T H R E E

Modeling
in the Time Domain
SOLUTIONS TO CASE STUDIES CHALLENGES
Antenna Control: State-Space Representation
For the power amplifier, = . Taking the inverse Laplace transform, ea+150ea =

150vp. Thus, the state equation is

·
e a =- 150e a +150v p

For the motor and load, define the state variables as x1 = m and x2 = qm. Therefore,

x1 = x2 (1)

Using the transfer function of the motor, cross multiplying, taking the inverse Laplace transform,

and using the definitions for the state variables,

x2 = - x2 + ea (2)

Using the gear ratio, the output equation is

y = 0.2x1 (3)

Also, Jm = Ja+5()2 = 0.05+0.2 = 0.25, Dm = Da+3()2 = 0.01+0.12 = 0.13, =

= 0.8, and = 1.32. Using Eqs. (1), (2), and (3) along with the previous values, the

state and output equations are,

 0 1 0
x= x+ e ; y = 0.2 0 x
0 -1.32 0.8 a
3-2 Chapter 3: Modeling in the Time Domain

Aquifer: State-Space Representation

C1= qi1-qo1+q2-q1+q21 = qi1-0+G2(h2-h1)-G1h1+G21(H1-h1) =

-(G2+G1+G21)h1+G2h2+qi1+G21H1

C2= qi2-q02+q3-q2+q32 = qi2-qo2+G3(h3-h2)-G2(h2-h1)+0 = G2h1-[G2+G3]h2+G3h3+qi2-qo2

C3= qi3-qo3+q4-q3+q43 = qi3-qo3+0-G3(h3-h2)+0 = G3h2-G3h3+qi3-qo3

Dividing each equation by C and defining the state vector as x = [h h h ]T

i 1 2 3

[ ][ ]
−( G1 +G 2 +G3 ) G2 q i 1 + G21 H 1
0
C1 C1 C1
.
G2 −( G2 +G3 ) G3 q i 2 −q o 2
x= x+ u( t )
C2 C2 C2 C2
G3 −G 3 q i 3 −q o 3
0
C3 C3 C3

[ ]
1 0 0
y= 0 1 0 x
0 0 1

where u(t) = unit step function.

ANSWERS TO REVIEW QUESTIONS

1. (1) Can model systems other than linear, constant coefficients; (2) Used for digital simulation
2. Yields qualitative insight
3. That smallest set of variables that completely describe the system
4. The value of the state variables
5. The vector whose components are the state variables
6. The n-dimensional space whose bases are the state variables
7. State equations, an output equation, and an initial state vector (initial conditions)
8. Eight
9. Forms linear combinations of the state variables and the input to form the desired output
10. No variable in the set can be written as a linear sum of the other variables in the set.
11. (1) They must be linearly independent; (2) The number of state variables must agree with the order of
the differential equation describing the system; (3) The degree of difficulty in obtaining the state equations
for a given set of state variables.
12. The variables that are being differentiated in each of the linearly independent energy storage elements
 Solutions to Problems 3-3

13. Yes, depending upon the choice of circuit variables and technique used to write the system equations.
For example, a three -loop problem with three energy storage elements could yield three simultaneous
second-order differential equations which would then be described by six, first-order differential equations.
This exact situation arose when we wrote the differential equations for mechanical systems and then
proceeded to find the state equations.
14. The state variables are successive derivatives.

SOLUTIONS TO PROBLEMS
1.

The state variables are i 2, i 4 and v o.

d i2 d i4 d v0
We have that =v 1 , 3 =v 2 and 0.5 =i 5.
dt dt dt

Applying KVL around the external loop one gets

v i−2i 1−3 i 3−i 5−v o=0

And in each one of the nodes

i 3=i 1−i 2 and i 5=i 3−i 4

Substituting

v i−2i 1−3 (i 1−i2 )−(i 3−i 4 )−v o=0

or

v i−2i 1−3 (i 1−i2 )−((i 1−i 2 )−i 4 )−v o=0

Solving for i 1 one gets

4 1 1 1
i 1= i 2+ i 4 + vi − v o
6 6 6 6

Thus
3-4 Chapter 3: Modeling in the Time Domain

d i2 −4 1 2 1
v 1= =v i−2 i 1= i − i + v+ v
dt 3 2 3 4 3 i 3 o

Also
−1 1 1 1
i 3=i 1−i 2= i + i + v− v
3 2 6 4 6 i 6 o

and
−1 5 1 1
i 5=i 3−i 4 = i2− i 4 + v i− v o
3 6 6 6
d vo −2 5 1 1
=2 i5 = i2 − i 4 + v i− v o
dt 3 3 3 3

