1

University of Hertfordshire DAG 29/09/03

Faculty of Engineering & Information Sciences
Control Systems Formula Sheet


1. SYSTEMS MODELLING


Time Domain Laplace Domain
f(t)
F(s) or L[f(t)]
a.f(t) + b.g(t) (a and b constant) a.F(s) + b.G(s)
= = f t f t
df
dt
( )
&
( )
sF s f ( ) ( ) 0
= f t
d f
dt
( )
2
2

s F s sf f
2
0 0 ( ) ( ) ( )
d f
dt
n
n

s F s s f s f f
n n n n
( ) ( ) ( ) . . . . ( )
( )

1 2 1
0 0 0
f t dt
t
( )
0

1
s
F s ( )
f(t-T) (T is a time delay) e
-sT
F(s)
(t) 1
1
s
1

t
n
n
s
n
!
+1

e
-at
1
s a +

t
n
e
-at

( )
n
s a
n
!
+
+1

sin(bt) b
s b
2 2
+

cos(bt) s
s b
2 2
+

e
-at
sin(bt)
( )
b
s a b + +
2
2

e
-at
cos(bt)
( )
s a
s a b
+
+ +
2
2


Initial and Final Value Theorem

| | | |
f f t sF s
t s
( )
lim
( )
lim
( ) 0
0
= =


| | | |
f f t sF s
t s
( )
lim
( )
lim
( ) = =
0


2


Transfer Function Definition

| |
| |
G(s) =
y(t)
u(t)

Y(s)
U(s)

The Laplace Transform of the output
The Laplace Transform of the input
=
L
L


=

Assuming all initial conditions are zero.



System Transfer Function

Differential Equation

Gain

G(s) = K


y = Ku

Integrator

G s
s
( ) =
1


y u dt =

.


First Order

G s
K
sT
( ) =
+ 1



T
dy
dt
y Ku + =

Second Order

G s
K
s s
n
n n
( ) =
+ +

2
2 2
2



d y
dt
dy
dt
y K u
n n n
2
2
2 2
2 + + =


System

Unit Step Response (assuming all ics = 0)

First Order

y K e
t
T
=
|
\

|
.
|

1


Second Order


( ) ( ) y K e t t
n
t
d
n
d
d
= +
|
\

|
.
|

(
(

cos sin

where
d n
= 1
2
,

n
t
=

max
1
2
,

overshoot e =

1
2
,
( ) ( )
( ) ( )

=
+
ln
ln
overshoot
overshoot
2
2
2



3

Rules for Block Diagram Manipulation

G
A
(s) G
B
(s)

G
A
(s)G
B
(s)

+
G
A
(s)
G
B
(s)

G
A
(s) G
B
(s)

G(s)


=
G(s)
G(s)


G(s)


1
G(s)
G(s)

G(s)
+

G(s)
G(s)
+


G(s)

G(s)
1
G(s)
+


G(s)
+
H(s)


G(s)
1 + G(s)H(s)




4
2. SYSTEM PERFORMANCE



Steady State Output

y
t
y t
s
sY s
ss
=

=

lim
( )
lim
( )
0



Steady State Error

=
ss
e u - y
ss





Steady State Error for Unity Feedback Systems

Step
u(t) = 1
Ramp
u(t) = t
Parabolic
u(t) = t
2
/2

Type 0

e
K
ss
p
=
+
1
1



e
ss
=

e
ss
=

Type 1

e
ss
= 0

e
K
ss
v
=
1



e
ss
=

Type 2

e
ss
= 0

e
ss
= 0

e
K
ss
a
=
1




Position Error Constant

| |
K G s
p o
=

lim
s 0
( )

Velocity Error Constant

| |
K sG s
v o
=

lim
s 0
( )

Acceleration Error Constant

| | ) s ( G s
0 s
lim
K
o
2
a

=


5
3. SIMPLE CONTROLLERS


Controller
G
c
(s)

