1 Chapter 3: Modeling in the Time Domain

Chapter 3
Modeling in the Time Domain

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 2 Chapter 3: Modeling in the Time Domain

Chapter Objectives
In this chapter you will learn the following:
• How to find a mathematical model, called a
state-space representation, for a linear, time-
invariantsystem.
• How to convert betweentransfer function
and state-space models
• How to linearize a state-space
representation

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
3 Chapter 3: Modeling in the Time Domain

Introduction
Two approaches are available for the analysis and
design a feedback control systems. The first,
which we beganto study in Chapter 2, is knownas
the classical, or frequency –domain technique.
This approach is based on converting a system’s
differantial equation to a transfer function, thus
generating a mathematical model of the systemthat
algebraically relates a representation of the output
to a representation of the input. Replacing a
differential equation with an algebric equationnot
only simplifies the representation of individual
subsystems but also simplifies modeling
interconnected subsystems.
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 4 Chapter 3: Modeling in the Time Domain
Introduction cont.
• The primary disadvantage of the classical approachis its
limited applicability: it can be applied only to linear, time
invariant systems or systems that can be approximated as
such.
• A major advantage of frequency-domain techniques is that
they rapidly provide stability and transient response
information. Thus, we can immediately see the effect of
varying system parameters until an acceptable design is met.
• With the arrival of space exploration, requirements for
control systems increased in scope. Modeling systems by
using linear, time-invariant differentialequation and
subsequent transfer functions become inadequte.

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
5 Chapter 3: Modeling in the Time Domain
Introduction cont.
• The state-space approach (also referred to as the modern,
or, time-domain approach) is a unified method for
modeling, analyzing, and designing a wide range of
systems. For example, the state-space approach can be
used to reprsent nonlinear systems that have backlash,
saturation , and dead-zone. Also, it can handle,
conveniently, systems with nonzero initial conditions.
Time varying systems canbe represented in state-space.
Many systems do not have just a single input and a single
output. Multiple-input, multiple-output systems can be
compactly representedin state-space with a model similar
in form a compexity to that used for single-input, single-
output systems.

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 6 Chapter 3: Modeling in the Time Domain
Some observation
1. We select a particular subset of all possible systemvariables and
call the variables in this subset state variables
2. For nth-order system, we wite n simultaneous, first order
differantial equations in terms of the state variables. We call this
systemof simultaneous differantial equations state equations.
3. If we know the initial condition of all of the state variables at t0 as
well as the system input for t ≥ t0, we can solve the simultaneous
differential equations for the state variable for t ≥ t0.
4. We algebraically combine the state variables with the system’s
inputs and find all of the othersystem variables t ≥ t0 . We call this
algebraic equation the output equation.
5. We consider the state equations a viable represntation of the
system. We call this representation of the system a state-space
representation.

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 7 Chapter 3: Modeling in the Time Domain
The General State-Space Representation
Now that we have represented a physical network in state space and have a
good idea of the terminology and the concept, let us summarize and
generalize the representation for linear differential equations. First we
formalize some of the definitions that we came across in the last section.
Linear combination: A linear combination of n variables, xi, for i = 1 to n, is
given by the following sum, S:
S = Knxn + Kn-ıxn--1 + ... + K1x1
where each Ki is a constant.
Linear independence: A set of variables is said to be linearly independent if
none of the variables can be written as a linear combination of the others.
For example, given x1, x2, and x3, if x2 = 5x1 + 6x3, then the variables are
not linearly independent, since one of them can be written as a line ar
combination of the other two. Now, what must be true so that one variable
cannot be written as a linear combination of the other variables? Consider
the example K2x2 = K1x1 + K3x3. if no xi = 0, then any xi can be written as a
linear combination of other variables, unless all Ki = 0. Formally, then,
variables xi, for i = 1 to n, are said to be linearly independent if their linear
combination, S, equals zero only if every Ki = 0 and no xi = 0.
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 8 Chapter 3: Modeling in the Time Domain

