Solution: 7.9.9.11
R
+
C
G
c
G
p
Σ
-
Figure 1: Cascade Compensation with UnityFeedback
For the system of Figure 1 wehave
G
p
=
4000(s+6)(s+15)
s(s +8)(s+3;j12)(s+3+j12)
:
The system is unusual in that it has twozeros, and these zeros greatly reduce
the complexityofthe design. The designer of the
ap control has also done
us a big favor by providing an enormous amountofgain.
We rst consider a compensator with tworealzeros and tworeal poles,
G
c
(s)=
1:6(s+1:2)(s+3:3)
s(s +140)
:
Wehavechosen to add another integrator. Since the requirementonK
v
is
so large, wemightaswell go ahead and make the system type 2. For this
gain, the closed loop system is
T
c
(s)=
6400(s+1:2)(s+3:3)(s+6)(s+15)
(s +40:01)(s+6:974)(s
2
+3:273s+2:808)(s
2
+103:7s+ 2911)
:
The tworeal poles are not dominant. Wehavetwosetsofcomplex poles
s
2
+3:273s+2:808 = (s +1:6365;j:3603)(s+1:6365+ j1:3503);;
and
s
2
+ 103:7s+ 2911 = (s +51:85;j14:91)(s+51:85+ j14:91):
The root locus is shown in Figure 2 The rst set of complex poles is almost
critically damped. The second set of complex poles are marked by the boxes
in Figure 2 The step response of this system is shown in Figure 3. If we
1
-140 -120 -100 -80 -60 -40 -20 0 20
-50
-40
-30
-20
-10
0
10
20
30
40
50
Real Axis
Imag Axis
Figure 2: Root locus for Proposed Compensator
2
Time (sec.)
Amplitude
Step Response
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Figure 3: Step Response for K
c
=1:6
3
Time (sec.)
Amplitude
Step Response
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Figure 4: Step Response for K
c
=2:5
4
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0
-6
-4
-2
0
2
4
6
Real Axis
Imag Axis
Figure 5: Root Locus 0 < 4000K
c
< 75)
increase the gain of the compensator to 2.5, then
T
c
(s)=
10;;000(s+1:2)(s+3:3)(s+6)(s +15)
(s +21:12)(s+6:756)(s+2:152)(s+1:44)(s
2
+122s +8059)
:
The step response is shown in Figure 4. As can be seen, weget a little more
overshoot and a little less undershoot. Wecertainly meet the specications
on percentovershoot and rise time with either of these systems.
As an alternativeapprochwetry cancelling the complex poles. Wecould
build suchacompensator using the Tow-Thomas biquadratic opamp circuit
from Chapter 3. Consider the compensator
G
c
(s)=
K
c
(s +3;j12)(s+3+j12)
(s +7)(s +20)
:
Then Figure 5 shows the rlocus for 0 < 4000K
c
< 75. If wechoose
K
c
=
75
4000
=0:01875
5
Time (sec.)
Amplitude
Step Response
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Figure 6: Step Response for 4000K
c
=75
6
Time (sec.)
Amplitude
Step Response
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Figure 7: Step Response for Near Cancellation
Then the step response is shown in Figure 6. Since wenever achieveperfect
cancellation weconsider
G
c
(s)=
K
c
(s +3:2;j13)(s+3:2+j13)
(s +7)(s +20)
:
The step response is shown in Figure 7 Wenow just makesome minor
adjustments. The following dialogue yields a compensator that gives the
step response in Figure 8.
EDU>g = zpk([-6 -15 -3.2+j*13 -3.2-j*13],[0 -6.2 -8 -20 -3+j*12 -3-j*12],60)
Zero/pole/gain:
60 (s+6) (s+15) (s^2 + 6.4s + 179.2)
---------------------------------------
s (s+6.2) (s+8) (s+20) (s^2 + 6s + 153)
EDU>tc = feedback(g,h)
Zero/pole/gain:
7
Time (sec.)
Amplitude
Step Response
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Figure 8: Step Response for Near Cancellation for Final Compensator
60 (s+15) (s+6) (s^2 + 6.4s + 179.2)
----------------------------------------------------------------
(s+18.8) (s+5.924) (s^2 + 10.12s + 60.82) (s^2 + 5.349s + 142.8)
EDU>step(tc,2)
EDU>print -deps pitchstepe.eps
EDU>
The compensator is
G
c
(s)=
0:015(s+3:2;j13)(s+3:2+j13)
(s +6:2)(s+20)
:
There are, of course, better solutions whichtheinterested student can nd
with a little eort.
8