Finally
d i 4 v 2 i 5+ v o −1 5 1 5
= = = i 2− i 4 + v i + v o
dt 3 3 9 18 18 18

[[ ] ][ ] [ ]
In matrix form we have
d i2 −4 −1 2 2
dt 3 3 3 5
i2
d i4 −1 −5 5 5
= i4 + v
dt 9 18 18 18 i
d vo −2 −5 −1 v o 1
dt 3 3 3 3

[]
i2
y=v o= [ 0 0 1 ] i 4
vo

2.
Let C1 be the grounded capacitor and C 2 be the other. Now, writing the equations
for the energy storage components yields,

di L
=v i−v C
dt 1

dv C i −i
1
=1 2
dt 2
dv C i 2−i 3
2
=
dt 2 (1)
 Solutions to Problems 3-5

[]
iL
x= v C
1

vC
Thus the state vector is 2
.
Now, find the three loop currents in terms of the state variables and the input.

v C =v C +2i 2
Writing KVL around Loop 2 yields 1 2 .
i 2 =0 .5 v C −0 .5 v C
Or, 1 2

i +2 i 2=v i
Writing KVL around the outer loop yields 3
Or,
i 3 =v i −2i 2 =vi −v C + v C
1 2

Also,
i 1−i 3 =i L . Hence, i 1=i L +i 3 =i L + vi −v C1 + v C 2

Substituting the loop currents in equations (1) yields the results in vector-matrix form,

v o =2i 2 =v C −v C 2
Since 1 , the output equation is
3-6 Chapter 3: Modeling in the Time Domain

3.
Equations of motion in Laplace:

Equations of motion in the time domain:

Define state variables:

Substituting Eq. (1) in (2), (3) in (4), and (5) in (6), we obtain, respectively:

dz 1
=z 2 (7)
dt
dz 3
=z 4 (8)
dt
dz 5
=z 6 ( 9)
dt

Substituting Eqs. (1) through (6) into the equations of motion in the time domain and
solving for the derivatives of the state variables and using Eqs. (7) through (9) yields the
state equations:
 Solutions to Problems 3-7

The output is x1 = z1.

In vector-matrix form:

4.

The impedance equations are:

( 2 s 2 +2 s+ 1 ) X 1 ( s )−s X 2 ( s )−( s+1 ) X 3 ( s )=0

−s X 1 ( s ) + ( s2 +2 s +1 ) X 2 ( s )−( s+1 ) X 3 ( s )=0

−(s +1) X 1 ( s )−( s+1 ) X 2 ( s ) + ( s2 +2 s +2 ) X 3 ( s )=f (t)

Taking the inverse Laplace transform

2 ẍ1 +2 ẋ 1+ x 1− ẋ 2 − ẋ 3−x 3=0

− x˙1+ ẍ2 +2 x˙2+ x 2− x˙3− x3 =0

− x˙1−x 1− x˙2−x 2+ ẍ3 +2 x˙3+ 2 x 3=f (t)

3-8 Chapter 3: Modeling in the Time Domain

1 1 +1 1
x¨1=− ẋ 1− x 1+ ẋ ẋ + x 3
2 2 2 2 3 2

x¨2= x˙1−2 ˙x2 + x 2 + x˙3+ x 3 =0

x¨3= x˙1+ x1 + x˙2 + x 2−2 x˙3−2 x 3 +f (t)

Define the state variables

z 1=x 1; z 2= x˙1; z 3=x 2; z 4 = ẋ2 ; z 5=x 3; z 6= x˙3

The equations are rewritten as

z˙1= x˙1=z 2
1 1 1 1
z˙2= x¨1=−z 2− z + z 4 + z 6 + z 5
2 1 2 2 2
z˙3= x˙2=z 4
z˙4 = ẍ2 =z1−2 z 4 −z 3+ z 6 + z 5

z˙5= x˙3=z 6

z˙6= x¨3=z 2 + z 1+ z 4+ z3 −2 z 6 −2 z 5

[ ] []
In matrix form
0 1 00 0 0
0
1 1 1 1
− −1 0 0
2 2 2 2
0
ż= 0 0 0 1 0 0 z+ f (t )
0
1 0 −1 −2 1 1
0
0 0 0 0 0 1
1
1 1 1 1 −2 −2

y= [ 0 0 1 0 0 0 ] z
 Solutions to Problems 3-9

5.