System variable: Any variable that responds to an input or

initial conditions in a system.
State variables: The smallest set of linearly independent system
variables such that the values of the members of the set at
time to along with known forcing functions completely
determine the value of all system variables for all t ≥ t0.
State vector: A vector whose elements are the state variables.
State space: The n-dimensional space whose axes are the
state variables. This is a new term and is illustrated in Figure
3.3, where the state variables are assumed to be a resistor
voltage, vR, and a capacitor voltage, vC. These variables form
the axes of the state space. A trajectory can be thought of as
being mapped out by the state vector, x(t), for a range of t.
Also shown is the state vector at the particular time t = 4.

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 9 Chapter 3: Modeling in the Time Domain

Figure 3.3
Graphic
representation
of state space
and a state
vector

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 10 Chapter 3: Modeling in the Time Domain
The General State-Space Representation
• The state and output equations can be written in vector-
matrix form if the system is linear
x = A x + B u
y = C x + Du
• For t ≥ t0 and initial conditions x(t0), where
x = state vector
x = derivative of the state vector with respect to time
y = output vector
u = input or control vector
A = system matrix
B = input matrix
C = output matrix
D = feedforward matrix
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
11 Chapter 3: Modeling in the Time Domain

• Equation (3.18) is called the state equation and the

vector x, the state vector, contains the state
variables. Equation (3.18) can be solved for the state
variables, which we demonstrate in Chapter 4.
Equation (3.19) is called the output equation. This
equation is used to calculate any other system
variables. This representation of a system provides
complete knowledge of all variables of the system at
any t > to.
• The choice of state variables for a given system is
not unique. The requirement in choosing the state
variables is that they be linearly independent and that
a minimum number of them be chosen.

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 12 Chapter 3: Modeling in the Time Domain

• As an example, for a linear, time-invariant, second-order

system with a singleinput, v(t), the state equations could
take on the following form:

dx1
= a11 x1 + a12 x2 + b1 v(t )
dt
dx2
= a21 x1 + a22 x2 + b2 v(t )
dt
• where x1 and x2 are the state variables. If there is a single
output, the output equation could take on the following
form:
y1 = c1 x1 + c2 x2 + d1 v(t )
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 13 Chapter 3: Modeling in the Time Domain

Figure 3.1 RL network

d i (t )
L + R i (t ) = v(t )
dt
L [s I ( s ) − i(o )] + R I ( s ) = V ( s )
v(t ) = u (t ) : unit step V (s) = 1 s
i(t): state variable, v(t): input
 

1 1 1  i( 0)
I (s) =  − + d i (t ) 1
Rs s+ R  s+ R
  = [v(t ) − R i(t )]
 L L dt L
1
( )
i(t ) = 1 − e −( R L )t + i(0) e −( R L )t
R ©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
14 Chapter 3: Modeling in the Time Domain
RLC Circuit
di 1
L + R i + ∫ i dt = v(t )
dt C
i(t ) = dq(t ) dt
d 2q dq 1
L 2 +R + q = v(t )
dt dt C
We can convert the last equation into two simultaneous, first-order
differantial equations in terms of i(t) and q(t). We can take vL(t) as output
dq
=i
dt
di 1 R 1
= q − i = v(t )
dt LC L L
output :
1
v L (t ) = − q (t ) − R i (t ) + v(t )
C
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 15 Chapter 3: Modeling in the Time Domain
Figure 3.2 RLC network

0 1  q   0 
d q   1 R    +  1  v(t )
  = −
dt  i   L C L  i   
State-space   L
represantation x = A x + B u
in matrix form:  1  q 
v L (t ) =  − − R    + 1 . v(t )
 C  i 
y = C x +Du
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 16 Chapter 3: Modeling in the Time Domain

