]o(A) The above equation gives the required state feedback gain matrix K for arbitrary positive integer n, 0 1]fB:AB:AtR .. eect AOUBY OCA) This equation is called Ackermann's formula for the evaluation of state feedback gain matrix K.Modern Control Theory 5-19 Pole Placement Technique 5.5 Selection of Location of Desired Closed Loop Poles The locations of desired closed loop poles must first be chosen in the pole placement design technique. The commonly used method for such selection of poles is based on the experiences obtained from root locus design of placing a dominant pair of closed loop poles, The other poles are selected in such a way that they are away from dominant closed loop poles in left half of s plane. If the dominant closed loop poles are placed away from jo or imaginary axis then the response obtained from the system is very quick but at the cost of system becoming nonlinear which should be avoided. The other method of selection of poles is the quadratic optimal control technique. In this method, the desired closed loop poles are determined in such a way that it balances between the acceptable response and the amount of control energy required. The high speed response indicates that large amount of control energy is required. In addition to this a larger and heavier actuator is required which increases the cost. imp Example 5.1: Consider the system defined by x = Ax+Bu o 1 0 0 where A={0 0 1), B=/0 -1 -5 -6 1 By using the state feedback control u = — Kx, it is desired to have the closed loop poles ats =-14)2, s = - 10. Determine the state feedback gain matrix K. Solution : Let us first check the controllability matrix of the system. The controllability matrix, Q. = [B: AB: A7B] 0 0 1 O7}fo} fo B= 0]; AB=/0 0 1]{/o}=|1 | 1 -1 -5 -6][t] [-4) 0 ljs|-l - -6 -1 -5 -6]j[-1 -5 —-—4 +6 29 31Modern Control Theory 5-20 Pole Placement Technique 004 Q=]0 1 6]; [Qj=t0-Ns-1 1-6 3 The rank of matrix Q, is 3. The system is controllable and hence it is possible to place the poles arbitrarily. Now the state, feedback gain matrix K can be obtained by using any of the three methods discussed previously. Method 1 : Using transformation matrix T a1 0 Az=/0 01 “1-5 ~6 The characteristic equation for the system is, 100) fo 1 0 sa 0 |st-A} = |s}o 1 of-lo a 1 |[=|}o s -1 ee nn) 15 s+6 = s[s(s+6)+5]+][1] = 5 [s? +6s+5]+1 = so+6s745s+1=0 Comparing this equation with 53 +a)s? +a,s+a 5 =0 % a) 56, a) =5,a,=1 The desired characteristic equation is , (s+1-[2)(s+14)2)(s+10) = [cst0? +4] (s+10)=[5? +2545] [+10] = $9 4+10s? +20s+2s? +5s+50 = 5341257 4255450 2 6, =12, 6, =25, 8, = 0 K = [8;-a, 6,-a, 6,-a,]T* As the given stale equation is given in controllable canonical from T = 1 Tel ee K = [8;-a; 6,-a, 6,-a,] =(50-1 25-5 12-6] [a9 20 6] nm "Modern Control. Theory 5-22 Pole Placement Technique oo a}fo s[-1 -5 -6//0 0 1 6 29 31j|-1 -5 -6 -1 5 +6 =|6 2% 31 -31 -149 157 0 1 07 fo 1 ofa 4 6 01 =j0 0 1 . Qo ‘ ‘5 -1 5 -6 -} -5 6} [-1 5 ~ 6 29 31 1 5 + 001 1 0 100 (A) 6 2 BL i411 5 -6 wd a 1/|+50/0 1 0 “31-149 ~157 -6 29 31 -1 55 -6 o01 49° 20 6 = | 18 19 -16 -128 74 115 4 00 1 [B ab: AB] 01 -6 1-6 31 oo 17'f49 2 6 K = [0 0 1]/0 1 -6] | 18 19 16 1 -6 31} |-128 74 115 5 6 1]/f 49 2 6 = [0 0 1]/6 1 O}] 18 19 16 1 0 Of[-128 74 115 9 2 6 = fl o o]| 18 19 -16 -128 74 115 [49 20 6]Modem Control Theory 5-23 Pole Placement Technique ums Example 5.2 : It is desired to place the closed loop poles of the following system at s = -3 and s = — 4 by a state feedback controller with the control u =— Kx. Determine the state feedback gain matrix K and the control signal, (VTU : Aug.-2005) is (‘ oll y= [l o}x Solution : Equating the state equations with standard equations with standard form x=Ax4+Bu and y=Cx we get a=[f 5} 5 ep] oo ota Let us check the controllability matrix of the system Q = [B :AB] ae [ S)EH LI a=[ 4) The rank of above matrix is 2. Hence the given system is completely controllable. 0 The given equation is not in the controllabie canonical form because B is other than. ( we use method of transformation matrix T to get state feedback gain matrix. T=Q.W 0 2 a, 4 3.1) = : Wel) |= a: [: 4 ; ; al [ o} 0 1 <1 consider A = (S mE |sI-Aj= 1 s+ { 3 [86 +3) +1 =s° + 38+ 7 ating with s* + a, s + a, we get B 1 2 gelModern Control Theory 5-25 Pole Placement Technique 5.7 Concept of State Observer In case of state observer, the state variables are estimated based on the measurements of the output and control variables. The concept of observability plays important part here in case of state observer. Consider a system defined by following state equations x = Axt+Bu y= Cx Let us consider X as the observed state vectors. The observer is basically a subsystem which reconstructs the state vector of the system. The mathematical model of the observer is same as that of the plant except the inclusion of additional term consisting of estimation error to compensate for inaccuracies in matrices A and B and the lack of the initial error. The estimation error or the observation error is the difference between the measured output and the estimated output. The initial error is the difference between the initial state and the initial estimated state. Thus the mathematical model of the observer can be defined as, xe 1 AX4 Bu +K,(y-CX) = AK+Bu+K,y+KeCK (A-K,QX+Butk,y Here % is the estimated state and CX is the estimated output. The observer has inputs of output y and control input u. Matrix K, is called the observer gain matrix. It is nothing but weighting matrix for the correction term which contains the difference between the measured output y and the estimated output CX. This additional term continuously corrects the model output and the performance of the observer is improved. " Full order state observer The system equations are already defined as x = Ax+Bu yu & The mathematical model of the state observer is taken as,Modem Control Theory 5-26 Pole Placement Technique XK = AK+Bu+K,(y-CX) . . To determine the observer error equation, subtracting equation of X from x we get rk < {Ax+Bu]~[Ak+Bu+K,(y-Cx)] = A(x-8) -K, C(x-%) x-% = (A-K,C) (x8) Let e= x-k “ean es (A-K,Oe The block diagram of the system and full order state observer is shown in the Fig. 5.3. Full order state observer Fig. 5.3 The dynamic behavior of the error vector is obtained from the eigen values of matrix A-K,C. If matrix A-K,C is a stable matrix then the error vector will converge to zero for any initial error vector e(0). Hence X(t) will converge to x(t) irrespective of values of x(0) and X(0). If the eigen values of matrix A-K,C are selected in such a manner that the dynamic behaviour of the error vector is asymptotically stable and is sufficiently fast then any of the error vector will tend to zero with sufficient speed. If the system is completely observable then it can be shown that it is possible to select matrix K, such that A-K,C has arbitrarily desired eigen values. ic. observer gain matrix K, can be obtained to get the desired matrix A-K,C.Modern Control Theory 5-29 Pole Placement Technique Here T= Q.We[ Ch: Arce: (ayes. (A) Cw For the original system the abservability matrix is, (ANT CH = N NW = Wh = Wa [Crarct .. 4 u 4 i As w= W (ryt = (wy? Taking conjugate transpose of both sides of equation I Buty 8,78 8, Ay 8,1 And Se 7 nel *a(tly Baa ent = (Ty = sl kt =(T) i (Ty =(WN") 4 8-8, n-l @n-2 a2 8a3 Here fa, 4 1 0 0 0 N = [Ctrarcr:(ayice:. ay tory The second method is based on direct substitution method to get the state observer gain matrix K,. This is applicable for system having low order. The matrix K, is directly substituted in the desired characteristic polynomial. Let us consider on = Substitute this matrix into desired characteristic polynomial. [SI-(A-K,O = (8-04, ) (8-0 poe (Sty) By equating the like power coefficients of s on both sides, the values of K,,, K,g can be determined. This method is restricted for value of n at the most 3. The third method is use of Ackermann's formula. For the original system defined as Ky and x=Ax+Bu and y = Cx we have already derived the formula known as Ackermann's formula given as,Modern Control Theory 5-30 Pole Placement Technique A™ IB] o (A) 1] [B:AB:.. Now let us consider the dual system 2 Aw +0'V y = BY The Ackermann's formula for dual system is modified as K =[0 0... Vp cr: atc H(A)” CP o(as) Both observer gain matrix, K, = K*. | e yt 0 foc 0 cA 0 CA 0 uKy = Kt = O(a : Ey] = QA) : 3 . can? : [cart 1 cA" The desired characteristic polynomial for state observer is O(8) = (8-0, )(8-Oy )eneen(S-G,) watt, are desired cigen values, Here tt), a5. Equation (II) is Ackermann’s formula for determination of observer gain matrix K,. 5.11 Selection of Suitable Value of Observer Gain Matrix K, The feedback signal through the observer gain matrix K, acts as a correction signal to the plant model to account for the unknowns in the plant. If the unknowns involved are significant then the feedback signal through matrix K, is comparatively large. But if the output signal is affected by disturbances and measurement noises then output y is not reliable and feedback signal through matrix K, should be small. The effects of disturbances and noises involved in output y should be neatly studied while evaluating matrix K,. The observer gain matrix is dependent on the desired characteristic equation (S-G,)(5—C2)occe(SG,,) = 0. The selection of 0,0)... is many times not unique. But normally the observer poles should be two to five times faster than the controller poles. This makes the observation or estimation error converging to zero quickly. The observer estimation error decays two to five times faster than docs the state vector x. Due to this faster decay of observer error, the system response is dominated by controller poles.Modern Control Theary 5-31 Pole Placement Technique With the appreciable noise produced by sensor then the observer poles can be selected to be slower than two times the controller poles. The bandwidth of the system will then be less and noise will be smoothened. The system response is dependent on observer poles in such case. The observer poles are located to the right of controller poles in left half of s-plane so that the system response is influenced by observer poles. While designing the state observer, it is required to evaluate various observer gain matrices K, based on number of various desired characteristic equations. For each of these different matrices K,, the simulation tests can be run to check the overall system performance. Based on the performance, the best possible K, can be selected. In many systems, the best vaiue of K, is obtained after a compromise between fast response and sensitivity to disturbances and noise signals. ump Example 5.3 Consider the system represented by ao 1 0 0 r=]0 0 1] x +0lu i -h ~~ 1 y = [1 0 fx Design a full order observer such that the observer eigen values are at ~24 j2V3 and -5. (VTU: Aug-2005) Solution : fo 1 0 0 Aalo oI ; eel 3 c=ft 0 0] \-6 - ~6 1} Let us check the rank of dual system of the original system [Cre arc: (AYP Cr: (AN cry Here n = 3 nP Ch: Att: (ANP CH] 1 oo -6 [*] 0 @efol; act=|1 0 -n jo} = 0 jo 1 -6} [oj {o oOo -6)f0 0 (ary? = Av. A™=1/0 O =11}]1 O -11)/=/0 =11 60 oO 1 -6| [0 1Modern Control Theory 5-34 Pole Placement Technique 1 3 0)]f2 3 0 7 0 6 (ays [2 -1 2} ]2 -1 2/=j0 7 -2 9 1 O}](@ 1 0 2-12 7 0 6)fo] [6 cayrce = lo 7 -2})0)/=|-2 2-1 21} [2 00 6 {ce:atces(aytce] = |o 2 -2 102 The rank of above matrix 3. Hence the original system is observable. Ka Let K, = [Ky be observer gain matrix Kos The desired characteristic polynomial is, [s-(-3+i1)] [s-(-3-in)] [s-(-4)) = [+37 +1] [s+4] (s? #68410) (344) = 59 +65? +108 +48? +245 +40 = s3+10s?+34s+40 Now consider | sI-(A +K, -C) = s 0 0] [fi 2 0] [Ky erie tap ee|e ee 00 s} [lo 2 o s 00] [fi 2 0) foo Ky = |]o s o/-}]3 2 1]-]o 0 Ky 90s} {lo 2 0 00 Ky soo} fi 2 Ky = || s 0|-|3 -1 1-K, oos}|o 2 -K, 1 o s+] -14K 0 -2 s#KyModern Control Theory 5-37 Pole Placement Technique = (A-K,C) (x-¥) e = (A-K,.C)e » () Equations (I) and (II) can be combined and rewritten in matrix form as, x A-BK BK x “fo [ 0 ancl ( ~~ Om e Equation (III) shows the dynamics of observed state feedback control system. The characteristic equation for the system is sl-A+BK -BK 9 sI-A+K,C [sl-A+BK}|sI-A+K,C| = 0 The closed loop poles of observed state feedback system consists of the poles due to pole placement design alone and poles due to observer design alone. Thus the pole placement design and observer design are independent of each other, The design of each can be done independently and it can be combined to get observed state feedback control system. For n as the order of the plant, the order of the observer is also n while the resulting characteristic equation of the total closed loop system is of order 2n. To derive the transfer function of observer based controller, let us consider the system defined as x = Ax + Bu y= & Let the plant be completely observable and let the observed state feedback control is us- Kx wis AX+Bu+K,(y -CX) AK + B(-Kx)+K,(y-CX) 2X a (A-BK-K,.O)X+ Ky Taking Laplace transiv.m of above equation sk(s)-X(0) = (A-BK-K,C)X(s)+K, ¥(s)Modern Control Theory 5-38 Pole Placement Technique Assuming zero initial condition, [sl-(A-BK-K,C)] ¥(s) = K, ¥(s) Xs) = (sl-A+BK+K,C)*KY() Substituting this value of X(s) in equation for u, UG) = -K Xs) Us) = - K (sl - A + BK + KC)" K, Yo) The transfer function is given by, Us) _ “1 % - -K(sl-A +K,C+BK)'K, Us) _ _ “lp ie" K(sl-A +K,C+ BK) 'K, The block diagram representation of above system is shown in the Fig. 5.5 Ris)= 0+ -¥(s) us) ¥(s) Fig. 5.5 The transfer function obtained above is called observer based controller transfer function. Tt can be seen that the observer controller matrix A - K.C ~ BK may or may not be stable even though A — BK and A - K,C are selected to be stable. In certain cases, the matrix A - K,C - BK may be poorly stable or even unstable also. Examples with Solutions ‘> Example 5.5: A system represented by following state model is controllable but not observable. Show that the non-observability is due to a pole-zero cancellation in Clsl ~ A} (VTU: July/Aug.-2005) . o 1 «0 0 x=] 0 1 |x+lofu -6 -11 -6} [1 y= fi 1 o}xModern Control Theory §-39 Pole Placement Technique Solution : o 1 0 o x = 0 0 1 |x+lO}u -6 -ll -6 1 Y= f1 1 o}x o 1 0 0 Here, Az/0 oO 1]; B= |0)/; C= fl 1 0} ~6 -1l -6 1 s 00 o 1 0 s 1 0 [sE-A] = |0 s O|- oO 1a}=|O s -1 0 0 s| [-6 -ll -6 6 ll sro |sl-A| = s[(s+6)s+11]+1[6] = s?+6s* +11s+6 = (s+1)(s+2)(s+3) Adj of fsl-A si-ay? = Adiof I-A] |sl-A} iT [st +6s411 1 +5+6 * (ssljet2(523) 1 1 st+6stll s+6 1 ~ WHjerD era) ‘ -6 s(s+6) -68 =(lis+6) s? ‘ = C[st-A}'B+D = CIsi-AT'B = ¢ D=0) 3 1 fst4+os+1l s+6 1 [0 =f 1 oj) | s(s+6) 8 [lo (3+1)(s+2)(s+3) 6s -(i1s+6) 2 1 0 s?+6s+11-6 (54+6)+s(5+6) 145] A Ieee F1j(s+2)(5+3) t 1 = Ganesaery beModern Control Theory 5-40 Pole Placement Technique s+] (s41)(s+2)(s+ 3) Ys) _ stl TR = = —_—_ $+] _ Us) — (s#1)(s#2)(s+3) Clearly it is seen that the factor (s+1) is cancelled from numerator and denominator. +1 . s+] TF = 468° 41ist6 — [s(s+6)+11]s+6 Let us draw the simulation of the transfer function. Fig. 5.6 There is no coupling between x5 and y due to which state x3 can not be observed from measurement of output y. Thus the system is not completely observable due to pole-zero cancellation as matrix C contains one of the terms as zero. imp Example 5.6: A regulator system has the plant (VTU: Jan./Feb.-2006) a0 -6 1 x = |1 0-11} x4fOlu, y=[0 0 If. ol -6 0 i) Compute K so that the control law uw = ~ Kx + r{t), r(t) = referance input, places the closed loop poles at -2 + jVT2,-5. ) ii) Design an observer such that the eigen values of the observer are located at 2+ jV12,5. 6) iii) Draw a block diagram implementation of the control configuration. @ iv) Obtain the stale model of the observer based state feed back control system. @Modern Control Theory 5-41 Pole Placement Technique Solution : . oo 4 1 x = {1 0 -11) x+/0Ju o1 +4 0 y = [0 0 1)x oo -6 1 A=j1 0-11]; Bes /0}; C= fo 0 1] o1 -6 0 i) Let us first check the controllability matrix of the system the controllability matrix for n = 3 is given by, Q. = [B: AB: A*B] 1 00 ~6)f1 0 B=/0]; AB =]1 0 -11{/o] = |1 0 o1 -{]0 a 6) | + 23 0 ~6 36){1 0 A’B = |0 -11 60]}0] = jo 1 -6 25)(0 1 10 0 Q. =]o 1 0} |Q +10) =1e0 oo1 Rank of Q. = 3 The system is completely state controllable as the rank of controllability matrix Q.=n=3 Now the desired characteristics equation is, bs--8)l[s-(-2+#72)][s-(2-W2)] (s+5)[(s+2)-jvI2] [(s+2)+jvI2] (5+5)[(s+2)? +12] " W (s+5)[s? +4s+16]Modern Control Theory 5-46 Pole Placement Technique 00 ‘We) 1] AB = |0 -6 -5 | -5| J a 8 SS 0 Q. = {0 1 3); [Q,| = 0-1) =-1e0 1 3 Rank of Q, The system is completely state controllable as rank of matrix Q, = n. Hence it is possible to place the poles arbitrarily. Let us now find the state feedback gain matrix by use of transformation matrix T. 01 0 A=]|0 0 1 0-6 -5 The characteristics equation for the system is, s 0 0|//0 1 0 s -l O Jsl-A| = [10 s 0j-jo o 1fi=|jo s = 0.0 s} [0 +4 -5 0 6 s+5 s[s(s+5)+6]-(-1) [0] s[s? +5546] = 53 455? +6s Comparing the above equation with s? +a,s? +a,s+a, =0 a=5 ; ay=6; ag=0 The desired characteristic equation is, [s-Gi+f)][s-@ -)] b-(-3)] i [(s+1)-i]Ks+1) + jt] +5] [(s+1)? +1] +5) (3? +28+2)(s45) = 59475? 4128410 Comparing above equation with s°+ 3s" + 8,5 + 6, we have 6; = 7; 6, = 12; 8, = 10 K = [6;-a,; §,-a, 6,-a,] T7? As the given state equation is given in controllable form T = 1 TI 21Modern Control Theory 5-47 Pole Placement Technique K [10-0 12-6 7-5]1 100 [1 6 2])0 1 O}=f10 6 2] oo01 The state feedback gain matrix, K = [10 6 2] ‘> Example 5.8: Consider a linear system described by the transfer function Dd -Design a feedback controller with a state feedback so that closed loop poles are placed at - 2,~ 1+). (VTU ; July/Aug.-2006) Solution : The transfer furiction is given as , 10 10 10 The denominator is decomposed as below, s43s7+2s = [(s+3)s+2]s Let us simulate this denominator vit) Fig. 5.9 ‘The complete state diagram is as shown in the following Fig. 5.10 after simulating the numerator by shifting take off point. Fig. 5.10Modern Control Theory 5-49 Pole Placement Technique Now the desired characteristics equation is, (s-(2)[s-(-1 +] [s-CA-)] = (+2) (541 +71) (541-11) (54 2\[(s+1)? #1] = (s+2)(s? +2542) 342s? +25+2s? +4544 = sh 445" +6544 Comparing above equation with s+ 3,3" + 55 + 5, we have, 6, =4; 6,56 6,54 Let us now find state feedback gain matrix by direct substitution method. Let the desired state feedback gain matrix K be, K = [K, kK, K;] s 0 0] fo 1 0] fo t Consider, jsI-A+BK| = |]0 s 0/-|0 0 1)+{O0|[K, K, Ky] oo 0-2 -3] [1 soo} fo 1 o] fo o oO = {jo s O|-/0 0 2iH0 0 O {0 0 sj} [0 -2 -3) [K, K, K; s -l 0 =i]]/O os 1 K, 2+K, s+3+K, = s[s(s+3+K,)-(-1)(2+K;)+1[K,]] = s[s? +3s+Kys+241K, ]+K, = s94(3+K,)s?4+(2+K,)s+Ky Equating coefficients of above equation with those from desired equation, 3+Kj=4; Kjy=1l 2+K,=6 .K)=4 Kj=4 +. State feedback gain matrix, K =[4 4 1]Modern Contral Theory 5-50 Pole Placement Technique a> Example 5.9: Consider the system described by the state model x = Ax, Y = Cx where 11) A “| 1 2} Cali O] Design a full order state observer. The desired eigen values for the observer matrix are uy = 5, uy = -5. (vTUsulyrAug.