Proportional

K


Proportional + Integral

K
K
s
i
+
or K
s
s
+ |
\

|
.
|

where =
K
K
i


Proportional + Derivative

K K s
d
+

or ( ) K sT
d
1+ where T
K
K
d
d
=

Proportional + Integral + Derivative
(Three Term Controller)

K
K
s
K s
i
d
+ +

Lead Controller
b s
a s
K
+
+
where a < b

or
T s 1
sT 1
K
+
+
where < 1
Lag Controller
b s
a s
K
+
+
where a > b

or
T s 1
sT 1
K
+
+
where > 1


4. ROUTH STABILITY CRITERION

D(s) = a
n
s
n
+ a
n-1
s
n-1
+ a
n-2
s
n-2
+ a
n-3
s
n-3
+ a
n-4
s
n-4
+ a
n-5
s
n-5
+ ... = 0


s
n
a
n
a
n-2
a
n-4


s
n-1
a
n-1
a
n-3
a
n-5


s
n-2
b
n-1
b
n-3
b
n-5

. . . .
. . . .
. . .
s
0
h
n-1

1 n
3 n n 2 n 1 n
1 n
a
a a a a
b

=


1 n
5 n n 4 n 1 n
3 n
a
a a a a
b

=

6
5. ROOT LOCUS


No
Drawing Rules

1

All loci start for K = 0 at the Open Loop Poles and finish for K = at either Open Loop Zeros or s = .


2

There will always be a locus on the real axis to the LEFT of an ODD number of Open Loop Poles and Zeros.


3

If there are n Open Loop Poles and m Open Loop Zeros, there will be n-m loci ending at infinity on asymptotes
at angles to the real axis of

180 540 900

n m n m n m
, , , K


4

The asymptotes meet on the real axis at
=
p - z
n - m
i
i
n
i
i
m
= =

1 1

where p
i
is the position of the ith open loop pole and z
i
is the position of the ith open loop zero.


5

Where two real loci meet on the real axis they breakaway from the real axis at 90 to form two complex
loci, symmetrical about the real axis.

Where two complex loci meet on the real axis they break-in to form two real loci, moving in opposite
directions along the real axis.
The breakaway and break-in points are given by the roots of
dK
ds
= 0


6

The points of intersection of a locus with the imaginary axis can be determined by solving the equation

D j KN j ( ) ( ) + = 0

Remember both the real part and the imaginary parts must be satisfied in this equation. Hence this will give two
simultaneous equations one that will give values of while the other gives the K value at the crossing point.


7

The angle of departure of a locus from a complex open loop pole is given by;

d i
i
i d
n
i
i
m
= 180 +
=

=

1 1

where
i
is the angle from the ith open loop pole and
i
is the angle from the ith zero.

The angle of arrival of a locus at a complex zero is given by;

a i
i
n
i
i
i a
m
= 180 +
= =

1 1


7


Magnitude Condition

=
=
=
m
1 i
i
n
1 i
i
Z
P
K


Angle Condition

180 = -
i
i
n
i
i
m

= =


1 1



Lines of constant damping

=
2
- 1
where s = + j

NB Straight line through the origin of the s plane. Also lines makes an
angle cos
-1
with the negative real axis


Lines of constant undamped
natural frequency



2
n
2 2
= +

NB Circle with centre on the origin of the s plane and radius
n




6. FREQUENCY RESPONSE METHODS


u(t) = sin(t)
y(t) = R.sin(t + )
G(s)

R G j a b = = + ( )
2 2


= =
|
\

|
.
|

G j
b
a
( ) tan
1


G(j ) = a( ) + jb( )

g R = 20
10
log

y(t) = R.sin(t + )
u(t) = sin(t)
G
2
(s) G
1
(s)
G
3
(s)



R R R R =
1 2 3



g g g g = + +
1 2 3



= + +
1 2 3




8
Nyquist Diagrams of Common System Elements



G(s)