Applying the State-Space Representation

In this section we apply the state-space formulation
to the representation of mare complicated physical
systems. The first step in representing a system is to
select the state vector, which must be chosen
according to the following considerations:
• A minimum number of state variables must be
selected as components of the state vector. This
minimum number of state variables is sufficient to
describe completely the state of the system.
• The components of the state vector (i.e., this
minimum number of state variables) must be linearly
independent.
• Let us review and clarify these statements.
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
17 Chapter 3: Modeling in the Time Domain

Figure 3.4
Block diagram of a
mass and damper

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
18 Chapter 3: Modeling in the Time Domain

Figure 3.5
Electrical network for
representation in state
space

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
19 Chapter 3: Modeling in the Time Domain
Solution
Step 1: Label all of the branch current in the network, iL, iR, iC
Step 2: Select the state variablesby writing the derivative equation
dvC diL
C = iC , L = vL
dt dt
Step 3: Apply network theory, such of Kirchhoff’s voltage and
Current laws, to obtain iC and vL in terms of state variables, vC and
iL. At node 1,
iC = − iR + iL
1
= − vC + iL
R
around the outer loop,
vL = −vC + v(t )
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
20 Chapter 3: Modeling in the Time Domain
Solution cont.
Step 4: Using above equations, we can obtain the following state
Equations:
dvC 1
C = − vC + iL
dt R
diL
L = −vC + v(t )
dt
or
dvC 1 1
=− vC + iL
dt RC C
diL 1 1
= − vC + v(t )
dt L L
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 21 Chapter 3: Modeling in the Time Domain
Solution cont.
Step 5: Find the output equation. Since output is iR(t)
1
iR = vC
R
The final result for the state-space representation in vector-matrix
form as follows:

vC  − 1 (RC ) 1 C  vC   0 

 i  =  −   i  +   v(t )
 L  1L 0   L  1 L 
v 
iR = [1 R 0]  C 
 iL 

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 22 Chapter 3: Modeling in the Time Domain
Example 3.2: Find the state and output equation for the electrical
network shown in Figure 3.6 if output vector is
y = [vR 2 iR 2 ]
T
where T means transpose.
Solution: Immediately notice that this network has a voltage-
dependent current source.
Step1: Label all of the branch currents on the network, as shown in
figure

Figure 3.6 Electrical network for Example 3.2

©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 23 Chapter 3: Modeling in the Time Domain
Step2: Select the state variable by listing the voltage-current
relationship for all energy storage elements:
diL dvC
L = vL C = iC
dt dt
From this equations, select the state variables to be the
differentiated variables. Thus, the state variable x1=iL; x2=vC
Step3: Remembering thrt the form of the state equation is
x = Ax + Bu
Using Kirchoff’s voltage and current laws, we find vL; iC in
terms of the state variables and input current source.
Around the mesh L and C,
vL = vC + vR 2 = vC + R2 iR 2
But at Node 2, iR 2 = iC + 4vL Using this relationship
vL = vC + R2 (iC + 4vL )
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 24 Chapter 3: Modeling in the Time Domain
Solving for vL we get
1
vL = (vC + R2 iC )
1 − 4 R2

Notice that vC state variable, we only need to find iC in terms

of the state variables. We will then have obtained vL in terms
of the state variables.
Thus, at Node 1 we write sum of the currents as

iC = i (t ) − iR1 − iL
vR1
= i (t ) − − iL
R1
vL
= i (t ) − − iL
R1
Where vR1=vL
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 25 Chapter 3: Modeling in the Time Domain
From Node 1 and Node 2 equations

(1 − 4 R2 )vL − R2 iC = vC
1
− vL − iC = iL − i (t )
R1
Solving vL and iC
1
vL = [R2 iL − vC − R2i(t )]
∆
1 1 
iC = (1 − 4 R2 )iL + vC − (1 − 4 R2 )i (t )
∆ R1 