-20061 Solution : . [7 1 AU = 1 2 hed Let us check the controllability matrix of dual of given system for n = 2 Q = [cs atc] ve (HE 1 UV gj-1e0 Q@ =I y |!Qel = * Rank of Q, = 2. = Number of state variables The dual system is completely state controllable «Original system is observable. Kay] Let K, = | be observer gain matrix 2 ‘The desired characteristics equation is, [s-(-5)lfs-(-5)] = (5+5)(s+5) = s?410s+25 Comparing with x? +5,s+8, we have §, =10, 8, =25 Consider, wevone GE THE 2 [bake ofModern Control Theory 5-51 Pole Placement Technique (st1+K, (542) 1+K 3) (s#1+K, )(s+2)+1(-1+K,,) W 3? +s+K,)5+25+2+2K -1+K., s?+(1+K.) +2)s+(1+2K +K,2) = 8° +(34Ky )s+(1+2K +K.) Comparing the coefficients in above equation with those of desired characteristic equation. 34+Ky = 10; Ky=7 142K +K9 = 25; Ky =25-1-2K,, =24-2(7)=10 The observer gain matrix K, is given as, «= [7] ‘ua Example 5.10 : Consider the system defined by Y= Ax + Bu, where fo 10 0} As 0 6 1} B=lO| By using state feedback control u = -Kx, it us desired to have (1 35 4 1) the closed loop poles at s =-24)4 and s = ~10. Determine the state feedback gain matrix “K" by any one method. (VTU : JanJFeb.- 2008) Solution : We have 1 0 0 A= oo1 B=i0 a) a Let us check the controllability matrix of the system. Here n = 3. The controllability matrix , Q,=[B: AB : A7B] 0) ol if 0] B= {0 AB=]0 0 1 }Jo}=| 1] tj -1 -5 -6}{1] [-6] o 1 olfo 1 0 fo o 1 AXeAAz=lo o rf/o o f= -5 -6 -1 -5 -«|[-1 5 -6} [6 2 31* Modern Control Theory 5-52 Pole Placement Technique a’B = |-1 -5 -6/|0]=|-6 6 29 Sifla} [aa oo 1 o 1 6 + /Q.| =1[31-(36)]=-5 + 0 1 ~6 31 Q Rank of matrix Q, is 3. The system is completely state controllable as rank of matrix Q, =n. Let us now find state feedback gain matrix by direct substitution method. The desired characteristic equation is, [s-(-2+i)] [s-(-2-J9)] [s-(-10)] = [(s+2)-H](s+2)+ i]s +10] = [(s+27 +16] [s+10] = [s? +45+20][s+10] = $3 + lds? +605 +200 Comparing above equation with s7 +5,57 +8,5+3, we have 8, =14, 8, = 60, 6,= 200 Let the desired feedback gain matrix Kbe K = [K, Ky K,] Consider, |[s 0 0] fo 1 0} Jo jsI-A+BK} = |}o s o]-|0 0 1 |+/0][K, K, Ky] loo s} la 5 +] fp fs oo} fo 1 0] fo o 0 =|}0 s o|-lo 0 1/4/0 0 0 ilo s] |W -5 +] [K, K, K, s -l1 0 = 0 $ -1 14K, 54K, st+o+K.y = s[s(s+6+K,)+#(5+K, )]+1[1+K,] s[s? +(6+K)s+(5+K,)]+(1+K,) WW = s?+(6+K;)s” +(5+K,)s+(1+K,)Modern Control Theory 5-57 Pole Placement Technique c y'fo K. = o(A}) CA 0 cA7y [1 74 25 371 0 0} fo = {-18 41 7 {)/0 1 O}]0 42-95 -1]/0 0 1]}1 74 25 3) fo 3 K, = |[-18 41 7] /o] =]|7 -42 -95 -1|/|1 -1 This is the required observer feedback gain matrix by Ackermann’s formula. Review Questions |. Whet do you mean be pole placement problem? . Prove the necessary and sufficient condition for arbitrary pole placement? Write short note of coaluation of state feedback gain matrix. . What are the different methods of evaluating state feedback gain matrix? Explain any one method in detail. . Describe Ackermanit’s formula method for determining the state feedback gain matrix K. |. Write a note on selection of location of desired closed loop poles. . What do you mean by state observers? |. Write a note on full order state observer? ). What is dual probtem? . What és necessary and sufficient condition for state observation. . What are different methods of evaluation of state observer gain matrix K,? . Write a note on selection of suitable value of observer gain matrix K,. . Write a short note on observer based controller. Derive the transfer function of observer based controller. . Consider the system defined by x= Ar+Bu o 1 0 0 where A 0 0 14, B=|0 1-45 -6 1 By using ihe state feedback control u = -Kx, it is desired to have the closed loop poles at 5 = 2+ jf and s = -10. Determine the state feedback gaint matrix K. (Ans.: K=[199 55 8]Modern Control Theory 5-59 Pole Placement Technique 22. Consider the systemt represented by ool 2] 0 -6 -ll -6 1 x=]0 0 fal y = [t 0 o]x Design a full order observer sttch that the observer eigen values are at -2ip2V3 and - 5. Guly/Aug.-2005) 000Controllers 6.1 Background The concept of a control system is to sense deviation of the output from the desired value and correct it, till the desired output is achieved. The deviation of the actual output form its desired value is called an error. The measurement of error is possible because of feedback. The feedback allows us to compare the actual output with its desired value to generate the error. The error is denoted as e(t). The desired value of the output is also called reference input or a set point. The error obtained is required to be analysed to take the proper corrective action. The controller is an element which accepts the error in some form and decides the proper corrective action. The output of the controller is then applied to the process or final control element. This brings the output back to its desired set point value. The controller is the heart of a control system. The accuracy of the entire system depends on how sensitive is the controller to the error detected and how it is manipulating such an error. The controller has its own logic to handle the error. Now a days for better accuracy, the digital controllers such as microprocessors, microcontrollers, computers are used. Such controllers execute certain algorithm to calculate the manipulating signal. This chapter explains the basic discontinuous controllers such as on-off controller, continuous controllers such as proportional, integral etc. and composite controllers such as proportional plus integral, proportional plus derivative etc. Let us study first the general properties of the controller. 6.2 Properties of Controller Consider a control system shown in the Fig. 6.1 which includes a controller. The actual output is sensed by a sensor and converted to a proper feedback signal b(t) using a feedback clement, The sct point value is the reference input r(t). For example the actual output variable may be temperature but using the thermocouple as the feedback element, the feedback signal b(t) is an electrical voltage. This is then compared with reference input which is also an electrical voltage. The thermocouple senses the output temperature and produces the corresponding electrical em.f as the feedback signal. Hence actual output variable sensed and the feedback signal may be having different forms. (6 - 1)Modern Control Theory 6-2 Controllers Controller output e (p) Controlled output ott) Sensing Feedback element Fig. 6.1 Basic control system 6.2.1 Error The error detector compares the feedback signal b(t) with the reference input r(t) to generate an error. This gives an absolute indication of an error. For example if the set point for a range of 5 mV to 20 mV is 12 mV and the feedback signal is 11.8 mV then error is 0.2 mV. But actual vairable to be controlled may be different such as temperature, pressure etc. Hence to obtain correct information from the error, it is expressed in percentage form related to the controller operation. It is expressed as the percentage of the measured variable range. The range of the measured variable b(t) is also called span. Thus span = byax —Pmin Hence error can be expressed as percent of span as, t=-b ee, = po 100 P Binax ~Bmin Where ¢, = error as % of span ‘> Example 6.1: Tie range of measured variable for a certain control system is 2 mV to 12 mV and a setpoint of 7 mV. Find the error as percent of span when the measured variable is 6.5 mV. Solution: by, =12mMV, Buin =2mV, ba6SmV, r= 7mV r-b 7-65 ep = Ban -Dam «100 = Ti-3 x 100 = 5%Modern Control Theory 6-3 Controllers 6.2.2 Variable Range In practical systems, the controlled variable has a range of values within which the control is required to be maintained. This range is specified as the maximum and minimum values allowed for the controlled variable. It can be specified as some nominal values and plus-minus tolerance allowed about this value. Such range is important for the design of contollers. 6.2.3 Controller Output Range Similar to the controlled variable, a range is associated with a controller output variable. It is also specified interms of the maximum and minimum values. But often the controller output is expressed as a percentage where minimum contraller output is 0% and maximum controller output is 100 %. But 0 % controller output dees not mean, zero output. For example it is necessary requirement of the system that a steam flow corresponding to y's opening of the valves should be mininum. Thus 0% controller output in such case corresponds to the y* opening of the valve. The controller output as a percent of full scale when the output changes within the specified range is expressed as, Where p = controller output as a percent of full scale = value of the output Ujax = Maximum value of controlling variable Umin = minimum value of controlling variable 6.2.4 Control Lag The control system can have a lag associated with it. The control lag is the time required by the process and contraller loop to make the necessary changes to obtain the output at its setpoint. The control lag must be compared with the process lag while designing the controllers. For example in a process a valve is required to be open or closed for controlling the output variable. Physically the action of opening or closing of the valve is very slow and is the part of the process lag. In such a case there is no point in designing a fast controller than the process lag. 6.2.5 Dead Zone Many a times a dead zone is associated with a process control loop. The time corresponding to dead zone is called dead time. The time elapsed between the instant when error occurs and the instant when first corrective action occurs is called dead time.Modern Control Theory 6-4 Controllers Nothing happens in the system, during this time though the error occurs. This part is also called dead band. The effect of such dead time must be considered while the design of the controllers. 6.3 Classification of Controllers The classification of the controllers is based on the response of the controller and mode of operation of the controller. For example in a simple temperature control of a room, the heater is to be controlled. It should be switched on or off by the controller when temperature crosses its setpoint. Such an operation of the controller is called discontinuous operation and the mode of operation is called discontinuous mode of controller. But in some process control systems, simple on/off decision is not sufficient. For example, controlling the steam flow by opening or closing the valve. In such case a smooth opening or closing of valve is necessary. The controller in such a case is said to be operating in a continuous mode. Thus the controllers are basically classified as discontinuous controllers and continuous controllers. The discontinuous mode controllers are further classified as ON-OFF controllers and multiposition controllers. The continuous mode controllers are further classified as proportional controllers, integral controllers and derivative controllers. Some continuous mode controllers can be combined to obtain composite controller mode. The examples of such composite controllers are PI, PD and PID controllers. The most of the controllers are placed in the forward path of control system. But in some cases, input to the controller is controlled through a feedback path. The example of such a controller is rate feedback contreller. In this chapter, only continuous and composite controllers are discussed. 6.4 Continuous Controller Modes In the discontinuous controller mode, the output of the controller is discontinuous and not smoothly varying. But in the continuous controller mode, the controller output varies smoothly proportional to the error or proportional to some form of the error. Depending upon which form of the error is used as the input to the controller to produce the continuous controller output, these controllers are classified as, 1. Proportional control mode 2. Integral control mode 3. Derivative control mode Let us discuss these contral modes in detail.Modern Control Theory 6-5 Controllers 6.5 Proportional Control Mode In this control mode, the output of the controller is simple proportional to the error e(t). The relation between the error e(t) and the controller output p is determined by constant called proportional gain constant denoted as K,,. The output of the controller is a linear function of the error e(t). Thus each value of the error has a unique value of the controller output. The range of the error which covers 0% to 100% controller output is called proportional band, Now though there exists linear relation between controller output and the error, for a zero error the controller output should not be zero, otherwise the process will come to halt. Hence there exists some controller output p, for the zero error. Hence mathematically the proportional control mode is expressed as, pit) = K,et) +p, A) Where K, = Proportional gain constant Po = Controller output with zero error ‘The direct and reverse action is possible in the proportional control mode. ‘The error may be positive or negative because error is r-b and b can be less or greater than reference setpoint r. If the controlled variable i.e. input to the controller increases, causing increase in the controller output, the action is calied direct action. For example the output valve is to be controlled to maintain the liquid level in a tank. So if the level increases, the valve should be opened more to maintain the level. If the controlled variable decreases, causing increase in the controller output or increase in the controlled variable, causing decrease in the controller output, the action is called reverse action. For example simple heater control for maintaining temperature. If the temperature increases, the drive to the heater must be decreased. So if e(t) is negative then K,,e(l) gets subtracted from p, and e(t) is positive, then Kpe(t) gets added to p,, this is reverse action. So equation (1) represents the reverse action. But using negative sign to the correction term K,e(t), the direct action proportional controller can be achieved. The proportional controller depends on the proper design of the gain K,,. For fixed pg, if gain Kp is high then large output results for small error but narrow error band can be handled. Beyond these limits of the error, output will be saturated. If the gain is small then the response is smaller but large error band can be handled. This is shown in the Fig. 6.2.Modern Control Theory 6-6 Controllers Controller output High gain Small gain Error % Narrow 1 error band ' 1 ' ‘Wider error band Fig. 6.2 The proportional band is mathematically defined by, 6.5.1 Characteristic of Proportional Mode The various characteristics of the proportional mode are, 1. When the error is zero, the controller output is constant equal to py. 2. If the error occurs, then for every 1 % of error the correction of K,, % is achieved. If error is positive, K,,% correction gets added to p,, and if error is negative, K,,% correction gets subtracted from p,. 3. The band of error exists for which the output of the controller is between 0 % to 100 % without saturation. 4. The gain K,, and the error band PB are inversely proportional to each other. 6.5.2 Offset The major disadvantage of the proportional control mode is that it produces an offset error in the output. When the load changes, the output deviates from the setpoint. Such a deviation is called offset error or steady state error. Such an offset error is shown in the Fig. 6.3. The offset error depends upon the reaction rate of the controller. Slow reaction rate produces small offset error while fast reaction rate produces large offset error.Modern Control Theory 6-7 Controllers Output Setpoint Offset error D Timet Fig. 6.3 Offset error in porportional made The dead time or transfer lag present in the system further worsens the result. It produces not only the large offset at the output but the time required to achieve steady state is also large, The offset error can be minimized by the large proportional gain K,, which reduces the proportional band. If K,, is made very large, the proportional band becomes so small that it acts as an ON/OFF controller producing oscillations about the setpoint instead of an offset error. 6.5.3 Applications The proportional controller can be suitable where, 1. Manual reset of the operating point is possible. 2. Load changes are small. 3. The dead time exists in the system is small. 6.6 Integral Control Mode In the proportional control made, error reduces but can not go to zero. It finally produces an offset error. It can not adapt with the changing load conditions. To avoid this, another control mode is oftenly used in the control systems which is based on the history of the errors. This mode is called integral mode or reset action controller. In such a controller, the value of the controller output p(t) is changed at a rate which is proportional to the actuating error signal e(t). Mathematically it is expressed as, Where K, = Constant relating error and rate The constant K, is also called integral constant. Integrating the above equation, actual controller output at any time t can be obtained as,Modern Control Theory 6-8 Controllers pit) = K, o- (2) Where p(0) = Controller output when integral action starts ie. att = 0. The output signal from the controller, at any instant is the arca under the actuating error signal curve up to that instant. If the value of the error is doubled, the value of p(t) varies twice as fast ic. rate of the controller output change also doubles. If the error is zero, the controller output is not changed. The contro! signal p(t) can have nonzero value when the error signal e(t) is zero. This is because the output depends on the history of the errer and not on the instantaneous value of the error. This is shown in the Fig. 6.4. ett) alt) (a) Error (b) Controller output Fig. 6.4 Integral mode The scale factor or constant K, expresses the scaling between error and the controller output. Thus a large value of K, means that a small error produces a large rate of change of p(t) and viceversa. This is shown in the Fig. 6.5. If there is positive error, the controller output begins to ramp up. ErrorModern Control Theory 6-9 Controllers 6.6.1 Step Response of Integral Mode eit) The step response of the integral control mode is shawn in the A Fig. 6.6. The integration time constant is the time taken for the output to change by an amount equal to the input error step. This is shown in the Fig. 6.6. It can be scen that when error is positive, the output p(t) ramps up. For zero error, there is no change in the output. And when error is _ negative, the output p(t) ramps qT down. Fig. 6.6 Step response When error His constant 6.6.2 Characteristics of Integral Mode The integrating controller is relatively slow controller. It changes its output at a rate which is dependent on the integrating time constant, until the error signal is cancelled. Compared to the proportional control, the integral control requires time to build up an appreciable output. However it continues to act till the error signal disappears. This corrects the problem of the offset error in the proportional controller. For example, let us assume that the integral controller is used to control the armature current of a d.c. motor and to keep its value constant at 500 A. As long as the armature current is less than 500 A, the armature voltage, controlled by the controller, will increase. Thus the output of the controller will increase and will continue to do so till the error becomes zero i.e. armature current becomes 500 A. Then the controller output will remain at that value reached. This is possible because the output of the controller can remain at any value within its range, if the input is zero. The controller must not be overdriven as it will not then be effective. Thus far an integral mode, 1. If error is zero, the output remains at a fixed value equal to what is was, when the error became + 2. If the error is not zero, then the output begins to increase or decrease, at a rate K,% per second for every + 1% of error. In some cases, the inverse of K, called integral time is specified, denoted as T,.Modern Control Theory 6-10 Controliers It is expressed in minutes instead of seconds. 