1600
50

Writing the equations of motion,

Taking the inverse Laplace transform and simplifying,

4T

Defining the state variables as

· ·
x 1=q2 , x 2 =q2 , x 3 =q3 , x 4 =q3

Writing the state equations using the equations of motion and the definitions of the state

variables

In vector-matrix form,
3-10 Chapter 3: Modeling in the Time Domain

6.
Drawing the equivalent circuit,

10T

200(1/10)2 =2 N-m/rad
(1/10)(102 ) = 10 N-m/rad

Writing the equations of motion,

Taking the inverse Laplace transform,

(1)

(2)

(3)
From (3),

and (4)
assuming zero initial conditions.

From (1)

(5)
 Solutions to Problems 3-11

Substituting (4) and (5) into (2) yields the state equation (notice there is only one equation),

The output equation is given by,

7.
X 1( s ) X 2( s )
Solving Eqs. (3.44) and (3.45) in the text for the transfer functions F( s ) and F( s ) :
0 K M 1 s2 + D s + K 0
F M2 s2 + K K F
X 1 s = X 2 s =
M 1 s2 + D s + K K M 1 s2 + D s + K K
K M 2 s2 + K and K M 2 s2 + K

Thus,
X1 s = K
4 3 2 2
F s M 2 M 1 s +D M 2 s +K M 2 s +K M 1 s +D K s

and
2
X2 s M 1 s +D s +K
=
4 3 2 2
F s M 2 M 1 s +D M 2 s +K M 2 s +K M 1 s +D K s

Multiplying each of the above transfer functions by s to find velocity yields pole/zero cancellation at
the origin and a resulting transfer function that is third order.

8.
a. . Using the standard form derived in the textbook,

[ ] []
0 1 0 0 0
¿
x= 0
0 1 0
x + 0 r (t )
0 0 0 1 0
−80 −20 −15 −10 1
c= [ 80 0 0 0]x
b. Using the standard form derived in the textbook,
3-12 Chapter 3: Modeling in the Time Domain

[ ] []
0 1 0 0 0 0
¿ 0 0 1 0 0 0
x= 0 0 0 1 0 x+ 0 r (t )
0 0 0 0 1 0
−20 −1 −3 −7 −5 1
c= [ 20 0 0 0 0]x

9.
Program:
'a'
num=100;
den=[1 20 10 7 100];
G=tf(num,den)
[Acc,Bcc,Ccc,Dcc]=tf2ss(num,den);
Af=flipud(Acc);
A=fliplr(Af)
B=flipud(Bcc)
C=fliplr(Ccc)
'b'
num=30;
den=[1 8 9 6 1 30];
G=tf(num,den)
[Acc,Bcc,Ccc,Dcc]=tf2ss(num,den);
Af=flipud(Acc);
A=fliplr(Af)
B=flipud(Bcc)
C=fliplr(Ccc)

Computer response:
ans =

Transfer function:
100
---------------------------------
s^4 + 20 s^3 + 10 s^2 + 7 s + 100

A=

0 1 0 0
0 0 1 0
0 0 0 1
-100 -7 -10 -20

B=

0
0
0
1
 Solutions to Problems 3-13

C=

100 0 0 0

ans =

Transfer function:
30
------------------------------------
s^5 + 8 s^4 + 9 s^3 + 6 s^2 + s + 30