 R2 
∆ = − (1 − 4 R2 ) + 
 R1 
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 26 Chapter 3: Modeling in the Time Domain

Using these equation in state variable equation in Step 2

d  iL   R2 (L∆ ) − 1 (L∆ )   iL   − R2 (L∆ ) 
  =    +  i (t )
dt vC  (1 − 4 R2 ) (C∆ ) 1 (R1C∆ ) vC  (1 − 4 R2 ) (C∆ )

Step 4: Drive the output equation. Since the specified output

variables are vR2 and iR2
vR 2 = −vC + vL
iR 2 = iC + 4vL
In vector-matrix form

vR 2   R2 ∆ − (1 + 1 ∆ )   iL  − R2 ∆ 
i  =  1 ∆    +  i (t )
 R2   (1 − 4 R1 ) ∆R1  vC   − 1 ∆ 
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
27 Chapter 3: Modeling in the Time Domain

Translational Mechanical System

It is more convenient when working with mechanical systems
to obtain the state equation directly from the equation of
motion rather than from energy storage elements. For
example, consider an energy-storage element as a spring,
where F = Kx. This relationship does not contain the
derivative of a physical variable as in the case of electrical
networks, where i = C dv / dt capacitors, and V = L di / dt for
inductors. Thus, in mechanical systems, we change our
selection of state variables to be the position and velocity of
each point o early independent motion. In the example we
will see that although there are. energy-storage elements,
there will be four state variables; an additional linearly
independent state variable is included for the convenience of
writing the equations.
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 28 Chapter 3: Modeling in the Time Domain
Example 3.3: Find the state equation for the translational
mechanical system.
Solution: First we write differential equation for the network
in Figure 3.7 d 2x dx
M1 2
1
+D 1
+ Kx1 − Kx 2 = 0
dt dt
d 2 x2
− Kx1 + M 2 2
+ Kx 2 = f (t )
dt

Figure 3.7 Translational mechanical system

©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 29 Chapter 3: Modeling in the Time Domain
Sol. Cont. dx1
= v1
d 2 x1 dv1
=
2
dt dt dt
dx2 d 2 x2 dv2
= v2 2
=
dt dt dt
Complete dx1
the set of = v1
state equation dt
dv1 K D K
=− x1 − v1 − x2
dt M1 M1 M1
dx2
= v2
dt
dv2 K K 1
=+ x1 − x2 + f (t )
dt M2 M1 M2
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
30 Chapter 3: Modeling in the Time Domain

 x1   0 1 0 0  x1   0 
  
d  v1  − K M 1 − D M1 K M1 0  v1   0 
= + f (t )
dt  x2   0 0 0 1   x2   0 
      
 v2   K M 2 0 − K M2 0  v2  1 M2

 x1 
v 
y = x2 = [0 0 1 0] 1 
 x2 
 
 v2 

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 31 Chapter 3: Modeling in the Time Domain

Problem: Find the state-space representation of the electrical

network shown in Figure 3.8.
The output is v0(t)

Figure 3.8 Electric circuit for Skill-Assessment Exercise 3.1

©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
32 Chapter 3: Modeling in the Time Domain
Solution

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
33 Chapter 3: Modeling in the Time Domain

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
34 Chapter 3: Modeling in the Time Domain

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 35 Chapter 3: Modeling in the Time Domain

Problem: Represent the translational mechanical

system shown in Figure 3.9 in state-space, where x3(t)
is output

Figure 3.9 Translational mechanical system for Skill-

Assessment Exercise 3.2
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
36 Chapter 3: Modeling in the Time Domain
Solution

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
37 Chapter 3: Modeling in the Time Domain

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
38 Chapter 3: Modeling in the Time Domain

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
39 Chapter 3: Modeling in the Time Domain

Converting a Transfer Function to State Space

In the last section we applied the state-space representation to electrical
and mechanical systems. We learn how to convert a transfer function
representation to a state-space representation in this section. One
advantage of the state-space representation is that it can be used for the
simulation of physical systems on the digital computer. Thus, if we want
to simulate a system that is represented by a transfer function, we must
first convert the transfer function representation to state space. At first, we
select a set of state variables, called phase variables, where each
subsequent state variable is defined to be the derivative of the previous
state variable. In Chapter 5, we show how to make other choices for the
state variables.
Let us begin by showing how to represent a general, nth-order, linear
differential equation with constant coefficients in state space in the phase-
variable form. We will then show how to apply this representation to
transfer functions.