6.6.3 Applications The comparison of proportional and integral mode behaviour at the time of occurrence of an error signal is tabulated below, Acts immediately. Offset error always Actian according to K,, present. Larger the K,, smaller the error, Acts slowly. It is the Error signal always time intagral of the becomes zero. error signal. Table 6.1 It can be seen that proportional mode is more favourable at the start while the integral is better for steady state response. In pure integral mode, error can oscillate about zero and can be cyclic. Hence in practice integral mode is never used alone but combined with the proportional mode, to enjoy the advantages of both the modes. 6.7 Derivative Control Mode In practice the error is function of time and at a particular instant it can be zero. But it may not remain zero forever after that instant. Hence some action is required corresponding to the rate at which the error is changing. Such a controller is called derivative coniroller. In this mode, the output of the controller depends on the time rate of change of the actual errors. Hence it is also called rate action mode or anticipatory action mode. The mathematical equation for the mode is, Where Ky = Derivative gain constant. The derivative gain constant indicates by how much % the controller output must change for every % per sec rate of change of the error. Generally Ky is expressed in minutes. The important feature of this type of control mode is that for a given rate of change of errer signal, there is a unique value of the controller output.Modern Control Theory 6-11 Controllers The advantage of the derivative control action is that it responds to the rate of change of error and can produce the significant correction before the magnitude of the actuating error becomes too large. Derivative control thus anticipates the actuating error, initiates an early corrective action and tends to increase stability of the system improving the transient response. 6.7.1 Characteristics of Derivative Control Mode The Fig. 6.7 shows how derivative mode changes the controller cutput for the various rates of change of the error. eft) pt%) 100 % ' 1 1 4 ' ' ' ' ‘ ' ' t ' ' ' ' ‘ ' ' ' Controller output 50 % Time t Fig. 6.7 The controller output is 50% for the zero error. When error starts increasing, the controller output suddenly jumps to the higher value. It further jumps to a higher value for higher rate of increase of error, Then error becomes constant, the output returns to 50%. When error is decreasing i.e. having negative slope, controller output decreases suddenly to a lower value,Modern Control Theory 6-12 Controllers The various characteristics of the derivative mode are, . For a given rate of change of error signal, there is a unique value of the controller output. . When the error is zero, the controller output is zero. . When the error is constant ic. rate of change of error is zero, the controller| output is zero. When error is changing, the controller output changes by K, % for even 1 % per second rate of change of error. 6.7.2 Applications When the error is zero or a constant, the derivative controller output is zero. Hence it is never used alone. Its gain should be small because faster rate of change of error can. cause very large sudden change of controller output. This may lead to the instability of the system. 6.8 Composite Control Modes As mentioned earlier, due to offset error proportional mode is not used alone. Similarly integral and derivative modes are also not used individually in practice. Thus to take the advantages of various modes together, the composite control modes are used. The various composite control modes are, 1. Proportional + Integral mode (P1) 3. Proportional + Derivative mode (PD) 3. Proportional + Integral + Derivative mode (PID) Let us see the characteristics of these three modes. 6.9 Proportional + Integral Mode (PI Control Mode) This is a composite control mode obtained by combining the proportional mode and the integral made. The mathematical expression for such a composite control is, ' pt) = K, e()+K,Kjf e(tidt+p(o) o Where p(0) = Initial value of the output at t= 0Modern Control Theory 6-13 Controllers. The important advantage of this control is that one to one correspondence of proportional mode is available while the offset gets climinated due to integral mode, the integral part of such a composite control provides a reset of the zero error output after a load change occurs. Consider the load change occurring at t=t, and due to which error varies as shown in the Fig. 6.8. The controller output changes suddenly by amount V,, due to the proportional action. After that the controller output changes linearly with respect to time at a rate K, / Tj The reset rate is defined as the reciprocal of 7}. elt) Error 0 7 Time " ot ‘Controller 1 output t ' waren Tt TITTTTITTT __ Rise in autput due to integral : it i —————"_ action between times 1, and ty A Immediate rise due to proportional action 0 Time Integral acton time Fig. 6.8 Behaviour of P! controller The response shown in the Fig. 6.8 is for the direct action of the controller. The response of composite PI control mode for the reverse action is shown in the Fig. 6.9. In the reverse action, the proportional part is the image of the error. The sum of proportional plus integral action finally leaves the error to zero.Modern Control Theory 6-15 Controllers. 6.9.2 Applications The composite PI mode completely removes the offset problems of proportional mode. Such a mode can be used in the systems with the frequent or large load changes. But the Process must have relatively slow changes in the load, to prevent the oscillations. 6.10 Proportional + Derivative Mode (PD Control Mode) The series combination of proportional and derivative control modes gives proportional plus derivative control mode. The mathematical expression for the PD composite control is, de(t) pi) = Kye(t)+K Ky pi) The behaviour of such a PD control to a ramp type of the input is shown in the Fig. 6.10. Error e(l) Controller output Rise in signal due o proportional action between times ty and ty Immediate rise due to derivative action Time Fig. 6.10 Behaviour of PD controller The ramp function of error occurs at t=t,. The derivative mode causes a step Vy at ty and proportional mode causes a rise of V,, equal to Vy at ty. This is for direct action PD control.Modern Control Theory 6-19 Controllers " t 0.02(3t+ 7] + 0.02x 0. os f (3t+7)dt+0.5 0 at? ' josysorsesao-4| 9 67] +05 oO a ' [0.06% 1.5]+ [0,14]+ aero] F415)? +7x1 0.09 + 0.14 + 0.0111 + 0.5 = 0.7411 ie. p(t) = 74.11 % after t = 1.5 minutes. => Example 6.3: Suppose the error showit in the Fig. 6.14 is applied to a PD controller with K, = Gand K, = 04 see with p(0} = 25 %. Draw the graph of the controller output. ett) (%) oO 2 4 7 Fig. 6.14 Solution : The output of the PD controller is d pit) = Ky e()+K,Ky SA 25% it is necessary to obtain output response, dividing e(t) into three sections, Section I] ost s2 e(t) = 2t Slope = 2 de(t) a =? py) = 6% 2t+6x04x24+25 = 12+ 29.8 Thus there is an instantaneous change of 29.8 - 25 = 4.8 % produced by this error in the output at t = 0.Modern Control Theory 6-23 Controllers ey = te = 0 for the step input So there is a finite error for the ramp input ef magnitude A, which depends on & and For good time response, the system demands. 1. Less settling time 2. Less overshoot 3. Less rise time 4. Smalles Increasing K,, the steady state error for the ramp input can be reduced but it increases overshoot and settling time. This may lead to instability of the system. So to keep steady state error and overshoot well within the acceptable limits, the various composite controllers are used. Let us see the effect of PD, PI, PID and rate feedback controller on the time response of the second order system under consideration. steady state error. 6.14 PD Type of Controller A controller in the forward path, which changes the controller output corresponding to proportional plus derivative of error signal is called PD controller. ie. Output of controller = K eft) Taking Laplace = K E(s) + sTy E(s) = Efs)[ K + sTy] The T.F. of such controller is [K + s9,]. This can be realized as shown in the Fig, 6.17. Fig. 6.17Modern Control Theory 6-24 Controllers Assuming K = 1, we can write, _ (L+sTy) ae SO) = Ser2E0,) Cis) (1+ s Ty) a, and RG) + s2Ea, + af Ty]+ oF Comparing denominator with standard form, w, is same as in the previous P type controller. T, Because of this controller, damping ratio increases by factor “84. Lim Ky = 5p GUIHG) = = Lim ®, 26 Ky = jong 8 SUNHIS) = As there is no change in coefficients, error also will remain same. Key Point: Hence PD controller has following effects on system. i) It increases damping ratio. ii) “o,,' for system remains unchanged. iii) “TYPE' of the system remains unchanged. iv) It reduces peak overshoot. v) It reduces settling time. vi) Steady state error remains unchanged. Key Point: In general P.D. controller improves transient part without affecting steady state. 6.15 Pl Type of Controller A controller in the forward path, which changes the controller output corresponding to the proportional plus integral of the error signal is called PI controller.Modern Control Theory 6-25 Controllers je. | Output of controller = K e(t) + K;, fe(t) dt Taking Laplace = K E(s) + & E(s) = E(s) |x« | 2. The TF. of such controller is [+ 4 and can be realised as shown in the Fig. 6.18. Fig. 6.18 Assuming K = 1, we can write, Sts} = s(s+ 26.0,) (K, +3) a2 s? {s+2&a,) i.e. system becomes TYPE 2 in nature. ls) _ (K, +s) a} RG s3+2E@,s7 +s a2 + K, a2 ie. it becomes third order. and Now as order increases by one, system relatively becomes less stable as K; must be designed in such a way that system will remain in stable condition, Second order system is always stable. Key Point: Hence transient response gets affected badly if contraller is not designed properly. . Lim While =, = = Sp SIH) ==, ey = 0 Lim = 530% SHO =, e, =0 Key Point ; Hence as type is increased by one, error becomes zero for ramn type of inputs ie. steady state of system gets improved and system becomes more accurate in nature.Modern Controi Theory 6-26 Controllers Hence P| controller has following effects : i) I increases order of the system. ii) It increases TYPE of the system. iii) Design of K, must be proper to maintain stability of system. So it makes system relatively less stable. iv) Steady state error reduces tremendously for same type of inputs. Key Point: in general this controller improves steady state part affecting the transient part. 6.16 PID Type of Controller As PD improves transient and PI improves steady state, combination of two may be used to improve overall time response of the system. This can be realized as shown in the Fig. 6.19{a). Fig. 6.19(a) ‘The design of such controller is complicated in practice. 6.17 Rate Feedback Controller (Output Derivative Controller) This is achieved by feeding back the derivative of output signal internally using a tachogenerator and comparing with signal proportional to crror as shown, This is called minor loop feedback compensation. Output of controller = KE(t)- K, Output of the controller = K E{s) — sK, C{s) . taking Laplace This can be realized as es) KE(s) shown in the Fig, 6.19(b). Assuming K = 1, let us study its effect on same system which is considered earlier w with Gis) = =—s}---; ) = ee 2E I Fig. 6.19 (b)Modern Control Theary 6-30 Controllers Key Point: 1 Time constant “T’ is the time required by the system output to reach 63.2 % of its final value during the first attempt. The equation for the actual response c(t) is, in wa] .. For unit step Where = @, y1-8? Damped frequency of oscillations and 5 | radians. Note : As M, is the function of § alone, if M, is known, the equation of M, can be solved to obtain the corresponding value of damping ratio &. 6.19 Steady State Error Consider a simple closed loop system using negative feedback as shown in the Fig. 6.21. Ris) + E(s) Cs) Fig. 6.24 Where E(s) = Error signal, and B(s) = Feedback signal Now, F(s) = R(s) - Bis) But Bs) = Cs) H(s) E(s) = Ris) - C(s)H(s) and C(s) = E{s) - Gls) E(s) = R{s) — E(s)G(s)H(s) E(s) + E(s)G(s)H(s) = R(s)Modern Control Theory 6-31 Controllers = —*) sor nonunity feedback T+ Cis)His) Ris) = Item for unity feedback This E(s) is the error in Laplace domain and is expression in ‘s'. We want to calculate the error value. In time domain, corresponding error will be e(t). Now steady state of the system is that state which remains as t + . Steady state error, e,, = Lim e(t) tae Now we can relate this in Laplace domain by using final value theorem which states that, iim Fit) = Lim s F(s) Where F(s) = L { f(t) } Therefore, Lim s E{s) Where E(s) is L { e(t)] Substituting E(s) from the expression derived, we can write = tm — 386) ne TEI For negative feedback systems use positive sign in denominator while use negative sign in denominator if system uses positive feedback. From the above expression it can be concluded that steady state error depends on, i) R(s) Le. reference input, its type and magnitude. ii) G{s)H(s) i.e. open loop transfer function. iii) Dominant nonlinearities present if any. Examples with Solutions im Example 6.4: The figure shows PD controller used for the system. Determine the value of ‘Ty 80 that system will be critically damped. Calculate its settling time. R{s) cs) s@+16)Modern Contra! Theory 6-32 Controllers 1+sTy)4 Solution ; G(s} = aay (+sTy)4 s(s + 1.6) (l+sTya 1 EST Pe iese at, 644 s@=1.6) H(s) = 1 Comparing denominator with standard form, ow = 4, o,=2 and %o, =16+4Ty 2. LO+aTy ° a Now system required is critically damped, ie. § = 1 _ 16+4Ty = yt 4 = 16+4Ty eas 4 Ty = 06 and settling time = -— 4 6 Fo, = be To = tz17 2 sec. ‘um Example 6.5: A unity feedback system is shown in the following figure. i) In the absence of derivative feedback controller (K, = 0). Find § and t,' ii) Find K,, ifE is to be modified to 0.5 by use of controller. ion: = 1 = Solution : G(s) = 50x a3 for K, = 0 His) = 1 Cts) G(s) 50 Ris) «1+ G(S)H(s) 52 4 35450 wo = 50 +a, = 50 = 7.071 rad/sec.Modern Control Theory *“Controliers oO, % M. T, Ww V14 = 3.7416 rad/sec. MK, +16 HK. 1? 05 given 2x vid 0.1529 ©, {1-8 = 37416 f1-(0.5)? = 3.2403 rad/sec ©. = 0.9695 sec. OW et Vi-’ 4 5a, * 100 = 16.3 % = 2.1381 sec. jam Example 6.7: For the system shown determine % M,, and T, when it is excited by unit step input. If for the same system, PD controller having constant T, = 1/30 is used in forward path, determine new values of damping ratio, M, andT,. Draw respective wavefornis. Solution : Without controller, Gis) = Cs) Ro) 2 @ 28 Wy 100 serry HORT 5 100 s* + 12s +100 100». @,= 10 rad/see. 12 * §= 06 w, yl-€? = 10x 0.8 = 8 rad/sec. —-mnet _g2 oN oar y, 4 = g—— = 0.666 sec, $a,Modern Control Theory Controllers With controller Ris} os) - + 30)x 3.33 _ 60 = “saris savy 7 MO eT (s+ 30) 333 Ce) _ see 3.33(s + 30) = 33318 +30) Rs) |, +30) 3.53 92 41284 3.338 + 100 s* + 15.338 + 100 3(8+ 12) a, = 100 + @,= 10 rad/sec. w@, yi-=? = 10 (0.7665)? = 6.4224 rad/sce. elie? «100 = 2.353 % 2. Gis improved,ay = % My, = Overshoot decreased to 2.3 % from 9.47 %. 4 T, = gay = 05218 sec Comparison : Following figure shows comparison between system with controller and system without controller. 0.47 % oft) time time 6.5218 sec Without controtier With controllerModem Control Theory 6-37 Controllers @, = +10 rad/sec and 2Em, = 2 & = 0.3162 This is the required damping ratio. For error calculation : GSH) = y For ramp input magnitude is unity ie. A=1 . 8.5 im Sits 65) 7% rm 1 lim s G(s)His}= s40 t Case_b) The derivative feedback is introduced in the system. The system becomes, K A and H(s)=1 +1+K, | Cts) _ 8 WSst te Ry _ Re), A 05s? +(1+K,) s+ Ky 3[0.5s+ 14K, ]Modern Control Theory 6-40 My OO), But K, and 0.4037 Kp n= a, 1-2 6.8676 rad/sec @2 = 47.165 B+ Kp 2x a, 25449 wma Example 6.10: An integral controller is used for temperature control within a range 40 — 60°C. The set point is 48°C. The controller output is initially 12% when error is zera. The integral constant K, = — 0.2% controller output per second per percentage errar. If the temperature increases to 54°C, calculate the controller output after 2 sec for a constant error. Solution : For integral controller, ptt) pd) Now r J Edt pi) Alt = 2, pi) Ky J Epdt+p(0) 12%, Ky =-0.2% / sec / % error r-b Constant = 100 Prax —Pmin Set point b = Actual temperature = 54°C 60°C, Bain 48-54 40-40 ~ 100 =-3% Ey { as error is constant {- 0.2) © 30) () + 12 (0.2 x 30 x 2)+12=24% ‘This is the controller output after 2 sec. hm Example 6.11 : For a proportional controller, the controlled variable is a temperature with a range of 50 to 130°C with a set point of 73.5 °C. The controller output is 50% for zero error. The offset error is corresponding to a load change which causes 55% controller output, If the proportional gain is 2, find the % controller output if the temperaiure is 61°C. Solution : For a proportional controller, ptt) Where E,W Kp E,(0 * po Error, Kp = 2, pg = 50%Modern Control Theory 6-43 Controllers After end of 2 sec, integral term gets accumulated to, 4 PQ) = 2x22) 3" +40 = 48.8% 12" haa For2——*100 Where r = set point Bina Prnin 25 = b = 715°C This is temperature corresponding to offset error.Modern Control Theory 6-45 Controllers imap Example 6.15: A PID controller has Kp = 2.0. Ky = 2.2 sec”!, Kp = 2 see and PAO) = 40%. Draw the plot of controller output for error of Fig. 6.24. Error E,% Fig. 6.24 Solution : Kp = 2, K,=22sec', Kp =2 sec, P)(0) = 40 % From the given error plot, For 0-2sec, Ep = mjt Where my = Slope = E, = t for 0-2 sec. For2-4sec, E, = m,t+C) Two points on the line are (2, 2) and (4, -3) AL (2, 2), 2 = -25K 240, 1G s7 ee = -25t+7 for 2 —4 sec. For4-6sec, EF, = m3t+C, Two points on the line are (4, - 3) and (6, 0). Q-(-3) “ mz = op st 15 E, = +15t+C, At (4, -3), -3 = 15%4+C, 23=-9 ” E, = +15t-9 for4—6 sec.Modern Control Theory 6-46 Controllers Three mode equation for PID controller is, ee fre "4 dE, p= KpE, +KpK; J Epdt+K pK, + P(0) a a " t dE, 2E, st] B,dt+4—P +40 t For0-2sec, py, = asdf t dread (reso o 242.20? 