aa
262 Chapter 5 / Transient and Steady-State Response Analyses
If the input is a unit ramp, then the steady-state error is
Therefore, if kis chosen as
then the steady-state error for following a ramp input can be made equal to zero. Note that, if there
are any variations in the values of zand/orvndue to environmental changes or aging, then a
nonzero steady-state error for a ramp response may result.
A–5–26. Consider the stable unity-feedback control system with feedforward transfer function G(s).
Suppose that the closed-loop transfer function can be written
Show that
wheree(t)=r(t)-c(t)is the error in the unit-step response. Show also that
Solution.Let us define
and
Then
and
For a unit-step input,R(s)=1/sand
E(s)=
Q(s)-P(s)
sQ(s)
E(s)=
Q(s)-P(s)
Q(s)
R(s)
C(s)
R(s)
=
P(s)
Q(s)
AT 1 s+ 1 BAT 2 s+ 1 BpATn s+ 1 B=Q(s)
ATa s+ 1 BATb s+ 1 B p ATm s+ 1 B=P(s)
1
Kv
=
1
slimS 0 sG(s)
=AT 1 +T 2 +p+TnB-ATa+Tb+p+TmB
3
q
0
e(t)dt=AT 1 +T 2 +p+TnB-ATa+Tb+p+TmB
C(s)
R(s)
=
G(s)
1 +G(s)
=
ATa s+ 1 BATb s+ 1 BpATm s+ 1 B
AT 1 s+ 1 BAT 2 s+ 1 BpATn s+ 1 B
(mn)
k=
2 z
vn
=
2 zvn-v^2 n k
v^2 n
=slimS 0 s a
s^2 + 2 zvn s-v^2 n ks
s^2 + 2 zvn s+v^2 n
b
1
s^2
e(q)=r(q)-c(q)
Openmirrors.com