Solution: 7.9.5.2
For the system of Figure 1 wehave
R
+
C
G
c
G
p
Σ
-
Figure 1: Cascade Compensation with UnityFeedback
G
p
=
1
(s+1)(s+2)
and and G
c
(s)=
K
c
(s+1)(s+ z
2
)
(s+0:2)(s+20)
:
Figure 2 shows the evaluation of G
c
G
p
(s)ats = ;3+j2incomponent
form. That is, eachofthe vectors in Figure 2 is the polar representation
of one of the factors in G
c
G
p
evaluated at s = ;3+j2. The vector V
1
, for
instance, is the polar representation of the factor s in the denominator of
G
c
G
p
, V
2
the polar representation of the factor s +2,andsoon. Notice
that the pole at s = ;1hasbeen cancelled by the zero at s = ;1 and so
those twotermsdo not enter into the angle and magnitude calculations to
be made next. Thus,
G
c
G
p
(s) j
s=;3+j2
=
K
c
jV
3
j
6
(jV
1
j
6
1
)(jV
2
j
6
2
)(jV
4
j
6
3
)
=
K
c
jV
3
j
jV
1
jjV
2
jjV
4
j
6
(;
1
;
2
;
3
)
The evaluation of G
c
G
p
at s = ;3+j2hasnowbeen broken down into
acomposite magnitude and a composite angle. For the root locus to pass
through this point, wemust have
;
1
;
2
;
3
= ;180
:
Since all the angles except are known, wecanwrite
=
1
+
2
+
3
;180
=
h
180
;tan
;1
(2=2:8)
i
+
h
180
;tan
;1
(2=1)
i
+tan
;1
(2=17);180
= 144:46
+116:57
+6:71
;180
=87:74
1
θ
1
θ
2
-2
α
-3 -a
Im(s)
Re(s)
2
V
3
=
s + a
V
2 =
s + 2
V
1 =
s + 0.2
θ
3
V
4
=
s + 10
-20 -0.2
Figure 2: Satisfaction of Angle Condition For Desired Closed Loop Poles
Then
a = 3+
2
tan(87:74
)
=3:08
The gain that places a closed loop pole at s = ;3+j2, and another at
s = ;3;j2isobtained by solving
K
c
jV
3
j
jV
1
jjV
2
jjV
4
j
=1;; (1)
or
K =
jV
1
jjV
2
jjV
4
j
jV
3
j
: (2)
The gain to place the poles at this location is then
K =
p
11:84
p
5
p
293
p
4:0062
=65:8
The lead/lag compensator is then
G
c
(s)=
65:8(s+1)(s+3:08)
(s+0:2)(s+20)
2
-20
-3
-3.08 -2
2
-2
Im(s)
Re(s)
-0.2
Figure 3: Root Locus of Compensated System
Part b.
K
p
= lim
s!0
G
c
(s)G
p
(s)=lim
s!0
65:8(s+3:08)
(s+0:2)(s+2)((s+20)
=
65:83:08
0:2220
=25:325
Then the steady state error to a step input is
e
ss
=
1
1+K
p
=
1
26:325
=0:0395;;
or about 4%. Thus, the system will trackastep quite well.
Part c.
The system has three closed loop poles, under the assumption of perfect
cancellation. Weknow where twoofthethree closed loop poles are, namely
at s = ;3 j2. The next task is to nd the third closed loop pole. The
root locus in Figure 3 shows where that third closed loop pole must be. One
waytond the pole is to searchalongthe real axis between s = ;20 and
3
V
4
= s + 20 V
3
= s + 3.08
V
2
= s + 2
V
1
= s = 0.2
Re(s)
Im(s)
-20 -3.08 -2
-0.2
Figure 4: Calculation of Gain Along Real Axis
s -17 -16.5 -16.3 16.2
K 54.31 61.64 64.44 65.8
Table 1: SearchforClosed Loop Pole
s = ;3:08 until we nd a value of s that yields a gain of 65.8. Figure 4
shows how this calculation is made. That is
K
c
=
jV
1
jV
2
jjV
4
j
jV
3
j
Table 1 summarizes the searchfor the third closed loop pole. Thus, the third
closed loop pole is at approximately s = ;16:2. the closed loop transfer
function is
T
c
(s)=
65:8(s+3:08)
(s+16:2)(s+3;j2)(s+3+j2)
The following MATLAB program checks this result and generates the step
response shown in Figure 5
gcgp = zpk([-3.08], [-0.2 -2 -20],Kc)
tc = feedback(gcgp,1)
step(tc,3)
print -deps sr7952.eps
The damped frequency is 2 r./s. and the damping ratio is
= cos[tan
;1
(2=3)]= cos(33:69
)=0:83
Thus,if we had guessed the time response it would be as shown in Figure 6.
The formula for overshoot predicts less than 2% overshoot, but weknow
4
Time (sec.)
Amplitude
Step Response
0 0.5 1 1.5 2 2.5 3
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Figure 5: Step Response of Compensated System
1.5 s.
Over shoot = 10%
t
c(t)
Figure 6: Estimated Time Response
5
from experience that the formula can be too optimistic, so a good guess is
10%. The time to peak should be
t
p
=
!
d
1:5s.:
Wecan see that the actual response is actually better than our guess.
d
C(s) =
65:8(s+3:08)
s(s+16:2)(s+3;j2)(s+3+j2)
=
1
s
+
B
s+16:2
+
M
s +3;j2
+
M
s+3+j2
:
A = sC(s) j
s=0
= (s+16:2)
65:8(s+3:08)
(s+3;j2)(s+3+j2)(s+16:2)
s=0
= 0:962
B = (s+16:2)C(s)j
s=;16:2
= (s+16:2)
65:8(s+3:08)
s(s+3;j2)(s+3+j2)(s+16:2)
s=;16:2
=
65:8(;13:2)
(;16:2)(;13:2;j2)(;13:2+j2)
=
;868:56
;2887:488
= 0:299
M = (s+3;j2)C
c
(s) j
s=;3+j2
= (s+3;j2)
65:8(s+3:08)
s(s+3;j2)(s+3+j2)(s+16:2)
s=;3+j2
= 0:6305
6
;2:743 rad:
Thus, nally wehave
c(t)=
h
1+0:299e
;16:2t
+1:368e
;3t
cos(2t;2:743)
i
1(t)
The time response is shown in Figure 7. The actual per centovershoot
around twopercent, and the time to peak is 0.8 s.
6
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
0.00 1.00 2.00 3.00
Time in Seconds
Desired Steady State
Total Response
1.368exp(-3t)cos(2t-2.744)
0.3exp(-16.2t)
Figure 7: Step Response of Compensated System
7