R



Nyquist Diagram

Gain Term


K



K



0

K


Integrator

1
s



1

-90



Differentiator


s







90



First Order Lag


1
1+ sT




1
1
2 2
+ T




( )

tan
1
T

1


First Order Lead


1+ sT



1
2 2
+ T



( ) tan
1
T

1


Second Order Term

n
n n
s s
2
2 2
2 + +



( )
( )

n
n n
2
2 2
2
2
2 +

|
\

|
.
|

tan
1
2 2
2

n
n




Pure Time Delay

e
sT
D

1


T
D


(in radians)




9
Bode Plots of Common System Elements




G(s)



g (dB)

(deg)


Gain Plot


Phase Plot

Gain Term

K


20log
10
K

0

2 0 lo g 1 0 K


0


Integrator

1
s


20
1
10
log

|
\

|
.
|


-90

1
-20 dB/dec
0


-90
0


Differentiator

s

( ) 20
10
log

90

1
20 dB/dec
0


90
0


First Order Lag

1
1+ sT


20
1
1
10
2 2
log
+
|
\

|
.
|
|
T


( )

tan
1
T


1/T
-3dB
0
-20 dB/dec


1/T
-45
0
-90
10/T 0.1/T


First Order Lead

1+ sT

20 1
10
2 2
log +
|
\

|
.
| T

( ) tan
1
T


1/T
20 dB/dec
3dB
0


1/T
45
0
90
10/T 0.1/T


Second Order Term

n
n n
s s
2
2 2
2 + +


( )
( )
20
2
10
2
2 2
2
2
log


n
n n
+
|
\

|
.
|
|
|
|

|
\

|
.
|

tan
1
2 2
2

n
n

n
-40dB/dec
-20log10(2)
0

n
-90
180
0



10
7. DIGITAL CONTROL


Discrete Time Domain z Domain
(T = sample time period)

Laplace Domain
f(k) F(z) or Z[f(k)] F(s)
a.f(k) + b.g(k) a.F(z) + b.G(z) a.F(s) + b.G(s)
f(k+1) zF(z) - zf(0) e
sT
F(s)
f(k-1) z
-1
F(z) e
-sT
F(s)
(k) 1 1
(k-n) z
-n
e
-snT

1 z
z 1

1
s

k
( )
Tz
z 1
2

1
2
s

2
k
! 2
1

(

+
3
2
) 1 z (
) 1 z ( z
! 2
T

3
s
1

e
-ak
z
z e
aT

1
s a +

ke
-ak

( )
Tze
z e
aT
aT

2

( )
1
2
s a +

1 - e
-ak

( )
z( e
z z e
aT
aT
1
1

)
( )

( ) a s s
a
+

k - (1 - e
-ak
)/a
| |
( ) ) e z ( 1 z a
aTe e 1 ( ) e 1 aT ( z z
aT 2
aT aT aT

+ +

( ) a s s
a
2
+

sin(ak) z aT
z z aT
sin( )
cos( )
2
2 1 +

2 2
a s
a
+

cos(ak) z(z aT
z z aT

+
cos( ))
cos( )
2
2 1

2 2
a s
s
+

e
-ak
sin(bk)
ze bT
z ze bT e
aT
aT aT

+
sin( )
cos( )
2 2
2

( )
b
s a b + +
2
2

e
-ak
cos(bk)
z(z e bT
z ze bT e
aT
aT aT

cos( ))
cos( )
2 2
2

( )
s a
s a b
+
+ +
2
2

Zero Order Hold
(A/D and D/A
converters)

G
H
(z)