A=

0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1
-30 -1 -6 -9 -8

B=

0
0
0
0
1

C=

30 0 0 0 0

10.
a. Using the standard form derived in the textbook,

b. Using the standard form derived in the textbook,

22
3-14 Chapter 3: Modeling in the Time Domain

11.
Program:
'a'
num=[8 10];
den=[1 5 1 5 13]
G=tf(num,den)
[Acc,Bcc,Ccc,Dcc]=tf2ss(num,den);
Af=flipud(Acc);
A=fliplr(Af)
B=flipud(Bcc)
C=fliplr(Ccc)
'b'
num=[1 2 12 7 6];
den=[1 9 13 8 0 0]
G=tf(num,den)
[Acc,Bcc,Ccc,Dcc]=tf2ss(num,den);
Af=flipud(Acc);
A=fliplr(Af)
B=flipud(Bcc)
C=fliplr(Ccc)

Computer response:
ans =

ans =

den =

1 5 1 5 13

Transfer function:
8 s + 10
----------------------------
s^4 + 5 s^3 + s^2 + 5 s + 13

A=

0 1 0 0
0 0 1 0
0 0 0 1
-13 -5 -1 -5

B=

0
0
0
1
 Solutions to Problems 3-15

C=

10 8 0 0

ans =

den =

1 9 13 8 0 0

Transfer function:
s^4 + 2 s^3 + 12 s^2 + 7 s + 6
------------------------------
s^5 + 9 s^4 + 13 s^3 + 8 s^2

A=

0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1

0 0 -8 -13 -9

B=

0
0
0
0
1

C=

6 7 12 2 1

12.

The transfer function can be represented as a block diagram as follows:

1 2
3 2 s −3 s
s + 4 s +13 s+10
3-16 Chapter 3: Modeling in the Time Domain

The differential equation of the first box is

⃛x +4 ẍ +13 ẋ +10 x=r (t)

Define the state variables

x 1=x ; x 2= ẋ ; x 3= ẍ

so

x˙1=x 2

x˙2=x 3

x˙3=−4 ẍ−13 ẋ−10 x+ r ( t )=−4 x 3−13 x 2−10 x 1+r (t)

From the second box

y= ẍ −3 ẋ =x3 −3 x 2

In vector-matrix form

[ ] []
0 1 0 0
ẋ= 0 0 1 x+ 0 r (t )
−10 −13 −4 1

y= [ 0 −3 1 ] x

13.
a. G(s)=C(sI-A)-1B

[ ] []
0 1 0 0
A= 0 0 1 ; B= 0 ; C=[ 1 0 0 ]
−1 −3 −2 23

[ ]
s 2 +2 s +3 s +2 1
−1 1
(s I − A) = 3 2
−1 s ( s+2) s
s +2 s +3 s+1
−s −(3 s +1) s 2

23
Therefore, G ( s )= 3 2 . Note that in this case the result could have also been obtained
s +2 s + 3 s+1
by inspection.

b. G(s)=C(sI-A)-1B
 Solutions to Problems 3-17

[ ] []
−1 2 −6 5
A= −4 −5 0 ; B= 1 ; C=[ 2 0 2 ]
3 −3 7 2
−1 1
(s I − A) = 3 2 ¿
−s +s +11s−71
2
14 s +8 s−134
Therefore G ( s )= 3 2
.
s −s −11s +71

c. G(s)=C(sI-A)-1B

[ ] []
−2 8 7 1
A= 5 −4 2 ; B= −5 ; C=[ 7 2 1 ]
−9 −3 −1 −1

[ ]
s2 +5 s+10 8 s−13 7 s +44
1
(s I − A)−1= 3 2 5 s−13
2
s +3 s +65 2 s+39
s +7 s +43 s+ 481 2
−(9 s +51) −(3 s+78) s + 6 s−32
2
−4 s +318 s +166
( )
Therefore G s = 3 2
.
s +7 s +43 s+ 481

14.
Program:

'a'
A=[0 1 5 0;0 0 1 0;0 0 0 1;-7 -9 -2 -3];
B=[0;5;8;2];
C=[1 3 6 6];
D=0;
statespace=ss(A,B,C,D)
[num,den]=ss2tf(A,B,C,D);
G=tf(num,den)
'b'
A=[3 1 0 4 -2;-3 5 -5 2 -1;0 1 -1 2 8;-7 6 -3 -4 0;-6 0 4 -3 1];
B=[2;7;8;5;4];
C=[1 -2 -9 7 6];
D=0;
statespace=ss(A,B,C,D)
[num,den]=ss2tf(A,B,C,D);
G=tf(num,den)

Computer response:

ans =
3-18 Chapter 3: Modeling in the Time Domain

a=
x1 x2 x3 x4
x1 0 1 5 0
x2 0 0 1 0
x3 0 0 0 1
x4 -7 -9 -2 -3

b=
u1
x1 0
x2 5
x3 8
x4 2

c=
x1 x2 x3 x4
y1 1 3 6 6

d=
u1
y1 0

Continuous-time model.