444 -- () ' This is plotted for 0 - 2 sec. At the end of 2 sec, the integral term has accumulated to, t 27? PQ) = 4a] tdt+40 = sf] +40 = 48.8 % 0 0 ' For 2-4 sec, py = 2-2.547)+44f(2.5te7)dtr4 4 (2.5147) 4488 3 asta eda [1250 +n], 4x 2.54488 = ~5t+ 44 4af125(t? ~4)+7(1-2)}-104-488 = -5.5t7 +25.8t4+132 we (2) This is plotted for 2 - 4 sec. At the end of 2 sec, the integral term has accumulated to, W (4) aaf-125(1? —4)47(1- 2H, #488 = 444 % For 4 - 6 sec, ‘ Ps = 2(1.5t-9)+4.4] (1.5t-9)dt+4 fa 5t-9) +444 4 = 3t-18+44[075t? ~9] +4x154444 = 3t-18+ 444075 (1? -16)-9(-4)} +64 444 3.3t? —36.6t+138 2 BD This is plotted for 4 — 6 sec.Modern Control Theory 6-47 Controllers After 6 sec, error is zero hence the output will simply be the accumulated integral response providing a constant output. P\6) = safors(t? -16) -9(t—ayh +44 too = 31.2% The complete graph of controller output is shown in the figure. f i / Controtier 4 | 7 Scale | Gn x-axis ef units 1 sec Lony-axis —etunit=5% ebsites mp Example 6.16: A temperature control system has the block diagram given in Fig. 6.25. The input signal is a voltage and represents the desired temperature 0, is a unit step function and i) D(s)= 1 ii) Ds) = 1 + gi iii) D(s) = 1 + 0.3 s. What is the effect of the integral term in the PI controller and the derivative ferm in PD controlier on the steady state error 7 (VTU: July/Aug,-2006, 8 Marks) Fig, 6.25Modern Control Theory 6-48 Solution : For the given system, 200 Gls) = ins+2)' H(s) = 0.02, Rts) = be eneey e sR(s) ey = Lim sis) = Lim Toe aS i Dis) = 1 ey = Lim agg = sao - tame” ii) D(s) = wa 002 5x 202 ey = Lim — su ed Fresnires ya itt) Ds) = 17035 «0.02 = rapa 0.3) oy 7 14] —~—— (aes } — = 6.66x107> Due to PI controller, the steady state error reduces drastically while PD controller has no effect on the steady state error. Review Questions What is controller ? Explain its function i a system. Give the classification of controllers. |. State and explain the various properties of controller. aw N |. State the various continirous controller modes. Explain the proportional control mode. State its characteristics . Explain the integral control mode. Stale its characteristics. . Explain the derivative control mode. State its characteristics. . Explain how constant K, affect the output of Pt control mode. . Why derivative mode is called anticipatory control mode ? 10. State the various comiposite contro! modes. we aN AModem Control Theory 6-49 Controllers . Explain the PD contro! mode, stating its characteristics. . Write a note on three mode controller. . Discuss the effect of following controtiers on the second order control system, . Explain the effect of PI controller on typical second order system. . Explain the effect of PD controller on typical second order system. ‘Guly-2005) Explain the PI control mode, stating its characteristics. (July-2005) a) PI controller b) PD controller A feedback system which uses a rate feedback controller is shown in the figure. a) For Ky =10 in absence of derivative feedback (Kj, =0) determine the damping ratio and nalural frequency of oscillations. Alsa find the 3.5 error for unit ramp inj. b) Determine the constant K,, if the damping factor required is 0.6, with K, =10. With this value of Ky, determine 3. error for unit ramp input. (Ans, ; 0.316, 3.16 rad/sec, 0,2, 1.8, 0,38) ona(6 - 50)Nonlinear Systems 7.1 Introduction to Nonlinear Systems It has been mentioned earlier that a control system is said to be linear if it obeys law of superposition. Most of the control systems are nonlinear in nature and are treated to be linear, under certain approximations, from ease of analysis point of view. Let us discuss now the properties of nonlinear systems. In practice nonlinearities may exist in the systems inherently or may be purposely introduced in the systems, to improve the performance. Hence knowledge of properties of nonlinear systems and various nonlinearities is important. 7.2 Properties of Nonlinear Systems The various characteristics of nonlinear systems are, 1. The most important characteristics of a nonlinear system is that it does nat obey the law of superposition. Hence its behaviour with respect to standard test inputs can not be used as base to analyse its behaviour with respect to other inputs. Its response is different for different amplitudes of input signals. Hence while doing the analysis of nonlinear system, alongwith the mathematical model of the system, it is necessary to have information about amplitudes of the probable inputs, initial conditions etc. This makes the analysis of the nonlinear system difficult. 2. Linear system gives sinusoidal output for a sinusoidal input, may be introducing a phase shift. But nonlinear system produces higher harmonics and sometimes the subharmonics. Hence for sinusoidal input, the output of a nonlinear system is generally nonsinusoidal. The output consists of frequencies which are multiples of the input frequency i.e. harmonics. The subharmonics means the presence of frequencies which are lower than the input frequency. The input and output relations are not linear. 3. In linear systems, the sinusoidal oscillations depend on the input amplitude and the initial conditions. But in a nonlinear system, the periodic oscillations may exist (7-1)Modern Control Theory 7-2 Nonlinear Systems. which are not dependent on the applied input and other system parameter variations. In nonlinear system, such periodic oscillations are nonsinusoidal having fixed amplitude and frequency. Such oscillations are called limit cycles in case of nonlinear system. 4, Another important phenomenon which exists only in case of nonlinear system is jump resonance. This can be explained by considering a frequency response. The Fig. 7.1 (a) shows the frequency response of a linear system which shows that output varies continuously as the frequency changes. Similarly though frequency is increased or decreased, the output travels along the same curve again and again. But in case of a nonlinear system, if frequency is increased, the output shows discontinuity i.e. it jumps at a certain frequency. And if frequency is decreased, it jumps back but at different frequency. This is shown in the Fig. 7.1 (b). Output Output Peak resonant oy 2 {a) Linear system (b) Nonlinear system Fig. 7.1 Jump resonance 5. There is no definite criterion for judging the stability of the nonlinear system. The analysis and design techniques of linear systems cannot be applied to the nonlinear system. 7.3 Classification of Nonlinearities Basically the nonlinearities are classified based on the fact that whether they are inherent or purposely introduced in the system. Depending upon this the two classes of nonlinearities are, 1, Inherent nonlinearities. 2 Intentional nonlinearities, _t4 Inherent Nonlinearities The nonlinearities which are unavoidable in the control systems are called inherent nonlinearities. Let us discuss various types of inherent nonlinearities existing in practice.Modern Control Theary Nonlinear Systems. Output Output Input Input (a) On-off with deadzone (b) On-off with deadzone and hysteresis Fig. 7.7 7.4.5 Nonlinear Friction In any practical system, where there is a relative motion between the two moving surfaces, there exists a friction. There are various types of such friction. All of them are nonlinear ‘Viscous except viseous friction. The nonlinear friction is called Coulomb friction. ‘Output The coulomb friction is a force acting in 'naut— gpposite direction of motion but it is constant in magnitude irrespective of velocity. The viscous and coulomb friction are shown in the Fig. 7.8. It can be seen that viscous Coutom friction is linear while coulomb friction is constant irrespective of the input. Fig. 7.8 Friction The example of coulomb friction is the friction existing between the brushes resting against the commutator in an electrical motor, 7.4.6 Backlash This type of nonlinearity is generally found in the mechanical linkage where coupling is not perfect. A common example is the existence of free play between tecth of drive gear and driven gear of a gear train. This is shown in the Fig, 7.9 (a). The driven gear does not work until there is contact between drive and driven gear. So input member has to travel the distance B to have the output, as indicated by input-output characteristics shown in the Fig. 7.9 (6). Once the contact is made output follows the input. If the motion is to be reversed, then input member has to travel distance of 2B in reverse direction, as shown in the Fig. 7.9 (b).Modern Control Theary 7-6 Nonlinear Systems Drive gear 9, Driven gear O, (a) Fig. 7.9 Backlash This is the most complex nonlinearity from modelling point of view. This is also memory dependent nonlinearity. It is a double valued nonlinearity like on-off with hysteresis, A process control valve with hysteresis is another example of backlash type of nonlinearity. 7.4.7 Nonlinear Spring A linear spring has a linear equation for its force and displacement. Linear spring force = kx where k = positive spring constant and x i spring displacement The nonlinear spring is not governed by the equation (1) but its force and displacement are related by the equation, Nonlinear spring force = kx + kx? -. Q) where k = Positive spring constant and k’ = Nonlinear spring constant The k’ can be positive or negative. If k’ is positive, the spring is called hard spring while if k’ is negative, the spring is called soft spring, If k’ = 0, the spring is no longer a nonlinear but behaves in a linear fashion. The degree of nonlinearity in the spring is totally controlled by the magnitude of k’. If such a spring is used in the system consisting of mass and friction then it is experimentally observed that the frequency of the free oscillations increases or decreases as the amplitude decreases depending upon the value of k’.Modern Control Theory 7-7 Nonlinear Systems. Frequency KO Hard spring Linear ke Soft spring ‘Amplitude Fig. 7.10 Frequency-amplitude curves for roplnear spring / The frequency-amplitude curves for a system with nonlinear spring are shown in the Fig. 7.10. For k’ > 0, as the amplitude decreases, frequency also decreases. This is hard spring. For k’ < 0, as amplitude decreases, the frequency increases. This is soft spring. For a linear spring such characteristics is a straight line. These curves are often used to find out whether there exists a nonlinearity in the spring and also to find the degree of nonlinearity. 7.5 Absolute Value Nonlinearity In some systems, though the direction of the input is changed, the output direction remains same. The value of output is absolute irrespective of the direction of the input applied. The best example of such a nonlinearity is a relay operated by the solenoid. Irrespective of the direction of the current through the solenoid, relay gets operated. This is shown in the Fig, 7.11 (a) and (b). ‘Output Supply Solenoid ~ — —~., Input Relay (a) (b) Fig. 7.11 Absolute value nonlinearity Thus the output is not affected by the direction of the input applied.Phase Plane Method 8.1 Introduction The phase plane method is an important method used for the analysis of nonlinear systems. The phase plane method is applied normally to first order and second order of linear and nonlinear systems. The system behaviour is qualitatively analysed along with design of system parameters so as to get the desired response from the system. The periodic oscillations in nonlinear systems called limit cycle can be identified with this method which helps in investigating the stability of the system. 8.2 Limit Cycle In case of linear time invariant systems a periodic oscillation is sinusoidal. The amplitude of such oscillations is a function of excitation applied and initial conditions. If the system parameters are changed slightly ie. if the system poles are shifted from the imaginary axis of complex s plane, then the oscillations will no longer exist. In case of nonlinear systems, periodic oscillations are present but are not dependent on amplitude of applicd excitation. Also the oscillations are not much sensitive to parameter variations. A periodic oscillation is case of a nonlinear system is called limit cycle. Normally the limit cyeles are non-sinusoidal. The limit cycles which are having fixed amplitude and time period can be observed over a finite range of system parameters. The limit cycle is also called self excited oscillation which is observed in certain nonlinear systems. Consider a system described by following equation, Mx-p(l = x2) x¢-kx = 0 Here M, 1! and k are positive quantities. This is a nonlinear equation. For small values of x, the damping will be negative which indicates the energy is actually put in the system. For large values of y, the damzing is positive which removes the energy from the system. Thus it can be expected that such system may exhibit a sustained oscillation. Now as it is not a forced system, these oscillations are called self excited oscillanons or limit cycle. (8-1)Modern Control Theory 8-2 Phase Plane Method 8.3 Jump Resonance The phenomenon of jump resonance is seen only in case of non linear systems. This can be explained by considering the frequency response plots. The Fig. 8.1 shows frequency response plot for a linear system while the Fig. 8.2 shows frequency response plot for nonlinear system. Response Fig. 8.1 Linear system Fig. 8.2 Nonlinear system Consider a nonlinear system excited by sinusoidal input of constant amplitude. If the input frequency is increased then jump (or discontinuity) occurs in the response amplitude. With decrease in frequency also, jump can be observed but at a different frequency. A nonlinear system is one which behaves quite differently for various input functions. Consequently the logical design of a nonlinear system requires a complete description of the input signals. This dependence of the system behaviour on the actual input functions is demonstrated by the jump resonance phenomenon which is observed in certain closed loop systems with saturation. Let us consider a nonlinear equation describing a certain system. Mxt fat kyxtkyx? = F cos ot If the magnitude of the response for above system is plotted against frequency, then corresponding frequency response plot for k, > is shown in the Fig. 8.3. Fork > Fig. 6.3 Jump resonanceModern Control Theory 8-3 Phase Plane Method Due to the presence of a nonlinear term k,x*, the resonant peak is bent towards the higher frequencies. As the input frequency is gradually increased from zero with input amplitude fixed, the measured response follows the curve through the points A, B and C. But at C, an increment in frequency results in a discontinuous jump down to the point D, after which with further increase in frequency, the response curve follows through DE. If the frequency is decreased, the response follows the curve EDF with jump to B occurring at F and then move to A. For certain range of frequency the response function is double valued. For k, <0, the resonant peak bends towards lower frequencies which is shown in the Fig. 8.4. Similar jump resonance takes place with the difference that there is upward jump when frequency is increased and downward jump when frequency is decreased. Response D For ky < 0 Fig. 6.4 Jump resonance The overall curve shows the familiar hysteresis or jump resonance phenomenon. It is well known in nonlinear mechanics and is frequently abserved in high gain servo systems. In such cases, the response is multivalued and depends not only on the present value of the input but also on the past history. 8.4 The Phase Plane Method The equation of a second order nonlinear system having free motion has the following form, ¥- hy, H¥+80. yy = 0 At any particular moment, the state of the system can be represented with the help of a point having coordinates (y, y) in rectangular coordinate system, This coordinate plane is called a 'phase plane’.Modern Control Theory 8-5 Phase Plane Method +g yay yyY Fig. 8.5 The starting point A with initial condition but no initial velocity, the system comes to equilibrium i.e. to the origin with damped oscillatory behaviour. 8.5 Basic Concept of Phase-Plane Method Consider the differential equation for a linear second order control system d*yit) dy(t) _ 42 +a—F + Bytt) = ult) Here u is input while y is output. The phase plane method is basically a graphical method of solving second order linear as well as non linear system equations. The co-ordinate plane whose axes correspond to dependent variable y(t) and its first derivative y(t) is called phase plane. In phase plane plot y(t) and yt) are plotted in a two axes plane. The trace of the phase plane plot as time t increases is called trajectory. The starting point of the trajectory depends on the initial conditions imposed on the system. For different scts of initial conditions, various trajectories can be plotted. The complete dynamic behaviour of the system for various initial conditions can be obtained from the trajectories plotted in phase plane. ‘The dynamic performance and analysis of given second order system differential equation can be made based on destination point of trajectories. The method has limitation. of its application restricted to second order as for higher order system equations, three or higher dimensional plane is needed which is not much convenient to construct and visualise. But transient response of second order system excited by step input can be studied by this method. This method is thus applicable to linear time functions and can not be applied to systems excited by sinusoidal signals. The phase plane method can be considered as a special case of state space. With the use of slate variables known as phase variables, the state of second order system can be defined.Modern Control Theory 8-7 Phase Plane Method In phase plane analysis, the trajectorics can be constructed analytically, graphically or experimentally. The analytical method is straight but not useful from practical point of view. The graphical methods are used when it is not possible to solve the given differential equation analytically. The graphical methods are applicable to both linear and non linear equations. 