B.

ENG (HONS) ELECTRICAL & ELECTRONIC ENGINEERING

NORTHUMBRIA UNIVERSITY, UNITED KINGDOM
Assignment Of
EN0631 Industrial Instrumentation and Modern Control Level

Northumbria University Student Number:

KDU student number:
Date of Submission:
Subject Lecturer:

12040328
21108
16/03/2015
Dr. Robin Tan

k
di(t)
R
1
i(t) 1 (t) u(t)
dt
L
L
L
d(t) k 2

i(t)
dt
J

Q1) A simplified model of a dc motor, is given

by
where i(t) = armature motor current, R = armature resistance (5 ohms),
u(t) = input voltage, L = armature inductance (200mH),

(t) = motor

speed,
J = motor inertia (0.02 kgm2), k1 = back-emf constant (0.1 V/rad/s)

and

k2 = torque constant (0.1 Nm/A).

a) By setting x1(t) = i(t) and x2(t) = (t) write the system in state-space
form by using the above numerical values.
(2 marks)
b) Assuming that the speed is measured, that is the output y(t) = (t),
state whether the system is stable and/or output/state controllable?
(3 marks)
c) Design an open-loop controller such that the speed has a constant
value of 0.25 rad/s. Show your result via simulation using Simulink.
Explain why open-loop control is not recommendable in practice.
(5 marks)
d) We wish to design an output feedback controller of the form u(t) =ky
+v in order to regulate the speed at a constant value. Give the range
of values of k for which the closed-loop system is stable.
(10 marks)
e) Show your result by carrying a simulation of your result using Simulink.
(10 marks)
f) Now, we wish to design an state feedback controller of the form u(t)
=kx +v in order to regulate the speed at a constant value. Show that
the range of values of k for which the closed-loop system is stable
larger than that obtained in Question d) for the output feedback. Give
the advantage for using the state feedback over the output feedback.
(10 marks)
g) For a fixed set of closed-loop poles of your choice, compare the state
feedback to the output feedback through simulation using Simulink.
(10 marks)

Q1 Answer
Q1 a)
Let

x 1 ( t )=i ( t )
1 ( t )=

di(t)
dt

---------------------------(a)

x 2 ( t )= ( t )
2 ( t )=

d (t)
dt

-----------------------------(b)

From (a) & (b)

1 ( t )=

R
K1
1
i(t)
( t ) + u(t )
L
L
L

1 ( t )=

5
0.1
1
x 1(t)
x 2( t)+
u( t)
200 m
200 m
200 m

2 ( t )=

k2
i(t )
J

2 ( t )=

0.1
x 1(t ) -----------------------------(d)
0.02

Form state-space equation using (c) & (d)

[ ][

][ ] [ ]

1 ( t )
25 0.5 x1 ( t ) 5
=
+ u
5
0
2 (t)
x2 ( t ) 0

-----------------------------(c)

Q1 b)

y (t )= ( t )

[ ]

( t ) =[ 01 ] x 1 ( t )
x2 ( t )

State control M =

[ B : AB ]

][ ]

AB= 25 0.5 5
5
0
0

[ ]

125
25

M= 5 125
0
25

Det = 125
0

The system is state controllable

Output control M =

[ CB : CAB ]

[]

CB=[ 0 1 ] 5
0
=0

][ ]

CAB=[ 0 1 ] 25 0.5 5
5
0
0
= 25

M =[ 0 25 ]

rank = 1, the system is output controllable

Stability of system

G ( s )=C [ SI A]1 B
1

{[ ] [

]} [ ]

G ( s )=[ 0 1 ] S 1 0 25 0.5
0 1
5
0

[0 1]

{[ ] [

s 0 25 0.5
0 s
5
0

[ 0 1 ] S +25 0.5
5
S

[ SI A ] =

]} [ ]
5
0

]} [ ]
5
0

1
S 0.5
( s+25 ) ( s ) +2.5 5 S+ 25

1
S 0.5
s +25 s+2.5 5 S +25

s
s +25 s +2.5
5
2
s +25 s +2.5

5
0

0.5
s + 25 s+2.5
s +25
2
s + 25 s+2.5

s
s +25 s+2.5
G ( s )= [ 0 1 ]
5
2
s +25 s+2.5

25
s +25 s+2.5

5
s + 25 s+ 2.5
2

)
)[ ]

0.5
s +25 s+ 2.5 5
s+25
0
2
s +25 s+ 2.5
2

][ ]

s+ 25
5
s +25 s+2.5 0
2

Therefore, the system is stable.

