Control Systems Engineering

Steady-State Errors

Department of Electrical Engineering

Mutah University

1
Introduction

We now focus on an important design specification,

which is : the steady-state error.

We define steady-state error to be the difference between

input and ouput as t .

We will see that control system design typically means we will

have to make trade-offs between the desired transient,
steady-state, and stability specifications.

c
2
Test Inputs

Table below shows the standard test inputs typically used for
evaluating steady-state error.

Table 7.1.

c
3
Choosing a Test Inputs
The test inputs we will choose for our steady-state analysis
and design depends on our target application.

Figure 7.1.

c
4
Steady-State Error and Stable Systems

The calculations we will be deriving for steady-state apply

only to stable systems.

Unstable systems represent loss of control in steady-state as

the transient response swamps the forced response.

As we analyze and design a system for steady-state error, we

must constantly check the system for stability.

c
5
Evaluating Steady-State Errors : Let us examine the concept of staedy-state error.

In this figure, a step input and two possible

outputs are shown. Output 1 has zero
steady-state error, and output 2 has a finite
steady-state error. A similar example which
has a ramp input is shown in the following
figure.

Output 1 has a zero steady-state error.

Output 2 has a finite steady-state error.
Output 3 has a infinite steady state error
as time goes to infinity. Let us now look at
the error from the perspective of the most
general block diagram.

Since the error is the difference between

the input and the output of a system, we
assume a closed loop system, the error
E(s) is the difference between the output
C(s) and the input R(s) for a unity feedback
system as shown in the figure.
 STEADY STATE ERROR FOR UNITY FEEDBACK SYSTEMS
Steady-state error can be calculated from a systems closed loop transfer function, T(s),
or the open loop transfer function, G(s), for unity feedback systems as shown in figure.

Figure 7.3
Closed-loop control
system error:
a. general representation;
b. representation for
unity feedback systems

We begin by deriving the systems steady-state error in terms of the closed loop transfer
function, T(s), in order to introduce the subject and definitions. Next we obtain insight
into the factors affecting steady-state error by using the open loop transfer function,
G8S), in unity feedback systems for our calculations. Later in this lecture we generalize
this discussion to nonunity feedback systems.
Steady-State Error in Terms of T(s) : Consider the figure(a). To find E(s), we write
E(s)=R(s) - C(s) , but C(s) = R(s)T(s).
Substituting C(s) into R(s) and solving for E(s) yields E(s) = R(s)[1-T(s)] . Although the
last equation allows us to solve for e(t) at any time t, we are interested in the final value
of the error, e(). Appliying the final value theorem, we obtain
e() = lim e(t ) = lim sE ( s )
t s 0

Substituting the last equation E(s) = R(s)[1 - T(s)] to the last equation we obtained yields
e() = lim sR( s)[1 T ( s)]
s 0
Example : Find the steay-state error for the system of the following figure if
T(s)=5/(s2+7s+10) and the input is a unit step.

s 2 + 7s + 5
Solution : R(s)=1/s , T(s)=5/(s2+7s+10) , E(s)=R(s)[1-T(s)] E ( s) =
s(s 2 + 7s + 10)
Since T(s) is stable and, subsequently, E(s) does not have right half plane poles or j
poles other than at the origin, we can apply the final value theorem.
e() = lim e(t ) = lim sE(s) e() = 1/ 2
t s0

Steady-state Errors in Terms of G(s) : Consider the feedback control system shown in
the figure. E(s) = R(s) C(s) and C(s) = E(s)G(s) E ( s ) = R ( s )
1 + G (s)

Applying the final value theorem yields

sR ( s )
e ( ) = lim
s 0 1 + G (s)
Sources of Steady-State Error
Steady-state errors can arise from nonlinear sources, such as
backlash in gears or motors requiring a minimum input
voltage before it starts to move.
Steady-state errors can also arise from configuration of system
and the input we apply.
Consider a step input applied to the system below which has
constant gain.
If a unity feedback system has a feedforward transfer function
G(s), then we can derive the transfer function E(s)
R(s) as follows:

C(s) = E(s)G(s) (1)

E(s) = R(s) C(s) (2)

c
Figure 7.4. 9
 Sources of Steady-State Error - II
Substituting equation 1 into equation 2 gives:
E(s) = R(s) E(s)G(s)
E(s)[1 + G(s)] = R(s)
E(s) 1
= (3)
R(s) 1 + G(s)