8.6 Isocline Method for Construction of Phase Trajectory The isocline method is a graphical approach for determination of phase portrait. Consider the state equation for a second order system i = AO x2)] t F(%1-%2)f ‘The slope m of trajectory at any point in the phase plane i given by, Ixy {y(X 2X9 OL me 2(%1-%2) ax) fy (Xp -%2) The paint (x; x9) of the phase plane has with it associated slope of the trajectory except at singular point where the slope is indetermiante. Isaclines are the lines in the phase plane corresponding to slopes of the phase portrait. In order to construct the phase trajectory, it is necessary to integrate the slope equation, This is very difficult except for few simple cases. We have, 4h [sess ): m,f,(%y. Xa) The above equation gives locus of all points in the phase plane where the slope of trajectory is m,. Such a locus is called isocline. The direction of phase trajectroy at any point (x1,x2) can be obtained from the sign of Ax, and Ax. for small increment in At. XX) Ay = Fyxp-Xq) At = fy(44.%2) widiky = f(xy, %) AE Consider once again second order system equation. yt) gn 32 sales By = ul) Subject to initial conditions y(t) = ygModern Control Theory 8-9 Phase Plane Method Based on various slopes of trajectories, isoclines are drawn as shown in the Fig. 8.7 Xx, 2 mem, Isocline 1 m= m, Isocline 2 NY c= tan” 'm, m=m + isociine 3 a= tan” Ms mem, Isocline 4 x Fig. 8.7 From the given initial conditions, the starting point is O which is the paint on isocline 1 corresponding to slope mj. The phase trajectory which is drawn from point on the isocline whose slope is m,. The phase trajectory which is drawn from point O on the isocline whose slope is m, to that having slope m, is a straight line whose slope is the average of m, and m, ie. me ‘This line intersects isocline 2 at point P which has a slope of m;, Thus we get line segment OP. This procedure is continued at point P to get subsequent line segments PQ, OR and so on as shown in the Fig. 8.7 Now consider undamped second order system such that = 0 d?y(t) dy(t) “ye Oa For simplicity let us select @, = 1, The above equation is ty +y) = 0 Let us now select the state variables x = ylt) x = yet) The state equations are, x == x ” xy = y()=-yit)=-%,Modern Control Theory B-10 Phase Plane Method Dividing above equations, xy = dxg/dt _ -xy dx, dx /dt x5 xgdxy = =x, dx, Integrating and rearranging above equation. 22. xp#xp = C Here C is arbitrary constant obtained from initial conditions. The phase portrait in phase plane is as shown in the Fig. 8.8. The above equation represents a circle. The phase portrait consists of concentric circles depending on different initial conditions ay? ya xf xfs ete) ( ty “Io “lo Initial condition x, 1 Fig. 6.8 8.7 Application of Phase-Plane Method to Linear Control System Before studying the application of phase-plane method to non-linear systems, let us see its application in linear systems. Let us consider second order linear control system as shown in the Fig 8.9 kK ss*t) Ris} Cs) Fig. 8.9Modern Control Theory 8-11 Phase Plane Method The transfer function is given as, —K_ K) coy the) kA) Rep ik s(Tjs+1)+K Steet s(Ts+1) In the forward path, the transfer function is cs) OK Eis) s(T,s+1) [s(Ts+1)] Gs) = KE(s) T,s"Cis)+sC{s) = KE(s) Taking inverse Laplace transform, = Ke But e = rec ee re e = rc 4. car-e ey] [fe Tj r-el+|r-e| = Ke J J TyeterKe = Let the input be step input rit) = R rt) = TQ) <0 The above equation becomes T,e+erKe = 0 fort>0 Now e (0) = r(0)-c (0) At t=0,c=0iee(@)=0 rQ=RModern Control Theory 8-13 Phase Plane Method When the value of K is adjusted in such a way as to get the system overdamped then the phase plane trajectory is as shown in the Fig. 8.11 att) aft) 7 é(t) (a) Step response (Overdamped case) {b) Corresponding phase portrait Fig. 8.11 For the linear systems, where analytical solution can be computed for ¢ and ¢ then it is not required to draw phase portrait. It can be effectively applied in case of non-linear systems. 8.8 Second Order Nonlinear System on Phase Plane Let us consider the Van der Pol's differential equation, ” . youd -y*)yry = 0 This equation describes the behaviour in many nonlinear systems. Let us select the state variables as x, = y and x) = y. The stale equations are given as, xy = Y =x, ts = 2 Xy SY SMC xq)R9 xy The equilibrium point of this system is origin of the phase plane. Now there are two cases viz. t > 0 and ft < 0. The phase portraits for both these cases is shown in the Fig. 8.12. When u > 0 and initial value of x, [i.c. x,(0)] is large then the system response is damped. The amplitude of x, (t)[=y;(t)] decreases till the system enters the limit cycle. This is shown by the outer trajectory. For small initial values of x,, the damping is negative.Modern Control Theory 8-14 Phase Piane Method <- Stable limit cycle A- Unstable Wenit cycie (ayn> 0 (op <0 Fig. 8.12 ‘The amplitude of x,(t)[=y ,(t)] increases till the system state enters the limit cycle which is shown by inner trajectory. ‘The limit cycle is stable as the paths in its neighbourhood are converging. The limit cycle for p <0 is unstable. With increase in order of the system, the complexity is involved in applying phase plane method. The method of phase trajectories is restricted normally to second order systems and it is special case of phase space defined for n'" order system. In case of time invariant systems, the total phase-plane contains trajectories with only one curve passing through every point of the plane except for some critical points. Through these critical points either none of the trajectories or infinite number of trajectories pass, For time varying systems, two or more trajectories may pass through a single point. The phase portrait then becomes complex and not easy to interpret. Hence the method of phase-plane is mostly applied to second order systems having constant parameters with no or constant input. Some simple time varying systems can be analysed by this method. 8.9 Different Types of Phase Portraits In this section, the phase portraits of some commonly known linear systems is presented. These portraits can be used for analysis of piecewise linear systems. 8.9.1 Phase Portraits for Type 0 System Consider a transfer function for a type 0 system. Ce) Cs RS” 7 yRasree Let us consider the input r(t) to be zero ic. unforced system ” [s? +2G0, +02 |C(s) = 0Modern Control Theory 8-15 Phase Plane Method C+2é0, C+e2C = 0 Let x, = Cand xz = C. Therefore the state model is given by, Ry ge Ry = 2a, ry wx, Eliminating the time variable from the two equations by dividing the equations, dx, 2G a,x, - dx, xy d If we consider the point at which x, = x, = 0 then the slope at this point a = 1 which is indeterminate. This point is called singular point. We will study the behaviour of trajectories which are closed to this singular point. To find out the singular points which are nothing but the roots of characteristic equation s* +2&@,5+0;, = (s—y,)(s~y,) = 0. Depending on the values of y, and 73, six different types of singular points are obtained. These are described as below. 8.9.1.1 Stable System with Complex Roots Let y+ -a+i8, ¥, =-a@-if} a>0and B>0 The output response in this case is c(t) = Cye™™ sin (Bt+C,) C, and C, are constants whose values can be obtained from initial conditions. The phase portrait can be obtained from the response equation on (x,,x3) plane as xy) = Gl X= C. The typical phase trafectory obtained is a logarithmic spiral into the singular point which is shown in the Fig. 8.13. This type of singular point is called a stable focus. { joaxis ip ™ © axis —— (a) Stable focus (b) Nature of roots in s plane Fig. 8.13Modern Control Theory 8-17 Phase Plane Method jo iB {a) Center (b) Nature of raots: Fig. 8.15 8.9.1.4 Stable System with Real Roots Let y, and y, be two real and distinct roots located in left half of s-plane. The response in this case is given by, cit) = Cyc"! +Cyet2t The response obtained from the system is overdamped. The singular point obtained after plotting the phase trajectory on (x, = y, Xz = ¥) plane for this case is called a stable node. The phase portrait contains two straight line trajectories having equations x5 (t)= 7X, (05 Xz (t) = 75%; (t) which satisfies the differential equation of the given system. Assuming that the term e¥2! decays faster in the transient response than the term e71t. As time t increases, x, Cel +0 and x, +7,Ce"l +0. All the trajectories are tangential at the origin tothe straight line trajectory »(t)= y,x,(t). The other trajectory is given for initial conditions x.(0) = 74x,(0)- All the trajectories are asymptotic to this straight line trajectory. If the roots are repeated for stable systems then there is a single trajectory whose slope depends on value of root.Modern Control Theory 8-18 Phase Plane Method vo) % + a oo Xp HX Xp = tah Fig. 8.16 Stable node Fig. 8.17 Nature of roots The phase trajectory and nature of roots is shown in the Fig. 8.16 and Fig. 8.17. 6.9.1.5 Unstable System with Positive Real Roots Let 7, and 7 be two real and distinct roots located in right half of s plane 7, is a smaller root. The singular point in this case is called unstable node. The phase portrait on (%; = ys xy = ¥) plane is shown in the Fig. 8.18. ja ” % oe Fig. 8.18 Unstable node Fig. 8.79 Nature of roots The trajectories emerge from the singular point and go to the infinity. The trajectories are tangential to the straight line trajectory x5(t) = ¥,Xx;,(t) at origin and asymptotic at infinity to other straight line trajectory. For the repeated roots, there is a single trajectory.Modern Control Theory 8-19 Phase Plane Method 8.9.1.6 Unstable System with One Positive and One Negative Real Root In this case we have two straight line trajectories which have slopes dependent on values of root. The singular point obtained in this case is called a saddle. The straightline obtained due to negative root gives trajectory entering into the singular point while due to positive part gives trajectory which leaves the singular point. The other trajectories approach towards the singular point close to incoming straight line then takes a curve away and leaves the singular point in its vicinity. It approaches the second straight line asymptotically. The phase portrait on (x; = y, %, = » plane is shown in the Fig. 8.20, jo Fig. 8.20 Saddle point Fig. 8.21 Nature of roots 8.9.2 Forced Second Order Type 0 System Let us consider a second order system excited by constant input with magnitude A. Cis) 7 +260,8+0%] = Aw C+2E@, C+ Cuz = Aw PC ae Slate = Ao? qr eon ae Feel = Aw The above differential equation can also be written in following form. SC-AN+280, Sic aysak(c- a) = The equation is similar to that obtained fer unforced system with the difference that the singular point is shifted from origin to point C = A on the Y axis. 8.9.3 Phase Portraits for Type 1 System ‘Consider the following transfer function which represents Type 1 system. This system is excited by constant input of magnitude A.Modern Control Theory 8-21 Phase Plane Method . The total equation becomes, Ext = A=x9(t)- AbA- x(lt tx {O)— A+ x9(0) + AlafA — x, (0)] 1 A-x zlxy- x} = “ley ~ sO ald 355 | The corresponding phase portrait for A > 0 is shown in the Fig. 8.22. Fig. 8.22 Phase portrait for A > 0 If initial conditions are zero i.e. x,(0) = x)(0) = 0 then A-x 2 2a] The corresponding phase trajectory is shown in the Fig. 8.22. The trajectory is asymptotic to the line x, = A. If the initial point is different (x,(0), x (0)] then the trajectory will have same shape but shifted horizontally by k units so as to pass through the point [x, (0), x (O)} The equation of trajectory in this case is given by, 1 A-Xx, oak) = aaah x )Modern Control Theory 8-22 Phase Plane Method The phase portrait for A < 0 is shown in the Fig. 8.23. The value of A = 0 is a special case which is unforced system and the phase portrait consists of number of straight lines having the slope of - L Fig. 8.23 Phase portrait for A < 0, Fig. 8.23 Phase portraits for A < 0 8.9.4 Phase Portraits for Type 2 System Consider the transfer function of linear system representing type 2 system ce) R(s) Js? Let the input applied be of constant magnitude, r(t) = A. The corresponding differential equation is given by, JO=a Let the state variable be x, = C, x, =C ‘The corresponding state model is given as, Leen x, = t-8Moder Control Theory 8-23 Phase Plane Method By eliminating the time variable we get, dx, A ay Ikgdxy = Ady, Integrating the above equation, J bgde, = J Adx, 2, GW) ' Ngan x4(t) " Where C is constant of integration obtained from initial conditions. 2 x) = 1a 4c 2 x0) = HO. : . 1330) * C= 1 MSH The nature of trajectory is parabolic for an initial state point [x,(0), x,(0)] passing through a point x, = C on the x-axis where C is given by above equation. This is shown in the Fig. 8.24 for A<0 and A>0. FY C<0 ayForA>o {b) ForA<0 Fig. 8.24Modem Contro! Theory 8-24 Phase Plane Method With increase in time t, each trajectory is shown in clockwise direction. The direction of phase trajectory is obtained from the relationship x, = xz. As x, increases with time in upper half of phase plane, the state point hence moves from left to right in the lower half of phase plane. When x, decreases with time, the state point moves from right to left. The time interval between the two points of a trajectory is given by At= Ax, /x>,,. With the aid of this equation the time scale can be given to the trajectories. But the phase portrait is generally used to get an idea about transients in the system. The phase portrait or A <0 is shown in the Fig. 8.24 (b). The unforced system with A = 0 is a special case. dx, Te = 0 4 dy =0 Integrating above equation J dx, = 0 x(t) = C Where C is constant of integration obtained from initial conditions. C = x0) «We have equation of trajectory as x(t) = x2(0). This equation represents the trajectories of straight lines parallel to x, axis. ‘mp Example 6.1: Determine the kind of singularity for each of the following differential equations. woe i) y+3yt+2y = 0 ii) y- 8 y+17y = 34 (VTU: Jan/Feb.-2005) Solution : oe i) y+3y+2y=0 Let the state variables be x, = y and x, = y The corresponding state model is . os sw Mp SY Exp Xp Sy S—3ey 2K Eliminating the time variable, dx _-3xy 2% dx, >Modern Control Theory 8-25 Phase Plane Method The characteristic equation is given by, s? +3542 = 0 (s+2)(s+1) 0 0 s = -2 s=-1 We have, y,; = -1 and 7, = -2 The nature of phase trajectory is as shown in the Fig, 8.25 and the type of singular point obtained is stable node. t m= a= Yak (0) tb) Fig. 8.25 The response in this case is given by y(t) = Cyet' +C,e72'. The transient term e!2* decays at faster rate than el', As time to, x, +Ce™ 30; x, = yell +0 The singular point node obtained in this case (0, 0) is stable in (y, » plane. ii) y-ayti7y = 34 Y-By+I7y-34 = 0 The above equation can be rewritten as, d?(y-2) dt? d = 8-2 +Ily-2) = 0Modern Control Theory 8-26 Phase Plane Method The characteristic equation is given by, (5-2)? - 8(s—2)+17(s—2) 0 (s? —4s+4)-85+16417s-34 = 0 s? 458-14 = 0 (s+7)(s-2) 0 4 = “7 =2 There are two roots one positive and one negative root. The singular point obtained in. this case is called saddle point which is shifted from origin by 2 units. The slopes of the two straight line trajectories with slopes dependent on root values. The nature of phase trajectory is shown in the Fig. 8.26. He? He2 o- (2,0) (— t Fig. 8.26 8.10 Singular Points of a Nonlinear System Singular point is a point on the phase plane where the slope of the trajectory is indeterminate. In the state model of a given system, when time variable is eliminated then we get, By equating the numerator and denominator equations to zero i.e. solutions of the equations f(x), x7) = 0 and x, = 0 gives the locations of the singular points along with its number. It is also important to know the behaviour of the trajectories in the vicinity of a singular point of a nonlinear system. The nonlinear equations are linearized at the singularModern Control Theory 8-27 Phase Plane Method point using say Taylor series expansion and then the nature of phase trajectory around the singular point is determined by linear system analysis which is described earlier. 8.11 Delta Method In isocline method, entire phase plane must be filled with line segments of different possible slopes. If only a single solution curve is needed, only a few of these lines are put to use, thus wasting a lot of effort. A technique of construction known as the delta method leads more directly to the solution. This method applies to the solution of equations of the type, XHitx, x,t) = 0 ef) Where f must be continuous and single valued but may be nonlinear and time dependent. ‘The equation (1) is written as So abc f(x, x, )-a@ix = 0 wa(2) The constant w} may be determined from the equation (1) itself or may have to chosen from other information. Once again it is convenient to introduce the definitions t = ayt dx . and v = S so as to give, ogy FORx+L(v, x H-abx] = 0 +(3) dv [x4 8(v, x, DO] or a . rts fd) where aux g = Lays ox 8) a Equation (4) is similar to that one used for the case of isocline methed. Because of the way, in which it is set up, the $ method is most immediately applicable to the equations with oscillatory solutions, although it is not limited to this class of equation. Function 6 of equation (5) depends upon v, x and t but it is assumed constant for small changes in these variables. Under this assumption, the variables of _ equation (4) can be separated and integrated to give, v?+(x+8? = constant = R? (say) +6) Equation (6) represents a circle of radius R and centre at (-4, 0). Thus for a suitable increment, the solution curve is the arc of a circle having these properties. The constructionModern Control Theory 6-28 Phase Plane Method is shown in Fig. 8.21 where [x(0), v(0)] represents a point on the solution curve at the time (0). These values used in equation (5) allow the calculation of &.This values of 5 determines the centre of the circular arc located on the x-axis. The radius R is automatically fixed. A short circular are represents a portion of solution curve. Actually it is more accurate to use the average values of v, x, t for that increment. Again, the allowable length of the arc is a compromise. v Fig. 8.27 Examples with Solutions > Example, 8.2 :A linear second order seroa is described by the equation C+ 280, Crane = O where § = 0.15, , = 1 rad/sec, C(O) = 15 and € = 0. Determine the singular point. Construct the phase trajectory, using the method of isoclines (VTU: July/Aug.