The steady state error

Two definitions of error

E(s) R(s) C (s)

nt

r t

G1 s

G2 s

or

bt

E(s) R(s) B(s)

H s

The steady state error of control system

ess lim e(t )

t

e(t ) L {E (s)}

ct

The steady state error

Here we define:

nt

r t

G1 s

G2 s

ct

E(s) R(s) B(s)

bt

H s

E s er s R( s) en s N s

1
G2 s H s
R( s )
N ( s)
1 G1 s G2 s H s
1 G1 s G2 s H s

Final Value Theorem

Suppose f (t ) has the Laplace transform F (s) and the

Limits lim f (t ) and lim sF ( s) exist. Then the final value
t
s0
Is
Hence e lim e t lim sE

lim f (t ) lim sF ( s)
t

s 0

ss

s 0

Relationship between system and steady-state error

only caused by input signal
Rs

Define: E(s) R(s) B(s)

Bs

Open-loop transfer function

Gs H s
s T s 1T

2 T s 1

K j s 1 l2 s 2 2 l l s 1
v

2 2
k

G s

C s

H s

K open loop gain

k k

In terms of to the number of the poles of G(s)H(s) at s=0

(corresponding with the number of the integral elements in the
open loop systems), we name:

=0 type 0 system
=1 type system
=2 type system

Relationship between system and steady-state error

only caused by input signal
Gs H s

2 T s 1

K j s 1 l2 s 2 2 l l s 1

s v Ti s 1 Tk2 s 2

k k

1
ess lim sE ( s) lim s
R( s)
s 0
s 0 1 G ( s ) H ( s )
1
lim s
R s

s 0
1 K s
s 1
lim
R s
s 0 s K

Relationship between system and steady-state error

only caused by input signal
Rs

Bs

G s

H s

1
ess lim sE ( s) lim s
R( s )
s 0
s 0 1 G ( s ) H ( s )
1) Step input

R
R s
s

r (t ) R 1(t )
R
R

s 0 1 G ( s ) H ( s )
1 KP

ess lim

The position error constant: K P lim G( s) H ( s)

s 0

C s

Relationship between system and steady-state error

only caused by input signal
Rs

Bs

G s

H s

1
ess lim sE ( s) lim s
R( s )
s 0
s 0 1 G ( s ) H ( s )
2) Ramp input

V
Rs 2
s

r (t ) V t
V
V
ess lim

s 0 sG( s ) H ( s )
KV

The velocity error constant:

KV lim sG(s) H (s)

s 0

C s

Relationship between system and steady-state error

only caused by input signal
Rs

Bs

G s

H s

1
ess lim sER ( s) lim s
R( s )
s 0
s 0 1 G ( s ) H ( s )
3) Parabolic input

A
R s 3
s

r (t )

1
At3
2

A
A
ess lim 2

s 0 s G ( s ) H ( s )
Ka

The parabolic error constant:

K a lim s 2G(s) H (s)

s 0

C s

Summy of the steady-state errors

Type of
system

0
1

Error constant

Kp

KV

0
K

Ka

0
0

Steady-state error
Step input

r (t ) R 1(t )

Ramp Input

r (t ) V t

ess

Parabolic Input
r (t )

1
At2
2

R
ess
1 K p

ess 0

V
ess
KV

A
Ka

ess 0

ess 0

ess

ess 0

ess 0

ess 0

Summy of the steady-state errors

We

emphasize often enough that, for these table results to be valid,

the closed-loop system must be stable.
By

using the method described, the steady-state error of any linear

closed-loop system subject to an input with order higher than the
parabolic function can also be derived if necessary.
As

a summary of the error analysis, Table shows the relations among

the error constants, the types of systems, and the input types with
reference to the following conditions: Rs
C s
E
G s

The system configuration

Bs

H s

The error defination E(s) R(s) B(s)

Open-loop transfer function

Gs H s

2 T s 1

K j s 1 l2 s 2 2 l l s 1

s v Ti s 1 Tk2 s 2

k k

Summy of the steady-state errors

As a summary, the following points should be noted when applying
the error-constant analysis just presented.

1. The step-, ramp-, or parabolic-error constants are significant for the

error analysis only when the input signal is a step function, ramp
function, or parabolic function, espectively.
2. Because the error constants are defined with respect to the
forward-path transfer function G(s)H(s), the method is applicable to
only the system configuration shown in the above figure . Because the
error analysis relies on the use of the final value theorem of the
Laplace transform, it is important to check first to see if sE(s) has any
poles on the j-axis or in the right-half s-plane.
3. The steady-state error properties summarized in the Table are for
systems with the above figure and the error defination only.

Summy of the steady-state errors

4. The steady-state error of a system with an input that is a linear
combination of the three basic types of inputs can be determined by
superimposing the errors due to each input component.
5. When the system configuration and the error defination differ from the
above , we can either simplify the system to the form of above figure or
establish the error signal and apply the final-value theorem. The error
constants defined here may or may not apply, depending on the dividual
situation.
6. When the steady-state error is infinite, that is, when the error increases
continuously with time, the error-constant method does not indicate how
the error varies with time. This is one of the disadvantages of the errorconstant method.
7.The error-constant method also does not apply to systems with inputs
that are sinusoidal, since the final-value theorem cannot be applied.

Steady-state error only caused by

the disturbance signals
nt
r t

G1

ct

G2

ess lim sE ( s )
s 0

lim sER ( s ) lim sEN ( s )

s 0

s 0

lim s
s 0

R( s)
G2
lim s
N (s)
s

0
1 G1G2
1 G1G2

e ssr e ssn

When R(s)=0 , N(s)=1/s

EN ( s)

G2
N ( s)
1 G1G2

G
essn lim s

2 N (S )
s 0
1 G1G2

set G1 ( s) K1 , G2 ( s)

K2
s

sK2
1
1
essn lim sEN ( s) lim

s 0
s 0 s K K
K1
1 2 s

Steady-state error only caused by

the disturbance signals
nt
r t

G1

G2

When R(s)=0 , N(s)=1/s

ct

G2
EN ( s)
N ( s)
1 G1G2

G
essn lim s

2 N (S )
s 0
1 G1G2

K2
set G2 ( s)
s
if G1 ( s)

K1 (1 s)
s

sK2
1
1
essn lim sEN ( s) lim
lim
s 0
s 0 s G ( s ) K
s 0 G ( s )
1
2 s
1
K1 0, K 2 0, 0

Stability and

essn=0