1 LECTURE NOTE

2 LECTURER: TOLA O. J
3 COURSE: CONTROL ENGINEERING III CODE: EEE 512 UNIT:2 LEVEL: 500
4 INTRODUCTION:

5 Classical control embodies methods such as root locus, Bode, Nyquist and Routh- Hurwitz. These
6 methods have in common the use of transfer functions in the complex frequency(s) domain,
7 emphasis on the use of graphical techniques, the use of feedback and the use of simplifying
8 assumptions to approximate the time response. The major limitation of classical control method
9 was the use of single-input, single output (SISO) methods.
10 Modern control theory represents a new approach to the analysis and design of complex control
11 systems. The complex control systems may have multiple inputs and multiple outputs and may be
12 time varying. This new approach is based on state-space method.
13 .
14 Advantage modern control theory over the classical control theory:
15 It is applicable to time-variant (ie time-varying) and time-invariant, linear or nonlinear
16 It facilitates the inclusion of initial condition in the design
17 More accurate than the transfer function
18 Analysis of multi input and multi output systems are easy
19 Gives information about controllability
20 It is applicable to all dynamic systems
21 It facilitates the design of system for optimal conditions with respect to given performance indices
22 Disadvantage
23 Complex techniques
24 Many computations are required
25
26
27
28
29
30
31
32
33
34

CONTROL ENGINEERING III EEE 512 ENGR. DR. O. J. TOLA

35 Linear and Nonlinear Systems
Nonlinear vs. Linear Systems
S/No Nonlinear Systems Linear Systems
1 More realistic Approximation to reality
Usually simpler to analyze and
2 Approximation to reality design
3 Can have multiple equilibrium points Only single equilibrium point
System stability depends on Initial condition System Stability is independent of
4 (IC) IC
5 Bifurcations (number of equilibrium No bifurcation
points and their stability nature can vary
with parameter values)
6 Chaos (very small difference in I.C. can No chaos
lead to large difference in output as
time increases. That’s why predicting
weather for a long time is erroneous!)
36 Basic Concepts of State Space Model
37 A state space representation of a system is a mathematical model where the inputs, outputs and
38 state (internal variables) are related by first order differential equations.
39
40 The following basic terminology involved in this module.

41 State: it is a group of variables, which summarizes the history of the system in order to predict the
42 future values.
43 State Variables: the smallest set of variable that determines the state of the system
44 The number of the state variables required is equal to the number of the storage elements present
45 in the system.
46 Examples - current flowing through inductor, voltage across capacitor

47 State Vector: it is the vector which contains the state variable as element.
48 State-space equations or simply state equations, is a description in the time domain which may be
49 applied to a very wide category of systems such as linear and nonlinear systems, time-invariant
50 and time-vary systems, systems with nonzero initial conditions and others.
51 The term state of a system refers to the past, present and future of the system. From mathematical
52 point of view, the state of a system is expressed by its state variables. Usually, a system is described
53 by a finite number of state variables, which are designed by x1 (t ), x2 (t ),..., xn (t ) and are define as
54 follows:

CONTROL ENGINEERING III EEE 512 ENGR. DR. O. J. TOLA

55
56
57
58
59
60 State-space description of linear systems
61 State-space description of linear systems

62
u (m ) Linear y(p)

63 
System
64
x (n )
65

66
 y1 ( t ) 
67
 u1 ( t )   x1 ( t )   y (t)
 x (t )
 u (t)  y(p)   
68 2

u (m)   2  x (n )   2   yi (t ) 
69
 u i (t)   x i (t)   
70
    y (
 p  t )
u ( t ) x
 n  ( t )
71  m 
72
Input vector State vector Output vector
73

74 In which x(t) is the n-dimensional state vector.

75 Whose n scalar components are called state variables. Similarly, the m-dimensional input vector
76 and p-dimensional output vector.
77 STATE EQUATIONS
78 A state-space representation for a linear time-invariant system has the general form

 x (t )  Ax(t )  Bu (t )

79  y (t )  Cx(t )  Du (t )
 x(t )  0
 0

80 Finally, for a specified initial time t0, the initial state x(t0)=x0 is the specified, constant n-
81 dimensional vector.
82 We can represent a general state space system as a Block Diagram

CONTROL ENGINEERING III EEE 512 ENGR. DR. O. J. TOLA

 83
84 State-Equation Block Diagram
85 Example
86 A system is described by the differential equation

d3y d2y dy
87 3
 5 2  9  10 y  6u (t )
dt dt dt
88 Where y is the output and u is the input. Obtain a state space representation of the system
89 Solution
90 Step I: write the differential equation so that the highest derivative is on the left hand side
91 
y  6u (t )  5 
y  9 y  10 y

92 Step II: choose the state variable as:

93 x1  y , x2  y , x3  
y

94 Step III: Find the derivative of each state variable

95 x1  y  x2

96 x2  
y  x3

97 x3  
y  6u (t )  5 x3  9 x2  10 x1

98 The las equation can be re-arranged to give

99 x3  10 x1  9 x2  5 x3  6u (t )