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 40 Chapter 3: Modeling in the Time Domain
Consider the differential equation
dny d n −1 y dy
n
+ an −1 n −1 +  a1 + a0 y = b0u
dt dt dt
A convenient way to choose state variables is to choose the
output, y(t), and its (n - 1) derivatives as the state variables. This
choice is called the phase-variable choice. Choosing the state
variables, xi, we get x1 = y
dy
x2 =
dt
d2y
x3 = 2
dt

d n −1 y
xn = n −1
dt
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 41 Chapter 3: Modeling in the Time Domain
And differentiating both sides yields
dy
x1 = = x2
dt
d2y
x2 = 2 = x3
dt
d3y
x3 = 3 = x4
dt

d n −1 y
xn −1 = n −1 = xn
dt
xn = −a0 x1 − a1 x2  − an −1 xn + b0u

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 42 Chapter 3: Modeling in the Time Domain
In matrix-vector form
 x1   0 1 0 0 0 0  0   x1   0 
 x   0 0 1 0 0 0  0  x   0 
 2    2   
 x3   0 0 0 1 0 0  0   x3   0 
 =   +  u
          
 xn −1   0 0 0 0 0 0  1   xn −1   0 
      
 xn  − a0 − a1 − a2 − a3 − a4 − a5  − an −1   xn  b0 
Finally, since the solution to the differential equation is y(t), or x1
the output equation is
 x1 
x 
 2 
 x3 
y = [1 0 0  0 0] 
  
 xn −1 
©2000, John Wiley & Sons, Inc.  
Nise/Control Systems Engineering, 3/e  x 
 n 
 43 Chapter 3: Modeling in the Time Domain
Problem: Find the state-space representation in phase variable
form the transfer function shown in Figure 3.10(a)

Figure 3.10 a. Transfer

function; b. equivalent block
diagram showing phase-
variables. Note: y(t) = c(t)
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
44 Chapter 3: Modeling in the Time Domain
Solution:
Step 1
C (s) 24
= 3
(
R( s ) s + 9 s 2 + 26 s + 24 )
Cross-multiplaying yields
(s 3
)
+ 9 s 2 + 26 s + 24 C ( s ) = 24 R( s )

Corresponding diff. Eq.

c + 9c + 26c + 24c = 24r
Step 2: Select the state variables.Choosing the state
variables as succesive derivatives
x1 = c
x2 = c
x3 = c

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 45 Chapter 3: Modeling in the Time Domain
Since output is x1, the combined state and output equations are

x1 = x2
x2 = x3
x3 = −24 x1 − 26 x2 − 9 x3 + 24r
y = c = x1
In vector-matrix form

 x1   0 1 0   x1  0
 x  =  0 0 1   x  + 0  r
 2   2   
 x3  − 24 − 26 − 9  x3  1
 x1 
y = [1 0 0] x2 
 x3 
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
46 Chapter 3: Modeling in the Time Domain

Decomposing a transfer function

The transfer function of the previous problem has a constant term

in the numerator. if a transfer function has a polynomial in s in the
numerator that is of order less than the polynomial in the
denominator, as shown in Figure 3.11(a), the numerator and
denominator can be handled separately. First, separate the transfer
function into two cascaded transfer functions, as shown in Figure
3.11(b); the first is the denominator, and the second is just the
numerator. The first transfer function with just the denominator is
converted to the phase-variable representation in state space as
demonstrated in the last example. Hence, phase variable Xı is the
output and the rest of the phase variables are the internal to the first
block, as shown in Figure 3.11(b). The second transfer function
with just the numerator yields