Transfer function:
75 s^3 - 96 s^2 - 2331 s - 210
------------------------------
s^4 + 3 s^3 + 2 s^2 + 44 s + 7

ans =

a=
x1 x2 x3 x4 x5
x1 3 1 0 4 -2
x2 -3 5 -5 2 -1
x3 0 1 -1 2 8
x4 -7 6 -3 -4 0
x5 -6 0 4 -3 1

b=
u1
x1 2
x2 7
x3 8
x4 5
x5 4

c=
x1 x2 x3 x4 x5
y1 1 -2 -9 7 6

d=
u1
y1 0
 Solutions to Problems 3-19

Continuous-time model.

Transfer function:
-25 s^4 - 292 s^3 + 1680 s^2 + 1.628e004 s + 3.188e004
------------------------------------------------------
s^5 - 4 s^4 - 32 s^3 + 148 s^2 - 1153 s - 4480
3-20 Chapter 3: Modeling in the Time Domain

15.

Program:
syms s
'a'
A=[0 1 5 0
0010
0001
-7 -9 -2 -3];
B=[0;5;8;2];
C=[1 3 4 6];
D=0;
I=[1 0 0 0
0100
0010
0 0 0 1];
'T(s)'
T=C*((s*I-A)^-1)*B+D;
T=simple(T);
pretty(T)
'b'
A=[3 1 0 4 -2
-3 5 -5 2 -1
0 1 -1 2 8
-7 6 -3 -4 0
-6 0 4 -3 1];
B=[2;7;6;5;4];
C=[1 -2 -9 7 6];
D=0;
I=[1 0 0 0 0
01000
00100
00010
0 0 0 0 1];
'T(s)'
T=C*((s*I-A)^-1)*B+D;
T=simple(T);
pretty(T)

Computer response:

ans =

ans =

T(s)

3 2
59 s - 148 s - 2241 s - 140
-----------------------------
4 3 2
s + 3 s + 2 s + 44 s + 7

ans =
 Solutions to Problems 3-21

ans =

T(s)

4 3 2
(- 7 s - 408 s + 1708 s + 14582 s +

5 4 3 2
27665) / (s - 4 s - 32 s + 148 s -

1153 s - 4480)
>>

16.

The equivalent cascade transfer function is as shown below.

Ka
d (s) K X(s) K F (s)
3 b
K K K s+
Ka
s 3+ 2 2
s + 1
s+ 0
K K K
3 3 3

K K K
... 2  1  0 Ka
For the first box, x + x+ x+ x = (t).
K K K K
3 3 3 3
 
Selecting the phase variables as the state variables: x =x, x =x, x =x.
1 2 3

Writing the state and output equations:

x1 = x2

x2 = x3

x3 = - x1- x2- x3+ (t)

y = (t) = x+ x = x1+x2
In vector-matrix form,
3-22 Chapter 3: Modeling in the Time Domain

0 1 0 0
0 0 1 0 K
 b
x = K K K x+ (t) ; y = 1 0 x
0 1 2
Ka Ka
- - - K
K K K
3 3 3 3
17.

Since Tm = Jeq + Deqm, and Tm = Kt ia,

Jeq + Deqm = Kt ia (1)

Or,

= - m + ia

But, m = L.

Substituting in (1) and simplifying yields the first state equation,

= - L + ia
The second state equation is:

= L
Since

ea = Raia+La +Kbm = Raia+La +KbL,

the third state equation is found by solving for . Hence,

= - L - ia+ ea

Thus the state variables are: x1 = L, x2 = L , and x3 = ia.

Finally, the output is y = m = L .