1

e
s
sT




11
8. STATE SPACE METHODS


Continuous


Digital

Standard
Form for State
Space Model

& x Ax Bu
y Cx Du
= +
= +



) k ( u D ) k ( x C ) k ( y
) k ( u G ) k ( x F ) 1 k ( x
+ =
+ = +

Transfer
Function
Matrix


( ) G s C sI A B D ( ) = +
1



( ) G z C zI F G D ( ) = +
1

Characteristic
Equation


( ) det sI A = 0


( ) det zI F = 0
State Time
Response

x t e x e Bu d
At A
t
( ) ( ) ( ) = +

(
(

0
0

x k F x F Gu k i
k i
i
k
( ) ( ) ( ) + = +
+
=

1 0
1
0

Transition
Matrix

| |
( ) ( ) t e sI A
At
= =

L
-1 1


L + + + + = =
! 3
t A
! 2
t A
At I e ) t (
3 3 2 2
At


F e
AT
=


| |
G A F I B =
1

State
Feedback
Equation


( )
& x Ax B u Kx = +


( ) ) k ( x K ) k ( u G ) k ( x F ) 1 k ( x + = +
Characteristic
Equation with
state feedback



( ) det sI A BK + = 0



( ) det zI F GK + = 0
Controllability
Matrix

| | c
n
M
= B AB A B A B M M MLM
2 1



| | c
n
M
= G FG F G F G M M MLM
2 1

Controllable
Canonical
Form

| |
& x
a a a a
x u
y b b b b x
c
n
c
n n c
=

(
(
(
(
(
(
+

(
(
(
(
(
(
=

0 1 0 0
0 0 1 0
0 0 0 1
0
0
0
1
0 1 2 1
0 1 2 1
L
L
M M M O M
L
L
M
L L



| | ) k ( x b b b b ) k ( y
) k ( u
1
0
0
0
) k ( x
a a a a
1 0 0 0
0 1 0 0
0 0 1 0
) 1 k ( x
c 1 n 2 n 1 0
c
1 n 2 1 0
c

=
(
(
(
(
(
(

+
(
(
(
(
(
(

= +
L L
M
L
L
M O M M M
L
L


12


Ackermanns
Formula for
State Feedback

If desired CE is
s s s
n
n
n
+ + + + =

1
1
1 0
0 L

| | ( ) K M A
c
=

0 0 1
1
K
where
( ) A I A A A
n
n n
= + + + +

0 1 1
1
L


If desired CE is
z z z
n
n
n
+ + + + =

1
1
1 0
0 L

| | ( ) K M F
c
=

0 0 1
1
K
where
( ) F I F F F
n
n n
= + + + +

0 1 1
1
L
State
Estimator
Equation


&
$
$ (
~
$) x Ax Bu P y y = + +


)) k ( y ) k ( y
~
( P ) k ( Gu ) k ( x F ) 1 k ( x + + = +
Characteristic
Equation for a
State
Estimator


( ) det sI A PC + = 0

( ) det zI F PC + = 0
Observability
Matrix

o
n
M
=
C
CA
CA
CA
2
1
L

(
(
(
(
(
(
(



o
n
M
=
C
CF
CF
CF
2
1
L

(
(
(
(
(
(
(

Observable
Canonical
Form

| |
& x
a
a
a
a
x
b
b
b
b
u
y x
o
n
o
n
o
=

(
(
(
(
(
(
+

(
(
(
(
(
(
=

0 0 0
1 0 0
0 1 0
0 0 1
0 0 0 0 1
0
1
2
1
0
1
2
1
L
L
M
L L O L M
L
M
L



| | ) k ( x 1 0 0 0 0 ) k ( y
) k ( u
b
b
b
b
) k ( x
a 1 0 0
a 0 1 0
a 0 0 1
a 0 0 0
) 1 k ( x
o
1 n
2
1
0
o
1 n
2
1
0
o
L
M
L
M L O L L
M
L
L
=
(
(
(
(
(
(

+
(
(
(
(
(
(

= +



Ackermanns
Formula for
State Estimators

If desired CE is
s s s
n
n
n
+ + + + =

1
1
1 0
0 L

( ) | | | |
P A M
o
T
=

1
0 0 1 K
where
( ) A I A A A
n
n n
= + + + +

0 1 1
1
L


If desired CE is
z z z
n
n
n
+ + + + =

1
1
1 0
0 L

( ) | | | |
P F M
o
T
=

1
0 0 1 K
where
( ) F I F F F
n
n n
= + + + +

0 1 1
1
L