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
47 Chapter 3: Modeling in the Time Domain
Figure 3.11 Decomposing a transfer function

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 48 Chapter 3: Modeling in the Time Domain

( )
Y ( s ) = C ( s ) = b2 s 2 + b1s + b0 X 1 ( s )
after taking the inverse Laplace transform with zero initial conditions
d 2 x1 dx
y (t ) = b2 2 + b1 1 + b0 x1
dt dt
But the derivative terms are definition with the phase variables
y (t ) = b0 x1 + b1 x2 + b2 x3

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 49 Chapter 3: Modeling in the Time Domain

Converting a transfer function with polynomial in numerator

Problem Find the state-space representation of the transfer

function shown in Figure 3.12(a).

Solution In this problem, the numerator has a polynomial in s

instead of just a constant term.
Step 1 Separate the system into two cascaded blocks, as
shown in Figure 3.12(b). The first black contains the
denominator, and the second black contains the numerator.
Step 2 Find the state equations for the black containing the
denominator.

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 50 Chapter 3: Modeling in the Time Domain

 x1   0 1 0   x1  0
 x  =  0 0 1   x  + 0  r
 2   2   
 x3  − 24 − 26 − 9  x3  1
Step 3 Introduce the effect of the block with the numerator. The second
black of Figure 3. 12(b), where b2 = 1, b1 = 7, and b0 = 2, states that
( ) ( )
C ( s ) = b2 s 2 + b1s + b0 X 1 ( s ) = s 2 + 7 s + 2 X 1 ( s )

Figure 3.12 a. Transfer function; b. decomposed transfer function;

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 51 Chapter 3: Modeling in the Time Domain

Figure 3.12 c. equivalent block diagram. Note: y(t) = c(t)

©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 52 Chapter 3: Modeling in the Time Domain
Exercise 3.3 Find the state eqations and output equation for the
phase variable representationof the transfer function
2s + 1
G (s) = 2
s + 7s + 9

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
53 Chapter 3: Modeling in the Time Domain

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 54 Chapter 3: Modeling in the Time Domain

Converting from State Space to a Transfer Function

In Chapters 2 and 3, we explored two methods of representing
systems: (1) the transfer function representation, and (2) the state-
space representation. In the last section we united the two
representations by converting transfer functions into state-space
representations. Now we move in the opposite direction and
convert the state-space representation into a transfer function.
Given the state and output equations
x = Ax + Bu
y = Cx + Du

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 55 Chapter 3: Modeling in the Time Domain

Taking the Laplace transform assuming zero initial conditions

sX( s ) = AX( s ) + BU(s)
Y( s ) = CX( s ) + DU( s )
Solving X(s)
(sI − A )X( s) = BU(s)
X( s ) = (sI − A ) BU(s)
−1

Where I is identity matrix.

Y( s ) = C(sI − A ) BU(s) + DU(s)
−1

[
Y( s ) = C(sI − A ) B + D U(s)
−1
]
U(s)=U(s) and Y(s)=Y(s) are scalars, we can find the transfer function
Y ( s)
= C(sI − A ) B + D
−1
T ( s) =
U ( s)
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 56 Chapter 3: Modeling in the Time Domain
Exercise 3.4: Convert the state and output equations shown
below
0 1 0  10
x =  0 0 1  x +  0 u
− 1 − 2 − 3  0 
y = [1 0 0]x

Solution:

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
57 Chapter 3: Modeling in the Time Domain
Solution