In vector-matrix form,

Deq K N
t 1
- 0 0
Jeq Jeq N
2 N
 0 2
x= 1 0 0 x+ ea ; y = 0 0 x
1 N
K N 1
b 2 Ra La
- 0 -
La N La
1
where,
 Solutions to Problems 3-23


L
x = L
ia

18.
Controller:

The transfer function can be represented as a block diagram as follows:

Rc (s) Xc (s) Yc (s)

K1
s a
sb

Writing the differential equation for the first box,

and solving for ,

From the second box,

Wheels:

The transfer function can be represented as a block diagram as follows:

Rw (s) c Xw(s)

sc

Writing the differential equation for the block of the form,

and solving for ,

The output equation is,

yw = xw
3-24 Chapter 3: Modeling in the Time Domain

Vehicle:

The transfer function can be represented as a block diagram as follows:

Rv (s) 1 Xv(s)

Writing the differential equation for the block,

and solving for ,

The output equation is

yv = xv

19.
Adding displacements to the figure,

xr xs xe

Writing the differential equations for noncontact,

d2 xr dx r dx s
+2 +2 x r −x s− =u( t )
dt 2 dt dt
dx r d2 x s dx s
− −xr + 2 + + x s =0
dt dt dt

Define the state variables as,

· ·
x 1= x r ; x 2= x r ; x3 = x s ; x 4 =x s

Writing the state equations, using the differential equations and the definition of the state variables,

we get,
 Solutions to Problems 3-25

· ·
x 1=x r =x 2
· ··
x 2 =xr =- 2 x 1−2 x 2 + x 3 + x 4 +u( t )
· ·
x 3 =x s =x 4
· ··
x 4 =x s =x 1 + x 2 −x 3−x 4

Assuming the output to be xs, the output equation is,

y=x 3
In vector-matrix form,

[ ] []
0 1 0 0 0
·
x = −2 −2 1 1 x+ 1 u(t )
0 0 0 1 0
1 1 −1 −1 0
y= [ 0 0 1 0]x
Writing the differential equations for contact,

d2 xr dx r dx s
+2+2 x r −x s− =u( t )
dt 2 dt dt
dx d2 x dx
− r −xr + 2 s + s + x s −z−x e =0
dt dt dt
dz dx
−x s + + z− e =0
dt dt
2
dz d x dx
−x s − + 2 e + 2 e +2 x e=0
dt dt dt

Defining the state variables,

· · · ·
x 1= x r ; x 2= x r ; x3 = x s ; x 4 =x s ; x 5= z ; x 6 =z ; x 7 =x e ; x 8 =x e

Using the differential equations and the definitions of the state variables, we write the state equations.
·
x 1=x 2
·
x 2 =- x 1−2 x 2 + x 3 + x 4 +u ( t )
·
x 3 =x 4
·
x 4 =x 1 + x 2 −x 3 −x 4 + x 5 + x7
·
x 5 =x 6
3-26 Chapter 3: Modeling in the Time Domain

Differentiating the third differential equation and solving for d2z/dt2 we obtain,

2
·
d 2 z dx s dz d x e
x 6= = − +
dt 2 dt dt dt 2

But, from the fourth differential equation,

d2 xe dz dx e
= xs+ −2 −2 x e = x3 + x 6 −2 x 8 −2 x 7
dt 2 dt dt
·

Substituting this expression back into x 6 along with the other definitions and then simplifying yields,
·
x 6 =x 4 + x 3−2 x 8 −2 x 7
Continuing,
·
x 7 =x 8
·
x 8 =x 3 + x 6 −2 x7 −2 x 8
Assuming the output is xs,

y=x s
Hence, the solution in vector-matrix form is

[ ] []
0 1 0 0 0 0 0 0 0
−1 −2 1 1 0 0 0 0 1
0 0 0 1 0 0 0 0 0
·
1 1 −1 −1 1 0 1 0 0
x= x+ u(t )
0 0 0 0 0 1 0 0 0
0 0 1 1 0 0 −2 −2 0
0 0 0 0 0 0 0 1 0
0 0 1 0 0 1 −2 −2 0
y= [ 0 0 1 0 0 0 0 0 ] x
 Solutions to Problems 3-27

20.

a.