100 Step IV: express the above equation in concise form

 x1  0 1 0   x1   0 
101
 x   0 0 1  x2    0  u 
 2 
 x3   10  9  5  x3   6 

102 In matrix notation it is x  Ax  Bu

CONTROL ENGINEERING III EEE 512 ENGR. DR. O. J. TOLA

 0 1 0 0   x1 
103

Where A  0 0 1 , B  0  and x   x2 
  
 10  9  5 6   x3 

104 Recall that the output is y  x1 in Step II

 x1 
105 y  1 0 0  x2 
 x3 

106 Example 2
107 Consider a system described by the following pair of differential equations

d 2 d
108 J 2
f  kmia  0 and
dt dt
d di
109 km  Ria  La a  V (t )
dt dt
110 Obtain a state-space representation of the system
111 Solution

112 Step I: Rewrite the pair of differential equation so that the highest derivative are on
113 the LHS
km f
114   ia  
J J
dia 1 R k
115  V (t )  a ia  m 
dt La La La

116 Step II: Choose the state variables as

117 x1   , x2  , x3  ia

118 Step III: find the derivative of the variables and express your results
119 x1    x2

k f
120 x2    m x3  x2
J J
dia 1 R k
121 x3   V (t )  a x3  m x2
dt La La La

CONTROL ENGINEERING III EEE 512 ENGR. DR. O. J. TOLA

122 Step IV: now the equations are expressed in closed matrix
 
0 1  
0 
 x1     x1   0 
 x   0  f k m     0  V (t )
123
   x  
J    
2 2
J
 x3     x3 
0  km  Ra  1
  La 
La La 

124 Example 3

125
126 Consider the network shown in Figure above, with initial conditions i L (0)  I 0 and
127 vc (0)  v0 . Determine the state equations of the network.

128 Solution
129 The loop equation
130 Hint: it is often convenient to associate state variables with the energy storage elements in the
131 network, namely, the capacitors and inductors.
132

133 Example
134 A simplified dc servomotor model is shown in Figure .
135 The input is the armature voltage v(t) and the output is the motor shaft angular displacement θ(t).
136 The constant parameters are armature circuit inductance and resistance L and R, respectively, and
137 motor shaft polar inertia and rotational viscous damping coefficient J and b, respectively. The
138 intermediate variables are armature current i(t), motor torque τ (t), and motor shaft angular
139 velocity ω(t) = θ(t) ˙. we have simplified the model; we ignore back emf voltage, and there is no
140 gear ratio or load inertia included. Derive a valid state-space realization of the system
141

CONTROL ENGINEERING III EEE 512 ENGR. DR. O. J. TOLA

142
143 Figure : DC servomotor model
144 Solution
145 This system in three steps: circuit model, electromechanical coupling, and rotational mechanical
146 model. For the circuit model, Kirchhoff’s voltage law yields a first-order differential
147 equation relating the armature current to the armature voltage

148
149 Motor torque is modeled as being proportional to the armature current, so the electromechanical
150 coupling equation is

151
152 For the rotational mechanical model, Euler’s rotational law results in the following second-order
153 differential equation relating the motor shaft angle θ(t) to the input torque

154
155 we use a transfer-function description to derive the state equations. We have

156

CONTROL ENGINEERING III EEE 512 ENGR. DR. O. J. TOLA

157

158
159 State-Space Description
160 we specify the state variables and derive the coefficient matrices A, B, C, and D. The results then
161 are written in matrix-vector form. Since we have a third-order differential equation, the state-
162 space dimension is n = 3, and thus we need to define three state variables xi(t), i = 1, 2, 3. We
163 choose

164

165
166 The state differential equation is

CONTROL ENGINEERING III EEE 512 ENGR. DR. O. J. TOLA

167
168 from which we identify coefficient matrices A and B:

169
170 The algebraic output equation is

171
172 from which we identify coefficient matrices C and D:

173
174 This is a three-dimensional single-input, single-output system with m = 1
175 input, p = 1 output, and n = 3 states.
176

177 ASSIGNMENT (I)

178 (1) For the following systems described by the given transfer functions, derive valid state-space
179 realizations (define the state variables and derive the coefficient matrices A, B, C, and D).
180

CONTROL ENGINEERING III EEE 512 ENGR. DR. O. J. TOLA

 Y (s) 1
181 (a) G ( s )   2
U ( s) s  2s  6
Y (s) s3
182 (b) G ( s )   2
U ( s) s  2s  6
Y (s) 10
183 (c) G ( s )   3
U ( s ) s  4 s 2  8s  6
Y (s) s 2  4s  6
184 (d) G ( s )   4
U ( s ) s  10s 3  11s 2  44 s  66

185 (2) Given the following differential equations (or systems of differential equations), derive valid
186 state-space realizations (define the state variables and derive the coefficient matrices A, B, C, and
187 D).
188 (a) y (t )  2 y (t )  u (t )