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
58 Chapter 3: Modeling in the Time Domain
Linearization
A prime advantage of the state-space representation over the
transfer function representation is the ability to represent systems
with nonlinearities, such as the one shown in Figure 3.13. The
ability to represent nonlinear systems does not imply the ability to
solve their state equations for the state variables and the output.
Techniques do exist for the solution of some nonlinear state
equations, but this study is beyond the scope of this course.
If we are interested in small perturbations about an equilibrium
point, as we were when we studied linearization in Chapter 2, we
can also linearize the state equations about the equilibrium point.
The key to linearization about an equilibrium point is, once again,
the Taylor series. In the following example we write the state
equations for a simple pendulum, showing that we can represent a
nonlinear system in state space; then, we linearize the pendulum
about its equilibrium point, the vertical position with zero velocity.

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 59 Chapter 3: Modeling in the Time Domain

Figure 3.13
Walking robots,
such as Hannibal
shown here, can
be used to explore
hostile
environments and
rough terrain,
such as that found
on other planets © Bruce Frisch/S.S./Photo Researchers
or inside
volcanoes.

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 60 Chapter 3: Modeling in the Time Domain

Problem:
First represent teh simple pendulum shown in Figure 3.14(a)
(which could be a simple model for the leg of the robot
shown in Figure 3.13) in state space: Mg is weight, T is an
applied torque in the θ direction, and L is the length of the
pendulum. Assume the mass is evenly distributed with the
center of mass at L/2. Then linearize the state equations about
the pendulum's equilibrium point-the vertical position with
zero angular velocity.
Solution First, draw a free-body diagram as shown in Figure
3.l4(c). Summing the torques, we get

d 2 θ MgL
J 2 + sin θ = T
dt 2
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 61 Chapter 3: Modeling in the Time Domain
Solution
where J is the moment of inertia of the pendulum around the
point. of rotation. Select the state variables x1 and x2 as phase
variables. Letting x1 = θ, and x2 = dθ/dt, we write the state
equations as
x1 = x2
MgL T
x2 = − sin x1 +
2J J
where x2 = d 2 θ dt 2
Thus, we have represented a nonlinear system in state space. it
is interesting to note that the nonlinear Equations represent a
valid and complete model of the pendulum in state space even
under nonzero initial conditions and even if parameters, such
as mass, are time varying.
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 62 Chapter 3: Modeling in the Time Domain

Figure 3.14
a. Simple pendulum;
b. force components
of Mg;
c. free-body diagram

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 63 Chapter 3: Modeling in the Time Domain
Solution cont.
However, if we want to apply classical techniques and convert
these state equations to a transfer function, we must linearize
them. Let us proceed now to linearize the equation about the
equilibrium point, x1 = 0, x2 = 0, that is, θ = 0 and dθ/dt = 0. Let
x1 and x2 be perturbed about the equilibrium point, or
x1 = 0 + δx1
x2 = 0 + δx2
we can use Taylor series expansion to linearize a function,
which express the value of function in terms of the value of
that funtion at a particular point. For small excursions of x
from x0, we can neglect higher order terms. We get
df
f ( x ) − f ( x0 ) ≈ (x − x0 )
dx x = x0
δf ( x) ≈ m x = x δx
0
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 64 Chapter 3: Modeling in the Time Domain
Solution cont.
d (sin x1 )
sin x1 − sin 0 = δx1 = δx1
dx1 x =0
1

sin x1 = δx1
Substituting these eqs. into state equations, we get

δx1 = δx2
MgL T
δx2 = − δx1 +
2J J

Which are linear and good approximation to state eqs. for

small excursions away from the equilibrium point.
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
 65 Chapter 3: Modeling in the Time Domain
Exercise 3.5
Problem: Represent the translational mechanical system shown in
Figure 3.15 in state space about the equilibrium displacement The
spring is nonlinear, where the relationship between the spring force,
fs(t), and the spring displacement, xs(t), is f s (t ) = 2 xs2 (t ) The applied
force is f (t ) = 10 + δf (t ) where δf(t) is a small force about the 10 N
constant value.
Assume the output to be the displacement of the mass, x(t).