-2006) Solution : Consider the equation C+2to, CHa 4 ° Given §=015,@,=1.C()=15, C = 0 C+ 201 0)CH12C = 0 C+03 CHC = 0 Taking Laplace transform, 57s) +0.35 Cis) +Cls) = 0 [s? +0.3541] Gs) = 0 Consider s7 40.3541 = 0 -0.34,(03) 400) _ -034)[1-9773] S12 8 De = -0.15£)0-9596Modern Control Theory 8-29 Phase Plane Method The roots are complex conjugate and located in left half of s plane. In time domain the response will be underdamped consisting of oscillations and the singular point is thus stabel focus. Consider again the differential equation 2, a OY 93 a8 +Qt) = dt? Let x= CY, m= AY) = Ct) =x, x2 = Gt) =- CW) -03 Cy =», 03x, _ Axp, | xp /dt mF Gk) 7 Gk, fat > MX = -x,-0.3x, (m+03)x, = =x, x = -|—_k. 2 m+0.3 |"! The above equation is an equation of isucline. At initial point (1.5,0) O= 1 4s ~weoa 5) m = -0.3-c = - Due to this trajectory moves downwards Using the method of isocline, the phase portrait can be drawn as shown in the Fig. 8.28 Considering various values of m socline equation Xp = - 0.7692 xy Xg = — 0.1886 x, = [areModern Control Theory B-32 Phase Plane Method The type of singular point obtained is stable focus. Ji) Consider, y+ 3y+2y =0 Let the state variables be x, = y and x, =y The corresponding state model is Eliminating the time variable, dy Oxy ~ 2x dk, *2 The characteristic equation is given by, s?+3s+2 = 0 (s+2)(s+1) = 0 $ = -2; s=-l We have, 7, = -l andy, = -2 } jw —*———* - ne? net Xp Wek (a) (b) Fig. 8.30 The nature of phase trajectory is as shown in the Fig. 8.30 and the type of singular point obtained is stable node. ‘The response in this case is given by y(t) = C,e"l! +C,e"2!. The transient term e?2! decays at faster rate than e¥l', As time to, x, >Cye!l! 40; x, = ye!!! 90. The singular point node obtained in this case (0, 0) is stable in (y, y) plane.Modern Control! Theory 8-33 Phase Plane Method woe ili) Consider, y+3y-10 = 0 Let the state variables be x; = yj x, sy The corresponding state model is, xy BY EX Ky = y =-3y+10 =-3x, +10 Eliminating the time variable dx, _ -3x, +10 ox X The characteristic equation is given by, s*+3s-10 = 0 s?45s-2s-10 = 0 *. $(8+5)-2(s+5) = 0 (s +5) (8-2) = 0 ¥p=5;¥) =2 There are two roots one positive and one negative. The singular point obtained in this case is called saddle point. The slopes of the two straight line trajectories with slopes dependent on root values. The nature of phase trajectory is shown in the Fig, 8.31. Fig. 8.34Modern Control Theory 8-35 Phase Plane Method | Fig 8.32 IModern Control Theory 8-36 Phase Plane Method m 2 Yo = — 0.3848y, 63.49" 5 Yo = ~ 0.1785 yy 78.69" 10 Yg = — 0.0943y, 84.28° « ¥2 = 0 80° -1 Yo = 2.594 —a5° 2 Yq = O.7142y4 63.43" 6 Yo = 0.2272y, 78.69" 10 Yo = O.1063y, 64.28" = y= 0 780° The phase-plane plot is shown in the Fig. 8.32.(See Fig. 8.32. on previous page) imp Example 8.5: Draw the phase-plane trajectory for the following equation using isocline method : x4 2Gwx+ a? x =0 Given, & = 0.5, w= 1, Initial point (0, 6) (VTU: JanJ Feb.-2007) Solution: Consider the equation X+260, xtu2x = 0 Given that E= 05, a, =1,x= 6, =0 X+2(0-5)(1) xt 12x = 0 xtxtx = 0 Let y= x Yo =x v1Moder Control Theory 8-37 Phase Plane Method myz = “¥i- ¥2 (mtl)y2 = Yo = mil = -24 m= At initial point (6, 0). «The trajectory moves downwards Using the method of isocline, the phase portrait can be drawn considering various values of m. yg =~ 0.53 4 Yo # — 0.168 yy ¥q = ~ 0.080 y, ¥2=0 ya = 14 ¥q = 0.25 ¥4 ¥q = 0.1111 ¥4 ¥q = 0.0526 yy The phase-plane plot is as shown in the Fig. 8.33.Phase Plane Method Modern Control Theory Fig. 8.33(8 - 40)Liapunov's Stability Analysis 9.1 Introduction The state equation for a general time invariant system has the form x = f (x, u). If the input u is constant then the equation will have form x = F(x). For this system, the points, at which derivatives of all state variables are zero, are the singular points. These singular points are nothing but equilibrium points where the system stays if it is undisturbed when the system is placed at these points. The stability of such a system is defined in two different ways. If the input to the system is zero with arbitrary initial conditions, the resulting trajectory in phase-plane, discussed in earlier chapter, tends towards the equilibrium state. If the input to the system is provided then the stability is defined as for bounded input, the system output is also bounded. For linear systems with non-zero eigen values, there is only one equilibrium state and the behaviour of such systems about this equilibrium state totally determines the qualitative behaviour in the entire state space. In case of nonlinear systems, the behaviour for small deviations about the equilibrium point is different from that for large deviations. Hence local stability for such systems does not indicate the overall stability in the state space. Also the non-linear systems having multiple equilibrium states, the trajectories move from one equilibrium point and tend to other with time. Thus stability in case of non-linear system is always referred to equilibrium state instead of global term stability which is the total stability of the system. In case of linear control systems, many of the stability criteria such as Routh’s stability test, Nyquist stability criterion etc. are available. But these cannot be applied for non-linear systems. The second method of Liapunov which is also called direct method of Liapunev is the most common method for obtaining the stability of non-linear systems. This method is equally applicable to time varying systems, stability analysis of linear, time in variant systems and for solving quadratic optimal control problem. (9-1)Modern Control Theory 9-2 Liapunov's Stability Analysis 9.2 Stability in the Sense of Liapunov Consider a system defined by the state equation x = f(x, 1). Let us assume that this system has a unique solution starting at the given initial condition. Let us consider this solution as F(t: Xp, ty) where x = xp at t = tp and t is the observed time. Fltg : Xq ty) = Xp If we consider a state x, for system with equation x = f (x, t) in such a way that f(x, t) = 0 for all t then this x, is called equilibrium state. For linear, time invariant systems having A non-singular, there is only one equilibrium state while there are one or more equilibrium states if A is singular. In case of non-linear systems as we have seen previously there are more than one equilibrium states. The isolated equilibrium states that is isolated from each other can be shifted to origin ic. £(0, t) = 0 by properly shifting the coordinates. These equilibrium states can be obtained from the solution of equation [Og Hh=0 Now we will consider the stability analysis of equilibrium states at the origin. We will consider a spherical region of radius R about an equilibrium state x, such that x-x i)
0 and any real number 6 > 0, irrespective of how small it is, there is always a state xg in S(8) such that the trajectory starting at this states leaves S(z). 9.6 Graphical Representation The graphical representation of stability, asymptotic stability and instability is shown in the Fig. 9.1 (a), (b) and (c) respectively. The region S(5) is binding the initial state xp and the region S(e) corresponds to the boundary for the trajectory starting at xq. The correct region of allowable initial conditions are not specified for the definitions of stability that we have seen previously. Thus theseModern Control Theory 9-4 Liapunov's Stability Analysis definitions apply to the neighbourhood of the equilibrium state only if S(e) is not corresponding to the entire state plane. Se) Ste) ” C:) . CH) (a) Stability {b) Asymptotic stability (c) Instability Fig. 9.4 In Fig. 9.1 (c), the trajectory is leaving S(c) and shows that the equilibrium state is unstable. It can not be stated that the trajectory will go to infinity as it may approach a limit cycle outside the region S(e). In case of linear, time invariant system if it is unstable trajectories starting near the unstable equilibrium state will go to infinity which is not necessarily true in case of non linear systems. The definitions of stability that we have seen se far are not the only ones defining the stability of equilibrium state. There are other ways of defining stability also. In classical control theory, the systems which are asymptotically stable are called stable systems and those which are stable in the sense of Liapunov but are not asymptotically stable are called unstable. 9.7 Some Important Definitions In this section we will consider some important definitions which are useful in understanding Liapunov's stability criterion. 9.7.1 Positive Definiteness A scalar function F(x) is said to be positive definite in a particular region which includes the origin of state space if F(x) > 0 for all non-zero states x in that region and FQ) = 0 eg. F(x) = xq +2x5 9.7.2 Negative Definiteness A scalar function F(x) is said to be negative definite if — F(x) is positive definite. eg F(x) = — x? -(3x, +2x,)?Modern Control Theory 9-5 Liapunov’s Stability Analysis 9.7.3 Positive Semidefiniteness A scalar function F(x) is said to be positive semidefinite if it is positive at all states in the particular region except at the origin and at certain other states where it is zera. eg. F(x) = (xy + x2)? 9.7.4 Negative Semidefinite A scalar function F(x) is said to be negative semidefinite if - F(x) is positive semidefinite. 9.7.5 indefiniteness A scalar function F(x) is said to be indefinite in the particular region if it assumes both positive and negative values irrespective how small the region is eg. F(x) = x, xq +9 9.8 Quadratic Form A class of scalar functions which plays important role in the stability analysis based on Liapunov's second method is the quadratic form. Pu n 4 P vs ne eg. F(x) = X™PK = [xy xp... XA] 2 F 2 P, 7 Pan In P is real symmetric matrix and x is a real vector. 9.9 Hermitian Form If x is a complex n vector and P is a Hermitian matrix then the complex quadratic form is called Hermitian form. Py Pao Pin] . _—_ _|lPp Pho P. eg. FQ) = Pee gee) OB an | | 82 Pin Pan =: Pon} Lp; The stability analysis in state space the Hermitian form is commonly used than the quadratic from as it is more general. The positive definiteness of the quadratic form or Hermitian form F(x) can be determined by Sylvester's criterion which states the necessary and sufficient conditions for quadratic or Hermitian form to be positive definite isModern Control Theory 9-6 Liapunov's Stability Analysis Py eee rm IP, oP By ove P, Py > op! m SO... | iso : Pin roe Pan F(x) = x° Px is positive semidefinite if P is singular and all the principle minors are non-negative. F(x) is negative definite if - F(x) is positive definite. Similarly F(x) is negative semidefinite if - F(x) is positive semidefinite. 9.10 Liapunov's Second Method A system which is vibrating is stable if its total energy is continuously decreasing. This indicates that the time derivative of the total energy must be negative. The energy is decreased till an equilibrium state is reached. The total energy is a positive definite function, This fact obtained from classical mechanics theory is generalised in Liapunov's second method. If the system has an asymptotically stable equilibrium state then the stored energy decays with increase in time till it attains minimum value at the equilibrium state. But there is no simple way for defining an energy function. For purely mathematical system. This difficulty was overcome as Liapunov introduced Liapunov function method which is fictitious energy function. Liapunov functions depend on x), %, «+ %q and t It is given as F(x, x, «+ Xp #) or as F(x, t). In Liapunov's second method, the sign behaviour of F(x, t) and its time derivative F(x, ) = dF(x, Y/dt gives us information about stability, asymptotic stability or instability of an equilibrium state without requiring to solve the equations directly to get the solution. 9.11 Liapunov's Stability Theorem Consider a scalar function V(x), where x is n vector and is positive definite, then the states x that satisfy V(x) = C, where C is a positive constant, lie on a closed hyper surface in n dimensional state space at least in the neighbourhood of origin. This is shown in the Fig. 9.2. If V(x) is a positive definite function obtained for a given system such that its time derivative taken along the trajectory is always negative then V(x) becomes smaller and smaller in terms of C and finally reduced to zero as x reduces to zero. This indicates asymptotic stability of the origin. Liapunov’s main stability theorem is based on this and gives a sufficient condition for asymptotic stability.Modern Contro! Theory 9-7 Liapunov's Stability Analysis Liapunov's stability theorem is as given below. Consider a system described by equation X= f(x, ) where f(o, t) = 0 for all t. If there exists a scalar function V(x, t) having continuous first partial derivatives and satisfying the conditions such as V(x, t) is Positive definite and V(x, t) is negative definite then the equilibrium state at the origin is uniformly asymptotically stable. If V(x, t) — 0 as || x [| > where jx J] is norm of x, then the equilibrium state at the origin is uniformly asymptotically stable in Fig. 9.2 the large. The above theorem is a basic theorem of second method. The theorem 2 is stated below. Vincrease Consider the system described by x = f(x, t) where f (0, t) = 0 for all t 2 to. Lf there exists a scalar function V(x, t) having continuous first partial derivatives and V(x, t) is positive definite, V (x, t) is negative semidefinite V (6 (t : x t), t) does not vanish identically in t 2 ty for any ty and any xp # O where 6 (t : xp, tg) denotes or indicates the solution starting from xg at ty then the equilibrium state at origin of the system is uniformly asymptotically stable in the large. The equilibrium state at origin is unstable when there exists a scalar function U(x, t) having continuous, first partial derivatives and satisfying the conditions U (x, t) is positive * definite in some region about the origin and U (x, t) is positive definite in the same region. 9.12 Stability of Linear and Nonlinear Systems If the equilibrium state in case of linear, time invariant system is asymptotically stable locally then it is asymptotically stable in the large. But in case of a nonlinear system, the equilibrium state has to be asymptotically stable in the large for the state to’ be locally asymptotically stable. Hence the asymptotic stability of the equilibrium state of linear, time invariant systems and those of nonlinear systems is different. If it is required to tesi the asymptotic stability of any equilibrium state for a nonlincar system then the stability analysis of linearised models of non-linear systems is totally insufficient, The nonlinear systems are to be tested without making them linearized, There are various method based on Liapunov's second method such as Krasovskii's methodModern Control Theory 9-6 Liapunov's Stability Analysis which can be uscd to test sufficient conditions for asymptotic stability of nonlinear systems. We shall study in detail Krasovskii's method for asymptotic stability test. $9.13 Construction of Liapunov's Functions for Nonlinear Systems by Krasovskii's Method Liapunov's direct method is a most powerful tool available for doing the stability analysis of nonlinear systems. The stability analysis by Liapunov's method includes determination of a positive definite function V(x) called Liapunov's function. But unfortunately there is no universal technique or method for selecting Liapunov's function which will be unique for a special problem. Some of the Liapunov's function will give better results than other functions. There are various techniques available for construction of Liapunov's functions Similarly for a given function V(x), there is no general method which allows us to confirm that the function is positive definite. But if V(x) is in the quadratic form in x;’s then Sylvester's theorem can be applied to test the positive definiteness of the function. If it is not possible to obtain the Liapunov's function of required type then it does not indicate that the system is unstable. The theorems that we have seen so far just provides sufficient conditions for stability. The indication from the fact that Liapunov’s function of required type is not obtained is that the attempt to establish the stability of the system is failed. Krasovskii's method gives a sufficient condition for the equilibrium state to be asymptotically stable. Consider the system described by following equation. x = f(x); (0) = 0. The origin is a singular point. Let us consider Liapunov's function as, ve pe Here P is a symmetric positive definite matrix. fires Pt of ax oe Tf ae " Now,Modern Control Theory 9-9 Liapunov's Stability Analysis ] is called Jacobian Matrix given by of, of, of Oey Oey Oxy ay ay at, J = |ox; oxy Ox, of, aly af, | oxy Substituting fin expression for V we have, vie fy pee f rye = Pup Ps Pye Let Q=esPp+y As V is positive definite, for the system to be asymptotically stable, Q should be negative definite. If in addition V(x) is tending to infinity as norm of x tends to infinity (il x || +=) the system is asymptotically stable in the large. Alternatively if Liapunov function is taken as Vog = fbx). fx) J(x) is the Jacobian Matrix given earlier af, af, af, Ox, Oxy Ox, af, af, Joo = [xy Hey Oey hy a ox, oxy as, | We have f(x) = ae a = Ji) x = J) « £60) Now we have, ¥ (x) = £76) foe) + fod F(x) = [0 - fa] fog + FF G9 Joy feo = £) 76x) <0) + £ Gx) TO) - f) = £00) LI") + Joo] £69Modern Control Theory 9-10 Liapunov's Stability Analysis * > * Let fix) = J" G) + JO). If Ix) is negative definite then V(x) will also be negative definite. Therefore V(x) is Liapunov function and the origin is asymptotically stable. If V(x) is tending to infinity as || x || + then the equilibrium state is asymptotically stable in large. The Krasovskii's theorem is different from the normal linearization technique. It is not limited to small departures from the equilibrium stale. The Liapunov function ¥(x) and V(x) are expressed interms of f(x) or x rather than in terms of x. Krasovskii method gives sufficient condition for asymptotic stability of non-linear systems and necessary and sufficient condition in case of linear systems. An equilibrium state of a non-linear system may be stable even if the conditions stated above are not satisfied. So comments on stability of non-linear systems cannot be vigoursly made by applying this method. Let us have an illustration of Krasovskii's method consider a second order non-linear system described by following cquations Xp = By Oy) + Be ba) X= + bx, Let us consider that g)(0) = g2(0) = 0 and gy (x) and g(x) are real and differentiable. We will also assume that ley (1) + By Gehl? + fey + bxgf? 3 as |] x] > 2 Now we will obtain the sufficient conditions for asymptotic stability of the equilibrium state x = 0 fy = xy = £ (x1) = gy Gy) + B Ox) fy = X= fhe) =x + bey Let J(x) be the Jacobian matrix given as af, ah at a a 1 = By Dy] [yl +B262)1 Je, Bra) +B 0a)l fy Oty glx: bee] pehy the] Ox, Oxy 1 2 = [B100) 826%) 1 b a a Here Big) = PBLEY and (xg) = 28RD? Now we have, fd = J" G9 + Jo)Modern Control Theory 9-11 Liapunov's Stability Analysis. Bi) 1) fei) s200)| 820%) b| 1 b _ [Bie+siO) Teatro) | gb by)41 b+b | _ { 28s 1+g4 (x3) [+ghQ2) 2b By Krasovskii’s theorem if f(x) is negative definite then the equilibrium state x ~ 0 of the system considered is asymptotically stable in the large. Hence if g(x.) < 0 for all x, #0 and 4b g4(x,) - [1 + gy(xq)F > 0 for all x; #0, x, #0 then the equilibrium state at x = 0 is asymptotically stable in the large. This two conditions are sufficient for asymptotic stability. 