Q1 C)

= Ax+ Bu
A

[ ][]

x1 + 5 u
0.25 0

][ ] [ ]

25 0.5 x1 + 5 u
5
0
0.25 0

[][

0 = 25 x 10.125+5 u
0
5 x1

5 x1=0

---- equation

x 1=0
25 x 10.125+5 u

= 0 --- equation

0.125+ 5u=0

u=0.025

Comment: Open-loop control is not recommendable in practice because open-loop

system is a non-feedback system and it is a type of continuous control system in
which the output has no influence or effect on the control action of the input signal.
In other words, in an open-loop control system the output is neither measured nor

feedback for comparison with the input. Therefore, an open-loop system is expected
to faithfully follow its input command or set point regardless of the final result.
Furthermore, open-loop system has no knowledge of the output condition so cannot
self-correct any errors it could make when the preset value drifts, even if the results in
large deviations from the preset value.
Q1 d)

= Ax+ Bu
Given:

y=x 2

(State equation)

u(t)=ky (t)+ v (t)

so,

[ ]

( t ) =[ 01 ] x 1 ( t )
x2 ( t )

= Ax+ B(ky + v)

Ax + B(k x 2+ v )
Ax + B(kCx+ v )

Ax + BkCx+Bv
( A + BkC ) x+ Bv

new ( A+ BkC ) =

][]

25 0.5 5
+ k [ 0 1]
5
0
0

][ ]

25 0.5 + 5 k [ 0 1 ]
5
0
0

][

][ ] [ ]

( 25 0.5 + 0 5 k 5 k ) x 1
5
0
0 0 0 x2

][ ]

25 0.5+5 k x1
5
0
x2

[ ][

[ SI A ] =s 1 0 25 0.5+5 k
0 1

[ ][

s 0 25 0.5+5 k
0 s
5
0

[ SI A ] =

s+25 0.55 k
5
0

1
s +25 s +2.525 k
2

Rouths Criteriar
S2

2.5-25k

S1

25

S0

2.5 - 25k

2.525 k >0
k <0.1

Q1 e) 1st condition;
When k = 0.1

Comment: The system is marginally stable when k = 0.1.

2nd condition;
When k > 0.1

Comment: The system is unstable when k > 0.1.

.

3rd condition;
When k < 0.1

Comment: The system is stable when k < 0.1.

Q1 f)

Given:

u(t)=kx (t)+ v (t)

= Ax+ Bu

= Ax+ B(kx + v)
Ax + Bkx+ Bv

( A+ Bk )x + Bv

][]

new ( A+ Bk )=( 25 0.5 + 5 k )x

5
0
0

25 0.5 5
+ [ k 1 k 2 ]) x
5
0
0

][]

][

5 k 5 k2
( 25 0.5 + 1
)x
5
0
0
0

] []

25+5 k 1 0.5+5 k 2
x+ 5 v
5
0
0

[ ][

[ SI A ] =s 1 0 25+ 5 k 1 0.5+5 k 2
0 1

[ ][

25+5 k 1 0.5+5 k 2
s 0
0 s
5
0

[ SI A ] =

s+255 k 1 0.55 k 2
5
s

1
( s+255 k 1 ) ( s )[ (5 ) ( 0.55 k 2 ) ]

1
s+
255
k
(
s
)
[
(5 ) ( 0.55 k 2 ) ]
(
1)

1
s + (5 k 1 +25 ) s+ ( 2.525 k 2 )
2

Rouths Criteriar
S2

2.5 - 25k2

S1

-5k1+25

S0

2.5 - 25k2

5 k 1+25 >0
k 1 <5
2.525 k 2> 0
k 2 <0.1
Comment:

State feedback is easier but output feedback is more realistic.

State feedback gives complete control over placement of the closed loop
eigenvalues.
State feedback can position the poles at anywhere by using the pole placement
method.

Q1 g)
Set K1=4.5 (constant) , K2 as adjustable variable to observe the system respond.
1st condition;

Comment: From the result of simulink had a stable system because according to the
stable state is when K1<5 and k2<0.1. In other words, when k2 decrease, the result also
decrease. Therefore, the higher the value, the system is more stable.