Steady-State Error and T (s)

In Diagram below, we have E(s) = R(s) C(s).
We also have:
C(s) = R(s)T (s) (7)
Combining the two we get
E(s) = R(s) R(s)T (s) = R(s)[1 T (s)] (8)
We thus have
ess = lim s E(s)
s0
= lim s R(s)[1 T (s)] (9)
s0

Figure 7.3.

c
12
Steady-State Error and G(s)
From equation 3, we have

R(s)
E(s) = (10)
1 + G(s)
We thus have

ess = lim s E(s)

s0
R(s)
= lim s (11)
s0 1 + G(s)

Figure 7.3.

c
13
STEADY-STATE ERROR FOR NONUNITY FEEDBACK SYSTEMS
Control system often do not have unity feedback because of the compensation used to
improve performance or because of the physical model foe the system. A general
feedback system, showing the input transducer, G1(s), controller and plant, G2(s), and
feedback, H1(s), is shown in the following figure. Pushing the input transducer, G (s),
1
controller and plant, G2(s), and
feedback, H1(s), is shown in figure(b),
where G(s)=G1(s)G2(s) and
H(s)=H1(s)/G1(s). To convert a
nonunity feedback system to a unity
feedback system, form a unity
feedback system by adding and
subtracting unity feedback paths, as
shown in figure(c). This step requires
the input and the output units be the
same. Next combine H(s) with the
negative unity feedback as shown in
figure(d). Finally combine the
feedback system consisting of G(s)
and [H(s) 1] as shown in figure(e).
Notice that the final figure shows
E(s)=R(s)-C(s) explicitly and we can
use all algorithms axplained before.
Example : For the system shown in the figure, find the steady state error for a unit step
input.

100 1
Solution : Note that G (s) = and H(s) =
s ( s + 10) s+5
Using the equivalent unity feedback system block diagram shown in figure(e) in the
previous page, the equivalent forward path transfer function Ge(s) is calculated as
G (s) 100
Ge (s) = = 3
1 + G ( s ) H ( s ) G ( s ) s + 15 s 2 50 s 400
We can calculate the steady-state error anymore :
1 1
e() = = = 4
1 + lim Ge (s) 100 5
s 0 1+
400
The negative value for steady-state error implies that the output step is larger than the
input step.
 STEADY-STATE ERROR FOR DISTURBANCES
Feedback control systems are used to compansate for disturbance or unwanted inputs
enter a system The following figure shows a feedback control system with a disturbance,
D(s), injected between the controller and the plant. We now re-derive the expression for
steady-state error with the disturbance included.
The transform of output is given by
C(s) = E(s)G1(s)G2(s) + D(s)G2(s)
Substituting C(s) = R(s) E(s) into the output
equation and solving for E(s) yields
1 G2 ( s )
E ( s) = R( s ) D( s )
1 + G1 ( s)G2 ( s) 1 + G1 ( S )G2 ( s)

where we can think of 1/[1+G1(s)G2(s)] as a transfer function relating E(s) to R(s) and
-G2(s)/[1+G1(s)G2(s)] as a transfer function relating E(s) to D(s). To find steady-state
value of the error, we apply the final value theorem to equation of E(s) and obtain
s sG2 (s)
e() = limsE(s) = lim R(s) lim D(s)
s0 s0 1 + G (s)G (s) s0 1 + G (s)G (s)
1 2 1 2
where
s sG 2 ( s )
eR () = lim R( s ) and e D ( s ) = lim D(s)
s 0 1 + G ( s )G ( s ) s 0 1 + G ( s )G ( s )
1 2 1 2

The first term eR() is the steady-state error due to R(s), which we have already
obtained. The second term eD() is the steady-state error due to disturbance. Let us
explore the conditions on eD() that must exist to reduce the error due to disturbance.
Steady-State Error and Disturbances - III
The eR () term is the steady-state error due to input R(s)
that we have already seen.
The eD () term is the steady-state error due to D(s).
If D(s) = 1/s (step input), we have
1
eD () = (32)
1
lims0 + lims0 G1 (s)
G2 (s)
If we set R(s) = 0, we get from Eqn28 the transfer function:
E(s) G2 (s)
= (33)
D(s) 1 + G1 (s)G2 (s)

c
Figure 7.12. 25