9.14 The Direct Method of Liapunov and the Linear System For linear systems, Liapunov's direct method proves to be a simple method for stability analysis. Use of Liapunoy's method for linear systems is helpful in extending the thinking towards nonlinear systems. Consider linear system described by state equation x = AX The linear system described by above equation is asymptotically stable in the large at the origin if and only if for any symmetric, positive definite matrix Q there exists a symmetric positive definite matrix P which is the unique solution A "P+ PA=-Q The proof of above theorem can be given. For this we will assume the symmetric positive definite matrix P exists which is he unique solution of the equation V(x) = x! Px, Consider the scalar function, V(x) = x? Px Here V(x) > 0 for x #0 and V(0) = 0 We have, vod = oP xex' Px Yoo = AGO™Px4+x™PAx = x7 AT px +27 PA x " xt (ATP + PA) x =x! Ox ‘As Qis positive definite, (x) is negative definiteModern Control Theory 9-12 Liapunov's Stability Analysis Let norm of x define as Wx = &T Py? Vos) = xt? “ Vix) 9 eas |] x |] > 2 The system is therefore asymptotically stable in the large at the origin. The result is also necessary. To prove this, assume that the system is asymptotically stable and P is negative definite. ” Vix) = x! Px Wa) = - BT Px ex py = -[-x" ax] = x' Qk > 0 This is the contradiction as Vix) = x'Px satisfies instability theorem. Hence the conditions for the positive definiteness of P are necessary and sufficient for asymptotic stability of the system. The Liapunov's direct method applied to linear time invariant systems is same as the Hurwitz stability criterion. Examples with Solutions. ‘p> Example 9.4: Show that the following quadratic form is positive definite Vix) = x? +33 +4x3 + 2x yxy — 4x x5 —DxQxy Solution ; The above given V(s) can be written as fe 1 -2)[x, Vix) = xT Px efx xpxgl] 11-1 2 2-1 4]|x, Applying Sylvester's criterion we have, 81 8 1 -2 8B > 0; I ‘| 7>0;;1 1 -1js2050 2-144 As all the sticcessive principal minors of the matrix P ate positive, V(x) is positive definite.Modern Control Theory 9-13 Liapunov's Stability Analysis mm Example 9.2: Determine the stability of a non-linear system governed by the equations y= ey + 222 xy %2 =~ 2 Solution : Let us select the Liapunov's function to be V = xj +x3 We have, Vos xp +x} av av oxy av ax, dt Ox, "dt "ox, dt av av Now, oxy 7 2x xy 7 2x; d . aor = marx t axt xy dx . ae = ye-% Substituting, dv Gr = 2 Gy + 2xp xq) + a) Gx) = = 2x? (1-2 xy x2) 2x3 Thus {Y is negative definite if 1 — 2x xz > 0. Thus Liapunov’s criterion is satisfied and hence the origin of the system is asymptotically stable. weap Example 9.3: Determine the stability of the system described by the following equation . -1 1 beaesa=[) 4 Solution : To check the stability of the system described by equation X= Ax we will solve the equation AT P + PA = — Q where for any symmetric, positive definite matrix Q, there exists a symmetric positive definite matrix P. We can find out matrix P for any arbitrary choice of positive, definite real symmetric matrix Q. Let us select Q = I, where I is identity matrix * A’p+PA = -Q Let P= [pe P| Pu P22Modern Control Theory 9-14 Liapunov's Stability Analysis F al [Pu Pra |, [Pu mle ‘ = [ H| 1-4) [Pa Pas} [Po Pali-? 4 o -1 [Re —2Py “Pi ee bee -2Py Pu -2Px | .Pu-4Pa | Pi2 74 Pap Pa -2P2 Pa ~4 P22 -1 @ 0 =1 = ~2py) -2py -2Pq2 Pu SP a2 2] = Ct | Pu-SPy-2P2 Py 5P22 +P 9-1 But P is symmetric matrix ..py3 = P2, . [ py -4Py2 Pi “5P12~2Pm] _f-1 0 “TPy Py -2P22 Py: ~5P 22 ol 2pu-4p2 = -1 Pu~5Pi2-2 Paz = 0 2P12- 5 pa = -1 Solving the three equations using Cramer's rule, 2(25 +4) +4(-5) a= 2(29) -20 =-76 =-1(25 44) +4(0-2) dpi = 29-8 a 4 g| =-200-)41¢-8 Apr = ~ [tS=4 0-1-5 -2 A -1) =-26)+4-102) Apr = | 1-5 0} =-10-4-2 oO 2 -1] =-16 = Seu 2-37 7 Pu = A> 27s > 78 Apiy 1 ai Pi2 = Pa=—R> > Dg > 38Modern Control Theory Liapunov's Stability Analysis {37 1 78 | i i 78 78. Using Sylvester's criterion it can be seen that P is positive definite, Therefore origin of the sytem under consideration is asymptotically stable in the large ump Example 9.4: Use Krasovskii’s theorem to show that the equilibrium state x = 0 of the system described by My =- 3x, + xy = 3 Xp = Xp XQ 5 is asymptotically stable in the large. (VTU: Jan/Feb.-2005) Solution = we have, x, = fy (x) = - 3x, + x» xy = f(x) = xy — x - HP The Jacobian matrix is given by at, a J = ot, oxy ofp . 2 2 Rd ty 173% _f3 1 Jed, -i-3x3 Let Q = JP+Py Pu Piz ( 1 ' Let Ps = be identity matrix ee | O91 ° _f3 1 Jp opm oye 1 a= [PF asalle de If a3 3 1 -3 1 _ —t 2 Q- [; -a-axg|*]1 -1-ax3] "[2 -2-6x3Modern Control Theory 9-16 Liapunov's Stability Analysis For system to be asymptotically stable Q must be negative definite ic. - Q must be positive definite 6 2 6 -2 “Qe aly -2-6x3 | |-2 2+ 6x5 Using Sylvester's criterion, 6 > 0 and 12 + 36 x3 - 4 > 0 ie. 12 (1 + 3x3) > 4 then the equilibrium state x = 0 is asymptotically stable in the large. im Example 9.5 : Investigate the stability of the follawing non-linerar system using direct method of Liaupnov. xy =X) e-xy-at 2 1X] Xe (VTU: July /Aug.- 2005,Jan./Feb,-2005) Solution : Given that % = x2, %y =~ x1 - x? x Let the Liapunov function be, V = x7 + x3 * . . Vi = 2xy x +. 2%7 a = Dy xy + 2 xp x1 — RF MD) BS 2x, xy ~ Bey xq — 2 x7 XE = 242 = -2x2 x3 It can be seen that V <0 for all non-zero values of x; and x, Hence the function is negative definite. Therefore the origin of the system is asymptotically stable in large. tmp Example 9.6 : A second order system is represented by . ool x= Ax where A = [ | -l -1 Assuming matrix Q to be identity matrix, solve for matrix P in the equation ATP + PA = ~ Q. Use Liapunov theorem and determine the stability of the origin of the system. Write the Liapunov function V(x). (VTU: July/Aug.-2005, Jan/Feb.-2007) Solution : x= AXModern Control Theory 9-17 Liapunov's Stability Analysis Consider ATP+PA = -Q [? ‘fPu Pia], fPu Pajf 9 WL fA oO 1 o-l} [Par P2eJ LP21 P22 jl-} -3 oH f Piz -P2 ze perelel | Pui Pi2Pi2~P22J [Paz Pi2Pn} [0-1 [ Piz Pu Piz “P22 to ‘| Pu-Pi2-Px2 (Py. P22) 0-1 -2pp =-1 “Ppp =e 1/2 1 Pi ~ Piz ~ P22 = 9 * Pum P23 P= 5 l 2(Pi- Pm) = -1 Pam P2 F935 11,1 Pa = Patgegtgel 1 3 Pu = Prat Pa=5t+l=5 3 2] P= : *| 3 Vy - Using Sylvester's criterion, P can be tested for positive definiteness 31 3 22°31 ~=5 Zz? Ol feggsgre iz 1 the large. The Liapunov's Vod to) P is positive definite. Hence the equilibrium state at origin is asymptotically stable in function is 31 x =e UPxsk, x |? ? rm] 1 *2 5 1 2 1 xy = xy aX2 ah tee x, + +t x2 BN ea Fy 1X2 FXModern Control Theory Liapunov's Stability Analysis imap Example 9.7 : A system is described by the following equation : (VTU: Jan/Feb.- 2008) : a 2 tartan a-[! 2] Assuming matrix Q to be the identify matrix, solve for matrix P and comment on the stability of the system using the equation ATP +PA=-Q. Solution : To check the stability of the system described by equation x= Ax, We will solve the equation, ATP+PA =-Q where for any symmetric, positive definite matrix Q, there exists a symmetric, positive definite matrix Q, there exists a symmetric positive definite matrix P. We can find out matrix P for any arbitrary choice of positive, definite, real symmetric matrix Q. Let us select Q = 1, where | is Identity matrix. A™P+PA =-Q. Let Pe [Pn Pe Pa P22 ae [fT S} at 4] [2 "| Pu Pio], [Pu Pa [: “te 4 2-4) [Pa Pel [Pa Patti -4) [0 - [ee Pa "Pig + Pao }-[eet Pix ~2Pu > elt | “Pum 4Px 9 ~2Pi2 ~ 4P22 J [Pat P22 ~2Pu - 4Pm} LO -1 2p) +Py2*P2 -SP12-2Pu+Px |] _[-} 0 2p SP P22 ~2P2-2Py-BP2] LO -1 But P is symmetric matrix -.p)) =P, ‘ 2p tPit Py | —2Pn—SPi2 + P22 -[o “ “PSP i tP22 Piz -2Pa~8Pr} LO HL 2py+PitPa = pi -5P2 +P = 0 2p) - 2p) -8py = Wl Solving these three equations using Cramer's ruleModern Control Theory 9-19 Liapunov's Stability Analysis 2 1é«1 =-2(40-4 2)—1(16 + 2)+1(4—-10) A={-2 -5 1 ==2(42)-18-6 -2 -2 ~8] =-84-18-6=-108 11) 1] Se 1d0+2)-1(1)+1(-5) ap, =)0 5 1 80-1-5 =-86 ~1 -2 -8 2-21] +041) +116 +2)41(2) Apy =]-2 0 1] 2241842 2 -1 -6| = <1] ==25)-1(2)-1(4~10) ap,, = 0] =-10-2+6=-6 -1 - ‘Pu 86 | 86 Pu = a * Zios ~ 108 AP 2 18 Piz = Pa = e = ~08 AP 22 6 _ f Px =~ -108 ~ 108 36 = p = | 08 TOR oie 6 108 108 Using Sylvester's criterion, it can be seen that P is positive definite. Therefore origin of the system under consideration is asymptotically stable in the large. 2 mma Example 9.8 : Consider the system: with differential equation e+ket+k,e +¢=0, Examine the stability by Liapunov's method, given that k > 0 and ky > 0. (VTU: July/Aug.-2007) Solution: Consider the given equation, woe ad etke tke +e=0k>0,k; > 0 Let =e 3 ~kxy -kyx} x)Modern Control Theory 9-20 Liapunov's Stability Analysis As there is no specific procedure for selecting Liapunov's function but it has to be selected from experience. Let us select f = x} +x} as Liapunov's function df . : Se 2x, ky 42x) X at = 2xy( xy) #2x, (Rey ky x} -x;) = 2xyx, ~2ke} -2k, xd -2x x, 2 4 exe 2 4 ~Dkx} -2k x4 =—2kx} +k, x} The above equation is negative semidefinite and hence the system is stable. ta ~Example 9.9: Examine the stability of te system described by the following equation by Krasavaskii's thearem. XLS =x, - 3 XD = Xp -¥3 (VTU: July/Aug.-2007) Solution : We have, % = fy) = y x2 = AG) = x) x3 3G The Jacobian matrix is given by, af af = fo, O T= oe ag Oxy Oxy 1 0 ] yedy ~1-3x3 | Let Q = JTree 10 Let p = {Pu Pri, be identity matrix Pa Pa} [9 1 . 1 1 10 1 0) f-1 Qo eo [c ana] [a i}*fo A [; ana] fat a1 0 Fy? 1 “Jo -1-3x3]" [1 -1-3x3|7 |i -2-6xModern Control Theory 9-21 Liapunov's Stability Analysis O« [2 1 “ha -a(u+38)| For system to be asymptotically stable Q must be negative definite ic. - Q must be positive definite. -2 i 2 -1 “Q= 44 -2(1+3x3)|"}-1 2(14+3x3) Using Sylvester's criterion, 2>0 and 4+12x3 -1>0ie. 4412x} >1 ie 4(14+3x}) 21 then equilibrium state x = 0 is asymptotically stable in the large. ime} Example 9.10 : Deteremine whether or not following quadratic form is positive definite : (VTU: Jan.JFeb.-2007) Q(xy Xp) = 10x} 4x} +9 42x 4x9 -Ixqxg —ANX Solution : Consider the given equation Q(x1.xq) = Ox} 44x} +9 + 2x4 xg —Bxq x5 4x pS Consider, Pu Piz Pas} | ¥1 Q= [x%2%3]]Piz Par Pas] |%2 Pis P23 Pas} L¥3 4 = [Pr tPio%2Pis%3 Pr2%, +P22%2+P23%3° Pasi tP29%2*Pas%3] | é2 x 3 = Pus] + P2283 +P 33X34 2P 12%1%2 +2P23%2%3 +2P13%1%53 Equating coefficients of above equation with those from given equation Pop =4: Pal P= Po3 =-1 Py =-2 The given equation can be rewritten as, 10 1 -2} fx, Q = [x,y GQ) fl 4 =], 2-1 1 LXa9-22 Liapunov's Stability Analysis Modern Control! Thoory Applying Sylvester's criterion we have, 10 o,f? 59 alt ‘ x Por, 4 =39>0; - 2-11 10(4-1)-1(1-2}-2(-1+8) 30+1~-14=17 >0 As all the successive principal minors of the matrix P are positive, Q(x; xp) is positive definite. Using Lyapunov's direct method, find the range of K to guarantee in Example 9.11 (VTU: JanJFeb.- 2006) stability of the system shown in the Fig. 9.3. Solution : Fig, 9,3 The above diagram can be redrawn as, Fig. 9.4 ‘The state equations are, x, = —Kxy Xp XOXModern Control Theory 9-23 Liapunov's Stability Analysis Ry = X27 In matrix from above equations are written as, . “1 0 0 -K] fx, xp] = ]1 -1 0] |x, x o 1 -1] [x3 fo 0 -K © 10 Here Aszijil 1 0] sATs}O a2 1 lo 1-1 -K 0-1 To find the range of values of K for stability, we will solve the equation ATP+PA =-Q. 000 Let us select Qeslo 00 oo 1 This selection of Q is valid as this does not make x! Qx identically equal to zero except at the origin. 0 0 0} fx, -[x1 xp xq] 0 0 0} fxg} =x} 00 1} [x5 Consider ATP+PA =-Q O 1 8)[en Pe Pi] [Pu Pie Pi} {2 9 -Ki fe 0 -1 1] ]Pi2 Po Pos} + [Pi Por Pa] |i -1 0] =}0 -K 0 -1) [Piz Pox Pas Pis Pax Pal lO 0 -1y jo Solving the equation we get — 0 lt K(2-K) 2(2-K) P= 0 1 1 2(2-K) 2(2-K) =1 1 1 a(2-K) 2(2-K) (2-K)Modern Control Theory 9-24 punov's Stability Analysis For P to be positive definite it is necessary and sufficient that K(2-K)>0 K>0;2-K>0,2>K Thus for 0 < K < 2, the system is asymptotically stable. Review Questions Write a note on stability in the sense of Ligpunov. Define i) stability ii) Asyaptotic stability iii) Asymptotic stability in the large. Show the graphical representation of stability, asymptotic stability and instability. Define i) positive definiteness ii} negative definiteness iii) positive semidefiniteness iv) negative semidefiniteness v) indefiniteness Discuss Quadratic form and Hermitian form. Explain Liapunov's second method and Linpunoo's stability theorem. Write a note on stability of linear and nonlinear systems. oN aw Write a short note on Krasovskii’s method of constructing Liapenov's functions for nonlinear systems. © Examine the stability of the origin of the following system x = x kg = = 6x, -5 x9 10. Write a Liapunov function for the system allt) Determine the stability of the origin of the system. TL. Determine the stability of the equilibrium state of the following system. | Bos oxo 2a +2 2 = m-4e-1 1. 8 State and explain Liapunov’s theorems on i) Asymptotic stability ii) Global asymptotic stability and iti) Instability. 13. Use Krasouskii's theorem to show that the equilibrium state x = 0 of system described by xy = 3ry 4% Xp = xy 49-43Modern Control Theory 9-25 Liapunov's Stability Analysis 14, Investigate the stability of following non-linear system using direct method Liapunov. Xp EX yaya 15. A second order system is represented by x= Ax where A ={ a l-a-a Assuming matrix Q to be identity matrix, solve for matrix p in the equation AUP + PA = - Q. Use Liapunov theorem and determine the stability of Uke origin of system. Write the Liapunov function V(x). goo(8 - 26)Modern Control Theory & Chapterwise University Questions with Answer -1) don./Feb. duly/Aug. Jon./Feb. July/Aug. Jon./Feb. July/Aug. « = 2008 Jon./Feb. = 2005 + 2005 = 2006 = 2006 + 2007 2007qd) State Variable Analysis and Design Qi Ans. : Qz Ans. : a3 Ans. : a4 Ans. : Mention the disadvantages of conventional control theory and explain how these are overcome in modern control theory with particular reference to (i) Non-linear systems Gi) Time varying systent (iii) Analysis (iv) Design and (v) Computer applications. (QanJFeb.-2005, 10 Marks) Refer section 1.1. Explain the concepts of state variable, state and state model of a linear system. (JanJ/Feb.-2006, 6 Marks) Refer section 1.2. Define the concept of i) State ii} State variables iii) State space iv) State vector. Quly/Aug.-2006, Jan/Feb.-2008, 6 Marks; Jan./Feb.-2007, 10 Marks) Refer section 1.2. Compare classical control theory against modern control theory. Quly/Aug.-2006, 4 Marks) 2° Refer section 1.1. List advantages of modern control theory over conventional contro! theory. Quly/Aug.- 2007, 4 Marks) Refer section 1.1.2. Derive the equation of the vector model differential state equation. (Quly/Aug.- 2007, 2 Marks) Refer section 1.3. Mention the differences between state space techniques and classical approach. (Jan/Feb.-2008, 6 Marks) Refer section 1.1. aoa (P - 2)Modern Control Theory P-4 State Space Representation Q.6 —_Linearize the following equation in the neighbourhood of the origin. “8 = tan 29 - 3sinO + 2ue"/? +u3 Obiain the approximate response &t) for u = 0.02, with the system initially at equilibrium. QanJFeb.-2006, 6 Marks) Ans. 2, 28S tan! 20-3 sind + Que"/24u? dt? sin@ = .. in the neighbourhood of origin tan 720 = 20... in the neighbourhood of origin 2, ae, 29-304 2u0/2 +03 dt? 2 3 ul? ike ) 0 2 2 43 ao. -0+2u ies a Je dt? 2 a0 6 sn hi ae = -0+2u .» neglecting higher order terms. 8.9 = 2 lineari i ar = 2u » linearized equation Select X(t) = 6 Xq(t) = X10) Xai) = - 04 2u =~ X(t) + 2u Thus the state model is, ae) FO 1PM) 0) fa] ol baa]? 01 0 elModern Control Theory State Space Representation [sI- A] = Adj [sl- A] = 1 -ay = s?41 [slay? = : s?41 ZSR = LNe{s) BUS) where Ups) = 2 = ut =u {$ +Srct 8 st41 Ast +A 4 Bst45C = 0.04 ie A+B=0,C=0,A = 0.04, B =~ 0.04. } a um 2048 |. 9.04 - 0.04 cos t Ss? 41 Qt) = X,(t) = 0.04 ~ 0.04 cas t w= 0.02 Q7 — Choosing appropriate physical variables as state variables, obtain the state model for the electric circuit shown in Fig. 2. (JanJ/Feb.-2006, 8 Marks) Ans.: Refer example 2.29.Modern Control Theory P-6 State Space Representation as Ans. : ag Ans, : Q.10 Ans. : Y For the transfer function ~—= = —-— - (s+1)"(s+4) Obtain the state model in 1) Phase variable canonical form ii) Jordan canonical form (JanJ/Feb.-2006, 8 Marks) Refer example 2.30 Fig. 3 shows the block diagram of a speed control system with state variable feedback. The drive motor is an armature controlled de motor with armature resistance R,, armature inductance L.,, motor torque constant Kp, inertia referred to motor shaft J, viscous friction coefficient referred ta Hie motor shaft B, back emf constant K, and tachometer K,. The applied armature voltage is controlled by a three phase full-converler. ¢, is control voltage, ¢, is armature voltage, e, is the reference voltage corresponding to the desired speed. Taking X, = « (speed) and X,=i, (armature current) as the state variables, u =e, as the input, and y = was the output, derive a state variable mode for the feedback systei. Quly/Aug,-2006, 6 Marks) Fig. 3 Refer example 2.31 For a RLC network shown in Fig. 4, write the state ntadel in mairix notation choosing XD = WL and Xoft) = uclt) where X,(t) and Xp(1) are state cariables, oc(t) is output, v(t) is input. Quly/Aug.-2006, 8 Marks) Refer example 2.1Modern Control Theory P-7 State Space Representation an Qi2 Ans. = Q.13 Ans. : a4 Ans. : Y(s) For a transfer function given by G(s) = z 5 write the state model in the i) Phase s? +3542 variable form ii) Diagonal form. Quly/Aug-2006, 8 Marks) Refer example 2.32. State the properties of Jordan matrix. (July/Aug.