>> A=[-0.038 0.896 0 0.0015; 0.0017 -0.092 0 -0.0056; 1 0 0 -3.086; 0 1 0 0]

A=

-0.0380 0.8960 0 0.0015

0.0017 -0.0920 0 -0.0056

1.0000 0 0 -3.0860

0 1.0000 0 0

>> B = [-0.0075 -0.023; 0.0017 -0.0022; 0 0; 0 0]

B=

-0.0075 -0.0230

0.0017 -0.0022

0 0

0 0

>> C = [0 0 1 0; 0 0 0 1]

C=

0 0 1 0

0 0 0 1
3-28 Chapter 3: Modeling in the Time Domain

>> [num,den] = ss2tf(A,B,C,zeros(2),1)

num =

0 0.0000 -0.0075 -0.0044 -0.0002

0 0 0.0017 0.0001 0

den =

1.0000 0.1300 0.0076 0.0002 0

>> [num,den] = ss2tf(A,B,C,zeros(2),2)

num =

0 -0.0000 -0.0230 0.0027 0.0002

0 -0.0000 -0.0022 -0.0001 0

den =

1.0000 0.1300 0.0076 0.0002 0

b.

From the MATLAB results

 Solutions to Problems 3-29

z −(0 . 0075 s 2 +0 . 0044 s+0 . 0002)

( s )= 3
δB s (s + 0. 13 s 2 + 0. 0076 s+0 . 0002)
θ 0 . 0017 s +0 . 0001
( s )= 3
δB s +0. 13 s 2 +0 . 0076 s+0 . 0002
z −0 . 023 s 2 +0 . 0027 s +0 . 0002
( s )= 3
δS s (s +0. 13 s 2 +0 . 0076 s+0 . 0002)
θ −(0 . 0022 s+ 0 .0001 )
( s )= 3
δS s +0 . 13 s 2 +0 . 0076 s +0 . 0002
21.

a.
36
G( s)=C (sI−A )−1 B=9( s+10 )−1 4=
( s+10 )
b.

[ ]
1
0
[ ][] [ 41 ]
−1
s +10 0 4 s+10
G( s )=C (sI−A )−1 B=[ 9 0 ] =[9 0 ]
0 s+1 1 1
0
s+ 1

=
[ 9
s+10 ][ ]
0 4 =
36
1 s+10

c.

[ ]
1
0
[ ][] [ 40 ]
−1
−1 s+10 0 4 s+10
G( s )=C (sI−A ) B=[ 9 4 ] =[ 9 4]
0 s +1 0 1
0
s +1

[ ]
4
36
= [9 0 ] s +10 =
s +10
0
3-30 Chapter 3: Modeling in the Time Domain

22.
a.

d mSO
=k O 1 m A ( t )−(k O 2+ k O 3 )mSO ( t )+ k O 4 m IDO (t)
dt
d mIDO
=k O 3 mSO ( t )−k O 4 mIDO (t)
dt
d mV
=k L1 m A ( t )−(k L2 +k L3 )mV ( t )
dt
d mS
=k L3 mV (t )−k L 4 mS ( t )
dt

b.
 Solutions to Problems 3-31

23.

The armature loop equation is

d i a (t ) d θ m (t)
e a ( t ) =R a i a + La +Kb
dt dt

The motor load equation is

2
d θ m (t) d θm
T m (t )=J 2
+D + K θm
dt dt

Substituting the defined state variables one gets

e a ( t ) =R a x 3 + La ẋ 3+ K b x 2

Or
−K b Ra u
ẋ 3= x 2− x 3 +
La La La
ẋ 1=x 2

And

K m x 3=J ẋ2 + D x 2+ K x 1

Or
−K D Km
ẋ 2= x 1− x 2+ x
J J J 3

Putting these equations in matrix form:

[ ][ ] [ ]
0 1 0

[]
ẋ 1 K m x1 0
−K −D
0
ẋ 2 = J J J x2 + u
1
ẋ3 −K b −R a x 3
0 La
La La

[]
x1
y= [ 1 0 0 ] x 2
x3

24.