2nd condition;

Comment: The system is marginally stable.

3rd condition;

Commment: The system is unstable when K2 > 0.1, since the K2 <0.1 in calculation.

Set K2=0.05 (constant) , K1 as adjustable variable to observe the system respond.

4th condition;

Comment: The system is stable.

5th condition;

Comment: The system is marginally stable.

6th condition;

Comment: The system is unstable when K1 > 5.

Q2) The transfer function of a linear system is given by

H(s)

Y(s)
50

U(s) s s 250

a) Simulate the step response of the above system using Simulink.

Comment on the figure obtained.
(5 marks)
b) Give a state space representation of the above system.
(5 marks)
c) Give a discrete-time representation of the above system using a zeroorder hold.
(15 marks)
d) Analyse the stability and controllability of the discrete time model and
comment on the result obtained.
(5 marks)
e)

Assuming T=0.1s, design a state feedback of the form u(n)=Kx(n)

+v(n) so that the closed-loop system is stable. Confirm your result
using Simulink.
(10 marks)

f) For T=0.1s, calculate the value of v(n) such that the value of y(n)
converges to 5 in steady state. Validate your result using Simulink.
(10 marks)

Q2 Answer
Q2a)

Comment: The system is marginally stable.

Q2b)
H (s)=

Y (S)
50
=
S + 250 S U (S)
2

50 u(s )=Y (s)( S2 +250 S )

50 u(s )=Y ( S) S 2+Y (S)250 S
50 U (S )=

d 2 y (t ) dy (t)
+(
)250
dt
dt 2

x 1(t)= y (t )
x 2(t)= x1=

dy (t)
dt

d 2 y (t )
x 2(t)=
dt 2

x 2(t)+ 250 x 2 (t)

50 u(s)=
x 2(t)=250 x2 (t )+50 u (t)

x 1( t)=x 2 (t)
x 2(t)=250 x2 (t )+50 u (t)

State space model

[ ][

][ ] [ ]

1 ( t )
0
1
x1 ( t )
=
+ 0 u
2 (t) 0 250 x 2 ( t ) 50

[ ]

[ y (t)] = [1 0 ] x 1 ( t )
x2 ( t )

e A =L1 {( sI A) }

Q2c)

[ ][

1
sI A= s 0 0
0 s
0 250

s
1
0 s +250

s+ 250
s (s +250)

[ sI A ] =

1
s
0

1
s( s+250)
1
(s +250)

1
s(s+ 250)
s
s(s+ 250)

Partial faction,

1
A
B
= +
s (s +250) s s+250
1= A( s+ 250)+ Bs
Let,

s = -250
1 = -250B

s=0
1 = 250A

B = -1/250

{(sI A ) }=
1

F=e At =

e At B=

1
0

][ ]

BdT =

15 dT 15 e250 T dT
0

50 e250T dT
0

[[ ]]
[

1
(1e250 T ) 0
250
50
e250 T

1 1 250T
e
5 5
50 e250T

1
(1e250T )
250
0
e250T

[
[

1
( 1e250T )
250
e250T

1
(1e250T )
5
50 e250 T

At

A = 1/250

1
1 250 T
T+
e
5
1250
1
e250T
5

T
0

T
0

[
[

1
1 250 T
1
T+
e

1250
1250
5
1 250T 1
e
+
5
5

1
1
250T
T+
(e
1)
5
1250

=G
1
250T
(e
+1)
5

][

X 1 ; n+ 1 =
X 2 ; n+ 1

Q2d)

1
0

[ ][

][

][ ] [ ]

1 ( t ) = 0
1
x1 ( t ) + 0 u
2 (t) 0 250 x 2 ( t ) 50

[ y ( t ) ]=[ 1 0 ]

1
1
1 250T
T+
(e250T 1)
(e
1) X 1 ; n
1250
250
+ 5
Un
1
250T
X2; n
250T
(e
+1)
e
5

[ ]
x1 ( t )
x2 ( t )

State control M = [ B : AB ]

][ ]

1
0
AB= 0
0 250 50

[ ]

125
25

50
M= 0
50 12500

Det = -2500
0

The system is controllable

For stability,
s ( s+ 250 ) , s=0 , s=250

The system is marginally stable, as one of the pole on unit circle

Q2 e)

Q2 f)