189 (b) 
y (t )  3 y (t )  10 y  u (t )

190 (c ) 
y (t )  2 
y (t )  3 y (t )  5 y (t )  u (t )


y1 (t )  5 y1 (t )  10[ y2 (t )  y1 (t )]  u1 (t )
191 (d )
2 
y2 (t )  y 2 (t )  10[ y2 (t )  y1 (t )]  u2 (t )

192

193 Transfer function from state space models

194 We can obtain the state space equation from a given block diagram of a system
195

196

197 Given

 x (t )  Ax(t )  Bu (t )
198 
 y (t )  Cx(t )  Du (t )
199

Y ( s ) C ( sI  A)1 B
200 G (s)   D
U (s) ( sI  A)

201

202

203 ASSIGNMENT (II)

CONTROL ENGINEERING III EEE 512 ENGR. DR. O. J. TOLA

204 Find the transfer function of the system represented by the following state equations.

0 1 0   1
205 (i) x  0 0 1  x   1 u and y  1 0 0 x
0  4  6   5 
0 1 0 
206 (ii) x    x    u and y  1 0 x
 4  2   4
 2 0 1  1 
207 (iii) x  1  2 0  x  0  u and y   2 1  1 x  1 u
 
1 1  1 1 

208

209 EIGENVALUES, AND EIGENVECTORS

210 Characteristic equations play an important role in the study of linear
211 systems. They can be defined with respect to differential equations, transfer
212 functions, or state equations.
213 Characteristic Equation from State Equations
214 From the state-variable approach, we can write
adj ( sI  A) B
215 G (s)  C D
( sI  A)

C[adj ( sI  A)]B  sI  A D
216 G (s) 
sI  A

217 Setting the denominator of the transfer-function matrix G(s) to zero, we

218 get the characteristic equation

219 sI  A

220 Example
221 Consider the differential equation

d 3 y (t ) d 2 y (t ) dy(t )
222 5   2 y (t )  u (t )
dt 3 dt 2 dt
223 The characteristic equation of A is

s 1 0
224 sI  A  0 s  1  s 3  5s 2  s  2  0
2 1 s5

CONTROL ENGINEERING III EEE 512 ENGR. DR. O. J. TOLA

225 Eigenvalues
226 The roots of the characteristic equation are often referred to as the
227 eigenvalues of the matrix A.
228 The eigenvalues or the roots of the characteristic equation
229 of the matrix A are obtained by solving
230 for the roots of s 3  5s 2  s  2  0
231 The results are s  0.06047  j 0.63738, s  0.06047  j 0.63738, s  4.87906

232 Eigenvectors
233 Any nonzero vector P that satisfies the matrix equation

234  i I  A Pi  0
235 Where i i  1, 2, .. ..., n, denote the ith eigenvalue of A, called the
236 eigenvector of A associated with the eigenvalue i If A has distinct
237 eigenvalues, the eigenvectors can be solved directly from  i I  A  Pi  0

238 EXAMPLE Consider that the state equation that has the
239 coefficient matrices

1  1  1
240 A  B  D0
0  1 1
241 The characteristic equation of A is

242 sI  A  s 2  1

243 The eigenvalues are 1  1 and 2  1 Let the

244 eigenvectors be written as

 P11   P12 
245 P1    P2   
 P21   P22 

246 Substituting 1 and P1 in  i I  A Pi  0

0 1   P11  0 
247 0 2   P   0 
   21   

248 Thus, P21 = 0, and P11 is arbitrary, which in this case can
249 be set equal to 1.
250 Similarly, for 2  1

CONTROL ENGINEERING III EEE 512 ENGR. DR. O. J. TOLA

  2 1  P12  0 
251  0 0   P   0 
   22   

252 which leads to

253 2 P12  P22  0

254 The last equation has two unknowns, which means that
255 one can be set arbitrarily. Let P12 = 1, then P22 = 2. The
256 eigenvectors are

1  1 
257 P1    P2   
0 2
258

259 EXAMPLE Given the matrix

0 6  5 
260 A  1 0 2 
3 2 4 

261 The eigenvalues of A are 1  2, 2  3  1

262 Thus, A is
263 a second-order eigenvalue at 1. The eigenvector that is
264 associated with λ1 = 2 is determined

 2 6 5   P11 
265  1 I  A   1 2  2   P21   0

 3  2  2  P31 

266 Because there are only two independent equations, we arbitrarily set P11 = 2, and we have P21 =-1
267 and P31 = -2. Thus,

2 
268 P1   1 
 2 

269 For 2  1

CONTROL ENGINEERING III EEE 512 ENGR. DR. O. J. TOLA

  
1 
 
3
270 P2    
 7
 5
 
 7

271 For 2  1

 
 1 
 
22
272 P3    
 49 
 46 
 
 49 
273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

CONTROL ENGINEERING III EEE 512 ENGR. DR. O. J. TOLA