Let . Cross multiplying the transfer function,

3-32 Chapter 3: Modeling in the Time Domain

Taking the inverse Laplace transform,

Defining the state variables,

Writing the state equations,

In vector-matrix form,

25.

a. The output equation is

[]
θ
θ̇
y=lθ+x=Cx=[ l 0 1 0 ]
x
ẋ

b. Substituting the given values into the A matrix and B and C vectors yields,
 Solutions to Problems 3-33

[ ][ ]
0 1 0 0 0
29 . 8615 0 0 0 −1. 1574
A= B=
0 0 0 1 0
−0 . 9401 0 0 0 ; 0 . 4167
C=[ 0.36 0 1 0 ] ; D = 0

Using the following MATLAB M-file:

A=[0 1 0 0;29.8615 0 0 0;0 0 0 1;-0.9401 0 0

0];
B=[0;-1.1574;0; 0.4167];
C=[0.36 0 1 0];
D=0;
[numg, deng]=ss2tf(A, B, C, D, 1);
display G(s)=Y(s)/U(s);
G=tf(numg, deng)

The computer output (in the “Command” window) is:

G(s)=Y(s)/U(s)

G =

3.6e-05 s^2 - 2.378e-15 s - 11.36

---------------------------------

s^4 - 29.86 s^2

Continuous-time transfer function.

Note: As will be noted in chapter 6, a system with missing powers of ‘s’ and alternating signs

of its denominator coefficients is unstable and needs to be stabilized.

26.

a.

f 1=s−dT −(1−u1 )β Tv
3-34 Chapter 3: Modeling in the Time Domain

¿
f 2=(1−u1 ) β Tv−μT
¿
f 3 =(1−u 2 )kT −cv

∂f1
| =−d−(1−u1 ) βv|0=−d−(1−u10 ) βv 0
∂T 0
∂f1
¿ | =0
∂T 0
∂f1
| =(1−u 1 ) βv|0 =(1−u 10 )βv 0
∂v 0
∂f2
| =(1−u 1 ) βv|0 =(1−u10 ) βv 0
∂T 0
∂f2
¿ | =−μ
∂T 0
∂f2
| =(1−u 1 ) βT|0 =(1−u10 )βT 0
∂v 0
∂f3
| =0
∂T 0
∂f3
¿ | =(1−u 20 )k
∂T 0
∂f3
| =−c
∂v 0
∂f1
| =βT 0 v 0
∂u1 0
∂f1
| =0
∂u2 0
∂f2
| =−βT 0 v 0
∂u1 0
∂f2
| =0
∂u2 0
∂f3
| =0
∂u1 0
∂f3 ¿
|0 =−kT 0
∂u2
 Solutions to Problems 3-35

Then just by direct substitution.

b.

Substituting values one gets:

[ ][ ][ ] [ ][ ]
Ṫ −(d + βv 0 ) 0 −βT 0 T βT 0 v 0 0
¿ ¿ u1
Ṫ = βv 0 −μ βT 0 T + −βT 0 v 0 0
u
v̇ 0 k −c v 0
¿
−kT 0 2

[]
T
y= [ 0 0 1 ] T ¿
v

27.
a. The following basic equations characterize the relationships between the state, input, and output
variables for the HEV common forward path of the figure:

(1)

, where ,

,
b. Given that the state variables are the motor armature current, Ia(t), and angular speed,  (t), we re-
write the above equations as:

(2)
 3-36 Chapter 3: Modeling in the Time Domain

(3)

In matrix form, the resulting state-space equations are:

(4)

(5)

28.
The corresponding differential equation is:

¨
h(t)+0.0224 ˙
h (t)+196 ×10−6 h (t)=137.2 ×10−6 q(t−39)

We select the state variables as

x 1=h
x 2=ḣ
So we can write
ẋ 1=x 2
−6 −6
ẋ 2=−0.0024 x 2−196 × 10 x 1+132.2 ×10 q ( t−39 )

In matrix form:

[][
ẋ1
=
0
ẋ2 −196 ×10
−6
1
−0.0024 x 2 1
x1 0
][ ] [ ] −6
+ 137.2× 10 q(t−39)

x
h ( t )= y= [ 1 0 ] 1
x2 []
Alternatively this can also be written as:

[][
ẋ1
=
0
ẋ2 −196 ×10 −6
1 x1 0
−0.0024 x 2 1 ][ ] [ ]
−6
+ 137.2× 10 q(t)
 Solutions to Problems 3-37

h ( t+ 39 )= y (t +39)=[ 1 0 ]
[]
x1
x2