-2006, 6 Marks) Refer section 2.3.1. Obtain the state space representation model for the following electrical circuit in Fig. 5. Given R = 1 MQ and C = 1 ye. (JanJFeb.-2007, 10 Marks) R R eT — c uit) c= vit) _ [| Jt. Fig. 5 Refer example 2.22. Obtain the state space representation of the following system and draw ifs phase variable diagram : ¥+6Y +117 +6r =6u QanJFeb.- 2007, 10 Marks; July/Aug.-2007, 6 Marks) Taking Laplace transform of both sides and neglecting initial conditions we get, 6 UG) "53 46s? +11s+6 Then refer example 2.15, Qi5 Ans. : Obtain the state model of the electrical network shown in Fig. 6 selecting 'V', ‘i,’ and "ig! as state variables and voltage across R, and current |, through Ry are the output variables Y, and Y>. Quly/Aug.-2007, § Marks) Ry Egit) Fig. 6 Refer example 2.19 for the procedure. The direction of iy is opposite in this example, Thus the matrices A and B are,Modern Control Theory P-8 State Space Representation Ry 1 1 --§ 0 4 — uy R L Ly - a2 , 1) pe a i a Bel, i 1 ce ° 0 There are two outputs, Y, = Epft) = igR = RX, Yo C= QAG For the network shown in Fig. 7, choosing ij(t) = X,(0 and i,t) = X,0) as state variables, obtain the state equation and output equation in vector matrix form, QjanJFeb.-2008, B Marks) it) —igtt) Tah | yity Ans.: Refer example 2.33. Qi7y Obtain the state model in phase variable form and write the block diagram for the system represented by, (Qan/Feb.-2008, 6 Marks) ay od dy S68 411 + oy = 3dt. are Oe ae 14 Ans.: Refer example 2.16, Q.18 For the following transfer function obtain the state model in canonical form: Gan/Feb.-2008, 8 Marks) ¥(s) 6 so 46s? +115 +6 Ans.: Refer example 2.15. G(s) = Qo00ai Q2 Ans .: as Ans. : Q4 Ans. : as Matrix Algebra and Derivation of Transfer Function Determine the transfer matrix for the systemt QanJFeb.-2005, 10 Marks) x) 3 MGT # 8]fu) fut -1 ps | © |-2 offxs }"|-5 of fu} ‘Lye [8 1 fixe X2 Refer example 3.20. Find the transformation matrix P that transforms the matrix A into diagonal or Jordan 41 -2 form where A=|1 0 2]. 1-1 3 (Jan.JFeb.-2005, 10 Marks) Refer example 3.15. What are genralized eigen vectors? How are they determined? Quly/Aug.-2005, 5 Marks) Refer section 3.10. Conver! the following state model into canonical form Quly/Aug.-2005, 8 Marks) kell A)x, oj “13 -6 “le Y={1 ox Refer example 3.21. Convert the following square matrix A into Jordan canonical form using a suitable non-singular transformation matrix P. Quly/Aug.-2005, 7 Marks) o 1 a A=|]0 01 4+ -9 -6 Refer example 3.22. (P-9)Matrix Algebra & Derivation Modern Control Theory P-10 of Transfer Function a6 Consider the matrix 22 3 Az={1 1 1 1 3 -1 a7 Ans. : as Ans. = as Q.10 i) Find the eigen values and eigen vectors of A ii) Write the medal matrix itt) Show that the modal matrix indeed diagonatizes A. (Jan/Feb.-2006, 12 Marks) Refer example 3.23. Given the state model X = AX — Bu, y = CX { o 1 0 0 where A=| 0 0 1),B=|0) andC=[1 0 0] -1 -2 -3| 1 *) Gp, . _ ¥(s) ne oat i) Simulate and find the transfer function Ue) using Mason’s gain formula. ii) Determine the transfer fienction from the state model formulation. (Jan.fFeb.-2006, 7 Marks) Refer example 3.24. 1 22 3 The vector | 2| is an eigen vector of A=| 2 1 -G1 Find the eigen value of A -1 -1 -2 9 corresponding to the vector given. (uly/Aug.-2006, § Marks) Refer example 3.25. Show that the characteristic equation and eigen valves of a sytem matrix are wrvariant under linear transformation. Guly!Aug.-2006, 8 Marks) Refer section 3.9. Determine the transfer function of the given state vector differential equation below : Xi} fo t o]fm l Xo}=|0 00 1 [/X>j+}-t}e Xs\ Lo -4 -5] [x3] [5 Guly/Aug.-2007, 8 Marks)Matrix Algebra & Derivation Medern Control Theory P-12 of Transfer Function =f) _8t5 ~ Ys s(s+4) (stl) s(s#4) (541) = dy 848) 3 ~ 8s s(s+4)(s+1) s(s+4)(s+1) -lliust5 oy 5 “os s(s+4)(s+1) s(s+4) (s+1) =! ~ Ss 1 1 (s+4) (s+1) Ss (s+4d)@+l) s(erd) eri Tr = si+5st4-s sitds+4 “$44 64) 64d 41) an Obtain eigen values, eigen vectors and state model in canonical form for a system described by the fallowing state model : Quly/Aug.-2007, 10 Marks) xy Oo 4 Offs; 1 My fo f=] 3 0 2] lxg]+/O)umdy=fl 0 O)f x2 i] |[-12 -7 -6}[x3] [2 x3 Ans.: Refer example 3.14 for obtaining M. Once M is obtained find M“ and the new matrices B = M'B and C=CM. 1201 4.5 2.5 1 6.5 M=l-1 -4 -3], M’=|-3 -2 -1 5 ,Esfl 2 1. -l 13 25 15 1 45 Qi2 Obtain the model matrix for the matrix given below : (Jan/Feb.-2008, 4 Marks) o 1 0 Az/]3 0 2 -12 7-6 Ans. : Refer example 3.14.ai Ans. = Q2 Ans. : a3 Ans. : aa Solution of State Equations For a system represented by X = AX the response to one set of initial conditions is X(t) endl 2b = * it and another set of initial condition is X(t) = [‘ nt a o | Determine inatix A and the state transition matrix O(1). (Jan/Feb.-2005, 10 Marks) Refer example 4.19. Mention the conditions for complete controllability and complete observability of continous time systems. Using these, explain the principle of duality between controllability and observability. (Jan JFeb.-2005,Jan./Feb,-2008 10 Marks,July/Aug.-2007 6 Marks) Refer sections 4.12 and 4.13. Use contrallability and observability matrices to determine whether the system represented by the flow graph shown in Fig. 1 is completely controllable and completely observable. Fig. 1 (Janffeb.-2005, 10 Marks) Refer example 4.33. Given the time invariant system )-f SPE) ee off] (P14)Modern Control Theory P.15 Solution of State Equations Ans. : as Qs Ans. : ar an that u(t) =e! and y(t) = 2-ate|, find X4(1) and Xj(0. Find also X,(0) and X3(0), What happens if « = 0? Qan/Feb.-2005,10 Marks) Refer example 4.34. What is a state transition matrix ? List the properties of state transition matrix. (Quly/Aug.-2005, July/Aug,-2007,Jan.Feb.-2008,6 Marks; July/Aug.-2006, 5 Marks, JanJ/Feb.-2007, 10 Marks) Refer sections 4.3 and 4.4 Given the state model of a system : . Pa] lla * bel] | 5 J Y=f1 OX i) with initial conditions X(0) Determine : i) The state transition matrix. ii) The state transition equation X(t) and output Y(t) for an unit step input. iti) Inverse state transition matrix. (July/Aug.-2005, 14 Marks) Refer example 4.35. Explain the concept of controllability and observability. Quly/Aug.-2005, July/Aug.-2006, 6 Marks) Refer section 4.11. Determine the controllability and observability of the following state model. a1 0 I X=/]o 0 1|xsfolu ~6 -ll -6 1 y = [10 5 1X Quly/Aug.-2005, 8 Marks)Modern Control Theory P-16 Solution of State Equations Ans, : as Ans. : ato Ans, : ann Ans. : O12 Ans. : Refer example 4.36. A system represented by following state model is controllable but not observable. Show that the non-observability is due to a pole-zero cancellation in Clsl-AT"’ "B, . o 1 0 0 X=|0 O 11X+]0)u -— -ll - 1 Y = fi 1 ox (uly/Aug.-2005, 6 Marks) Refer example 4.37. List atleast three important properties of the state transition matrix. (JanJFeb.-2006, 3 Marks) Refer section 4.4. Consider the homogeneous equation Ks AX, where A is a 3 x 3 matrix. The following three solution for three different initial conditions are available, et eu Dee et |, |-2e | | bet 2e7 0 0 i) Identify the initial conditions ii) Find the state transition matrix iii) Hence or otherwise find the system matrix A. JanJFeb.-2006, 10 Marks} Refer example 4.38. Obtain the time response y(t) of the system given below by first transforming the state model into a ‘Canonical model’. 0 1 0 0 X=|0 0 1) X40) uY=l 0 oO] x ~6 -l1 -6 2 u is a unit step function and X" (0) = {0 0 2] QanJFeb.-2006, 12 Marks) Refer example 4.39.Modern Control Theory P-48 Solution of State Equations Qiz Obtain the state transition matrix using : i) Laplace transformation method and ii) Cayley-Hamilton method. For the system describe by, X(t) = [* ' }xo 4a 4 JanJFeb.-2007, 10 Marks) Ans.: Refer example 4.43. Q18 State the conditions for completely controtability and complete obseronbility. Determine the state controllability and observability of the system described by, . xy Oo 1 Ojfx] fo Xof=|0 0 F |fxy |+]0) fd ty] [6-1 -6lies] [1 3 “1 ye[4 5 us Lvs (Jan./Feb.-2007, 10 Marks) Ans. : Refer section 4.11 and example, 4.36 for the procedure. The system is completely controllable and completely observable. Q19 Obtain the observable phase variable state model of Y 33 +bs? bos TAS = Rose bis’ bas °b5. Demo the signal flow grap of Tt) sd eaye2 saye+ay Quly/Aug-2007, 8 Marks) Ans.: Refer example 4.46, TF. = Tes) = Q.20 Obtain the controllable phase variable form of the transfer function bos? +bys? +bgs+b tas? +agsta3 Quly/Aug.-2007, 6 Marks) Ans.: Refer example 4.47. Q.21 Compute efor the given matrix : 7 3 a-[ hae =| oa-(2 ‘| 0 6 “a 0 ~© 61 Quiy/Aug.-2007, 6 Marks)Modern Contro! Theory P.20 Solution of State Equations Ans. : Refer examples 4.27 and 4.28. Q.26 = Determine the complete time response of the system given by 0 kw et” *\xen where x0) =| an Yin) = [1-1] x0 we 4 4 (JanJFeb.-2008, 12 Marks) Ans.: Refer example 47 Q.27 Determine the controllability and observability using Kalman's test for the system described by, fo io fo A= 0 0 1] B=/0] C=ft0 0 oO] [0-2 -3 1 (Jan./Feb.-2008, 10 Marks) Ans.: Refer example 4.36 for the procedure and the given system is completely controllable and observable. qQaoModern Gontrol Theory P-23 Pole Placement Technique a Ans. : Qiz oi 0 0] Consider the system defined by x = Ax + Bu, where A=|0 0 1 | B= t By -1 -5 -6 1 using state feedback contro! u = Kx, it is desired to have the closed loop poles at s=-24)4 and s = -10. Determine the state feedback gain matrix “K" by any one method, Qan/Feb,-2008, § Marks) Refer example 5.10 nn) [fo Consider the system x=Ax+Bu and Y= Cx, where A=|0 0 1] B= [0 6 -ll -6 i and C= [1 0 0]. Determine the observer gain matrix by the use of : i) The direct substitution method. ii) Ackermann’s formula. Assume that the desired Eigen values of the observer gain matrix are My =— 24 [34641 and 2 == 2-j34641, 3 =-5. QanJjFeb.-2008, 8 Marks) Refer example 5.11. Qo00(6) Controllers at Ans. : Q2 Ans. : Qs Ans. : a4 Ans. : Qs Ans. : a6 Ans. : Explain effecis of a Pl controller on the static and dynamic response of a system. Quly/Aug.-2005, 5 Marks) Refer section 6.9. Consider # typical second order, type one system with unity feedback, being controlled by a PD controller and show that i) Damping increase with PD control. (July/Aug.-2005, 4 Marks) Refer section 6.14. A temperature control system has the block diagram given in Fig.1. The input signal is a voliage and represenis the desired temperature @. Find the steady-state error of the system when Q, is a unit step function and i) D(s) = 1 i D(y=14 ot iii) Dis) = 1+0.3 s. What is the effect of the integral term in the PI controller and the derivative term in PD controller on the steady state errar ? (uly/Aug.-2007, 8 Marks) Fig. 1 Refer example 6.16, What is a controlier ? Explain P, I, PI and PID controllers. (an/Feb.-2007, 10 Marks) Refer sections 6.1, 6.5, 6.6, 6.9 and 6.11 Define controller. Explain properties of P, Pi and PID controllers with the help of block diagram, (Quly/Aug.-2007, 8 Marks) Refer sections 6.1, 6.5, 6.9 and 6.11 Define controller. Explain PD, Pl and PID controllers. What are the advantages of PID controller ? QanJFeb.2008, 6 Marks) Refer sections 6.1, 6.10, 6.9 and 6.11 (P - 24) goaNonlinear Systems Qt Ans. : az Ans. : a3 Ans. : a4 as Ans, : as Ans, : Explain the following behaviour of non-linear systems : (i) Frequency amplitude dependence . (ii) Multivalued responses and jump resonances. (Jan/Feb.-2005, 10 Marks) Refer section 7.2. Discuss the basic features of the following non-linearities i) Non-linear friction ii) On-off controllers iit) Back lash Gan JFeb.-2006, 9 Marks) Refer sections 7.4.5 and 7.4.6. What are inherent nonlinearities ? Explain any three of them{July/Aug.-2006, 6 Marks) Refer section 7.4. Sketch the following non-linearities : i) Ideal relay ii) Relay with dead zone iii) Relay with dead zone and hysterisis iv) Relay with hysterisis v) Dead zone. (Quly/Aug.-2006, 4 Marks} Discuss the basic features of the Backlash non-linearities with suitable figures. Quly/Aug.-2007, 4 Marks) Refor section 7.4.6. Explain any three non-linearities in control systems. (anJFeb.-2008, 6 Marks} Refer section 7.4. o00 (P - 25)Qt Ans. : Q2 Ans. : Q3 Ans. = Qa Ans. : as Ans, : Phase Plane Method Determine the kind of singularity for each of the following differential equations. (i) 743i 2y=0 tt ot (i) y-Sys+17y= 34 (JanJ/Feb.-2005, 10 Marks) Refer example 8.1. What is a phase plane plot? Describe delta method or any other method of drawing phase plane trajectories. (Jan./Feb.-2005, 10 Marks) Refer sections 8.1 and 8.5. With reference to non-linear system explain : (i) Jump resonance (ti) Limit cycles (July/Aug.-2005, 6 Marks) Refer sections 8.3 and 8.2. What are singular points? Explain the classification of singular points based on the location of eigen values of the system. Quly/Aug,-2005; Jan./Feb.-2008, 8 Marks) Refer section 8.9, Tig. 1 shows phase portraits for type - 0 systems. Classify them into the categories. Stable focus, stable node, satdle paint and so-on. (See Fig. 1 on next page.) (JanJ/Feb.-2006;July/Aug.-2007, 6 Marks) Stable focus Fig. 2 Stable focus (P - 26)Modern Control Theory P-29 Phase Plane Method Unstable node 2 Wak Fig. 6 Unstable node Saddle point as Ans. : a7 Ans. = as Ans. : Fig. 7 Saddle point Explain the concept of jump resanance with a suitable example. (anJFeb.-2006, 5 Marks) Refer section 8.3, Explain the delta method of constructing phase trajectories. (anJFeb.-2008, 7 Marks) Refer section 8.11. Using isocline method, draw the phase trajectory for the system. 2, x 06% er =0 at? dt with x = 1 and & = 0 as initial condition, at (JanJFeb.-2006, 8 Marks) Refer section 8.4,Modern Control Theory P-30 Phase Plane Method Qs a0 Ans. : Qi Ans, : Q12 Q13 Ans. : a14 Ans. : Qs Ans. : “ * A Hinear second order servo is described by the equation C+2Guy,C+orC = 0 where €= 0.15, w, = T rad/sec, C(O) = 1.5 and C= 0. Determine the singular point. Construct the phase trajectory, using the method of isoclines. (July/Aug,-2006, 10 Marks) Refer example 8.2. Define singular point on a phase plane, Explain different types of singular points Quly/Aug.-2006, 10 Marks) Refer section 8.9, What are singular points? Explain different singular points adopted in non-linear control systems. anJFeb.-2007, § Marks) Refer section 8,9, Find out singular points for the following systems : ) X+05x42x=0 woe fi) y+3y +2y=0 iii) y+3y -10=0 (Jan./Feb.-2007, 12 Marks) Refer example 8.3. Draw the phase-plane trajectory for the following equation using Isocline method ; x +2Emx+e? x=0 Given, §=05, w=1, Initial point (0, 6). (JanJFeb.-2007, 12 Marks) Refer example 8.5. Discuss the basic features of the jmp resonance non-linearities with suitable figures : Jump resonance. Quly/Aug.-2007, § Marks) Refer section 8.3. Explain the construction of a phase trajectory by Delta method. (janJ/Feb.-2008, 6 Marks) Refer section 8.5. agg(9) Liapunov's Stability Analysis Qi Ans. : Q3 a4 Ans. = as Ans. = State and explain Liapunov's theorems on (i) Asymptotic stability {ii) Global asymptotic stability and (iif) Instability. QanJFeb.-2005; July/Aug.-2006; Jan./Feb.-2008, 10 Marks) Refer sections 9.3, 9.4 and 9.5. Use Krasovskii's theorem to show that the equilibrium state x = 0 of the system described by My =-3xy 4x2 Ey =x —X2—x} is asymptotically state in the large. (an.JFeb.-2005, 10 Marks) Refer example 9.4. Define : (i) Stability, (ii) Asymtotic stability (iii) Asymptotic stability in the large. (uly/Aug.-2005, 5 Marks, July/Aug.-2007, 6 Marks) Refer sections 9.29.3, 9.4 and 9.5. Investigate the stability of the following non-linear system using direct. method of Liapimov. Xy =Xx2 ¥y BN APN (uly/Aug.-2005, 5 Marks) Refer example 9.5. A second order system is represented by, 1 t=Ax where A=| 1 Assuming matrix Q fo be identity matrix, solve for matrix P in the equation T, . 1 ot A'P + PA = — Q. Lise Liapunov theorem and determine the stability of the origin of the system. Write the Liapunov function V(x). Quly/Aug.-2005, 10 Marks) Refer example 9.6. (P - 31)Modern Control Theory P-32 Liapunov's Stability Analysis a6 Ans. ar a8 Ans. : a9 Ans. Q.10 Ans, : ann Using Liapunovs direct method, find the range of K to guarantee stability of the systent showin in Fig. 1 (Jan/Feb.-2006, 14 Marks) Refer example 9.11 Choose an appropriate Liapunov function and check the stability of the equilibrium state of the system described by . X= Xy ia i GanJFeb.-2006, 6 Marks) Refer example 9.5. Determine whether or not following quadratic form is positive definite : Q(x, 42) = 10x? +403 4x3 + Qeyxg —2egxg-4epx3 GanJ/Feb.-2007, 10 Marks} Refer example 9.10. Explain with an example - i} Liapunow main stability thearem ii) Liapunop second method and iti) Krasouskii's theorem. GanJFeb.-2007, 10 Marks) Refer sections 9.10, 9.11 and 9.13. Find the Liapunov function for the system : ool xw=[ ks [cl -1 (JanJFeb.-2007, 8 Marks) Refer example 9.6. Explain Liapunov’s stability criterion. Guly/Aug,-2007, 6 Marks) Refer section 9.11Modern Control Theory P.~33 Liapunov's Stability Analysis. Q.12 Q43 Ans. : a4 Ans. : Consider the system with differential equation. e+kerkic3 +e 20 Examine the stability by Liqpunov's method, given that K > 0 and Ky >0. Quly/Aug.-2007, 6 Marks) Refer example 9.8. Examine the stability of the system described by the fallowing equation by Krasovdskit's, XY SOR) x2 = xy ty -} Quly/Aug.-2007, 8 Marks) Refer example 9.9. A system is described by the following equation : y= Ax Where A=| x¥=AX Here A= 1 —-4f Assuming matrix Q to be the identify matrix, solve for matrix P and comment on the stability of the system using the equation A P+ PA=-Q. (Jan/Feb.- 2008, 10 Marks) Refer example 9.7. Qo00Contents Pee eee (eee ee ee ie aed Pe ee RU Mare ee PUL S eke ee ee en ae femme seared + Solution of state equation, State transition matrix and its properties, Computation using Laplace transformation, Power series method, Cayley- Harnilion method, Concept of controllability and observability, Methods of determining the same, a eee en Pee ee neo ced placement, State regulator design, And design of state observer, Controllers- F, Pl, PID. Ne eld eee Le Be i «Phase plane method, Singular points, Stabiity of non-linear system, Limit eyeles, Construction of aces Liapunow stability eriteria, Liapunay functions, Direct method of Linpunov and the linear system, Hurwite criterion and Liapunov's direct method, Construction of Liapunav functions for nomlinear ee ee eed aieian [hele] a iear-a0 le) 3} Price INR 225, Sct ISBN 978-81-8431-506-6 | \) Technical Publications pune | | | | | it Residency, 412 Shaniwar Poth, Pune - 411030, M.S., India. Eee eC Oa Cenc I | | | sisa list Visit us at: Www.vtubooks.com

Modern Control Theory

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Modern Control Theory

Uploaded by

Copyright:

Available Formats

You might also like