Solution 6.8.1.9
Referring to Figure 1 First note that
X X
Re(s)
Im(s)
-3
-1
-2
ζ = 0.9 ω
d
Figure 1: Gain to place poles with =0:8
!
n
=
;;
where, in this case
=2
Then
!
d
= !
n
q
1;
2
=
p
1;
2
:
Thus in the present case
!
d
=
2
p
1;0:9
2
0:9
=0:9686
Then the dominantpoles will be at
s = ;2j0:9686:
1
The gain to place the poles at the desired locations is
K =
p
2
2
+0:9686
2
p
2
2
+0:9686
2
= 1:9383:
Then the closed loop transfer function is
T
c
(s)=
1:9383
(s +2;j0:9686)(s+2+j0:9686)
:
The MATLAB dialogue
EDU>g = zpk([],[-1 -3],1.9383)
Zero/pole/gain:
1.9383
-----------
(s+1) (s+3)
EDU>tc = feedback(g,1)
Zero/pole/gain:
1.9383
------------------
(s^2 + 4s+4.938)
EDU>step(tc)
EDU>print -deps sr6819.eps
EDU>
produces the step response shown in Figure 2 As can be seen there is a
steady state error of about 60%. Wecannd this is twoways. Since we
have unityfeedbackwecannd
K
p
= lim
s!0
1:9383
(s + 1)(s+3)
=
1:9383
3
= 0:6461:
Then
e
ss
=
1
1+K
p
2
Time (sec.)
A
mp
li
tu
d
e
Step Response
0 0.5 1 1.5 2 2.5 3
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
Figure 2: Unit step response of compensated system
3
=
1
1+0:6461
= 0:6075:
The other waytond the steady state error is to evaluae
e
ss
= = 1; lim
s!0
T
c
(s)
= 1; lim
s!0
1:9383
(s +2;j0:9686)(s+2+j0:9686)
= 1;0:3925
= 0:6075:
4