Figure 3.15 Nonlinear translational mechanical system for Exercise 3.5

©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
66 Chapter 3: Modeling in the Time Domain
Solution

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
67 Chapter 3: Modeling in the Time Domain

Solution cont.

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
68 Chapter 3: Modeling in the Time Domain
Solution cont.

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
69 Chapter 3: Modeling in the Time Domain
Solution cont.

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 70 Chapter 3: Modeling in the Time Domain

Figure 3.16
Pharmaceutical drug-level
concentrations in a human

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
71 Chapter 3: Modeling in the Time Domain

Figure 3.17
Aquifer system
model

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 72 Chapter 3: Modeling in the Time Domain
Summary
This chapter dealt with the state-space representation of physical systems,
which took the form of a state equation,
x = Ax + Bu
and an output equation,
y = Cx + Du
for t > to, and initial conditions x(to). Vector x is called the state vector
and contains variables, called state variables. The state variables can be
combined algebraically with the input to form the output equation, Eq.
(3.106), from which any other system variables can be found. State
variables, which can represent physical quantities such as current or
voltage, are chosen to be linearly independent. The choice of state
variables is not unique and affects how the matrices A, B, C, and D look.
We will solve the state and output equations for x and y in Chapter 4.
In this chapter, transfer functions were represented in state space. The
form selected was the phase-variable form, which consists of state
variables that are successive derivatives of each other. In three-
dimensional state space, the resulting system matrix, A, for the phase-
variable representation is of the form
©2000, John Wiley & Sons, Inc.
Nise/Control Systems Engineering, 3/e
73 Chapter 3: Modeling in the Time Domain
Summary Cont.
 0 1 0 
 0
 0 1 
− a0 − a1 − a2 

where the ai' s are the coefficients of the characteristic polynomial or

denominator of the system transfer function. We also discussed how to
convert from a statespace representation to a transfer function.
In conclusion then, for linear, time-invariant systems, the state-
space representation is simply another way of mathematically
modeling the system. One major advantage of applying the state-
space representation to these line ar systems is that it allows
computer simulation.

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
74 Chapter 3: Modeling in the Time Domain

Figure P3.1

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
75 Chapter 3: Modeling in the Time Domain

Figure P3.2

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
76 Chapter 3: Modeling in the Time Domain

Figure P3.3

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
77 Chapter 3: Modeling in the Time Domain

Figure P3.4

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
78 Chapter 3: Modeling in the Time Domain

Figure P3.5

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
79 Chapter 3: Modeling in the Time Domain

Figure P3.6

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
80 Chapter 3: Modeling in the Time Domain

Figure P3.7

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
81 Chapter 3: Modeling in the Time Domain

Figure P3.8

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
82 Chapter 3: Modeling in the Time Domain

Figure P3.9

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
83 Chapter 3: Modeling in the Time Domain

Figure P3.10
Gyro system

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 84 Chapter 3: Modeling in the Time Domain

Figure P3.11
Missile

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
85 Chapter 3: Modeling in the Time Domain

Figure P3.12
Motor and load

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
86 Chapter 3: Modeling in the Time Domain

Figure P3.13
Nonlinear
mechanical
system

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 87 Chapter 3: Modeling in the Time Domain

Figure P3.14
a. Robot with
television imaging
system
(©1992 IEEE);
b. vector diagram
showing concept
behind image-based
homing
(©1992 IEEE);
c. heading control
system

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
 88 Chapter 3: Modeling in the Time Domain

Figure P3.15
a. F4-E with
canards
(© 1992 AIAA);
b. open-loop flight
control system
(© 1992 AIAA)

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e
89 Chapter 3: Modeling in the Time Domain

Figure P3.16
Robotic manipulator
and target
environment
(©1997 IEEE)

©2000, John Wiley & Sons, Inc.

Nise/Control Systems Engineering, 3/e