Introduction to Control

Systems Technology

1
 The Control Problem
Fundamental Control Concepts Maintain a variable of process (h) at a
desired value (H) while rejecting the
effects of outside disturbances (Qin) by
manipulating another system variable
(Qout).

Examples:
• Heating and Cooling homes and offices
• Automobile cruise control
• Hold the position of a mechanical linkage
• Maintain level in a tank

h Qout depends on h
If Qout = Qin, h constant
Qout > Qin, tank empties
Qout < Qin, tank overflows

2
Basic Subsystems of Control
Feedback Control Subsystems
• Measurement
• Control decision
• System modification

Measuremen
t- sight glass

Control Decision
Human adjusts Qout
to maintain h =H

Reference
(setpoint)
h = control Process-
Final Control
variable Maintain tank
Element
level
Valve
3
Automatic Control Systems
Use sensors and analog or digital electronics to monitor and adjust system
Level
Measurement
Sensor Elements of Automatic Control
• Process – single or multiple variables
• Measurement – sensors
Controller • Error Detection – compare H to h
• Controller – generate corrections
• Final Control Element – modify process
Final control
element

Valve
Maintain Position
level
4
Block Diagrams
Automatic control systems use mathematical descriptions of
subsystems to reduce complex components to inputs and
outputs

Input signal Control System Output signal

Component

Energy
Source
(Optional)
Signals flow between components in system based on arrow direction

5
 Typical Component Block Diagrams

T (°C) Vin (V)

Temperature
Amplifier
Sensor
Vout (mV) Vout (V)

Valve Position
(% Open)
Control
Dc Motor
Valve
Armature V n (rpm) Liquid Flow
Va (V) (m3/s)

Error
L (ft) E (V)
Level Controller
Transmitter
Current Correction
(mA) V (V)

6
 Transfer Functions
Transfer function Transfer Function is the ratio of Laplace transform
of the output to the Laplace transform of the input. Consider all
initial conditions to zero.

7
 Transfer Functions
Examples initial conditions to zero. Eout(0)=0

 sEout  Eout  Ein

Eout 1
TF  
Ein  s  1

8
 Open-Loop Control
Open loop control modifies output based on predetermined
control values. There is no actual measurement of controlled
quantity.

Controller Tank level may

vary with outside
Final control disturbances and
element system changes

Valve Output Tank

setting flow Level
Control
Controller Tank
Valve

Disturbances 9
 Closed Loop Control
Closed loop control modifies output based on measured
values of the control variable. Measured value compared to
desired value and used to maintain desired value when
disturbances occur. Closed loop control uses feedback of
output to input.

Desired Valve Output

Value setting flow Tank
Level
Control
+ Controller Tank
- Valve

Level
Measurement

Measured
level 10
 Closed Loop Control
Example: Auto Cruise Control Disturbances
Fuel Mechanical
flow Power
Actual
Set Speed Speed
Fuel
+ Controller Engine Auto
- Injectors

Error
Speed
Sensor

Disturbances: Up hill/ down hill

Head wind/ tail wind

11
Generalized Closed Loop Control
Block Diagram of Servo Control-Example Positioning Systems

R E=R-Cm Controller C
+
- G=C/E

Cm
Measurement
H=Cm/C

Reference = R Measured Variable = Cm

Error = E
Controlled Variable = C
12
Generalized Closed Loop Control
Find overall transfer function using signal flow algebra

(Input)(Gain)=(Output)
C
For servo control Find
R
Cm
1 E  R Cm 3 H  C  H  Cm
C

2 C
G  EG  C 1 3 E  R C
CmH E
E

(R  C  H)  G  C
13
Overall Transfer Function-Servo
Control
C
Find
R
1 2 3
(R  C  H)  G  C
R G -CHG  C Multiply through by G

R G  C  CHG Add CHG to both sides

R  G  C  (1  G  H ) Factor C out of right hand side

R G
C Divide both sides by (1+GH)
(1  G  H )
G C
 Divide both sides by R
(1  G  H ) R

14
 Block Diagram Simplifications
R + C R C
G C G

- R 1 G  H

H
Block diagram at left simplifies to the above.
Can use this to reduce multiple loops into one
Block.
Also remember that blocks in series multiply
G  G1  G 2  G 3

G1 G2 G3 G1G2G3

Equivalent Block

15
 Block Diagram Simplification
Example
R + C
+
G1 G2 G3
- - C1
R1
H1

H3 H2

C1 G2 H
 Reduce the inner loop
R 1 1  G 2  H1
H  H 2  H3 Combine outer feedback block

Combine reduced inner loop with remaining forward gain blocks

16
 Block Diagram Simplification Example (1)
R + G2
C
G1 G3
- 1  G 2  H1

H=H2H3

 G2   G1  G 2  G 3 
G  G1   
 3 
G   Compute the value of G
1  G 2  H1  1  G 2  H 1 

 G1  G 2  G 3 
 1 G  H 
C

G
  2 1 

R 1 G  H  G1  G 2  G 3  Substitute the values of G and H into formula

1    H 2  H3 and simplify
 1  G 2  H1 
17
 Block Diagram Simplification Example (2)
 G1  G 2  G 3 
 1 G  H 
C

G
  2 1 

R 1 G  H G G G 
1   1 2 3   H 2  H3
 1  G 2  H1 

Multiply top and bottom of ratio by (1+G2H1) and simplify

 G1  G 2  G 3 
 1  G  H   1  G 2  H1 
C  1  G1  G 2  G 3
 2

R   G1  G 2  G 3    G1  G 2  G 3  H 2  H 3 
1  
 
 2 3
H  H   1  G  H  1  G  H   1  G  H    
   
2 1 2 1 2 1
  1 G 2 H 1    1 G 2 H 1 

C G1  G 2  G 3
 Answer
R 1  G 2  H1  G 1  G 2  G 3  H 2  H 3

Reduced R G1  G 2  G 3 C
block 1  G 2  H1  G 1  G 2  G 3  H 2  H 3

18
Models of electrical elements

19
 Impedance
• Generalized resistance to a sinusoidal alternating current (AC) I(s)

Z (s) : V ( s)  Z ( s) I ( s)

20
Kirchhoff’s Voltage Law (KVL)
The algebraic sum of voltage drops around any loop is =0.

21
Kirchhoff’s Current Law (KCL)
The algebraic sum of currents into any junction is zero.

22
Impedance computation

23
Impedance computation

24
Modeling example

25
Modeling example (cont’d)

26
Example: Modeling of op amp

27
Example: Modeling of op amp

28
Example: Modeling of op amp

29
Modeling example: op amp

By the formula in previous page,

30
Modeling example: op amp

31
Modeling example: op amp

1 1 1  RCs
  sC 
Zp R R
R
Zp 
1  RCs
32
Modeling example: op amp

33
Modeling – Mechanical Elements

34
Modeling – Spring-Mass-Damper Systems

35
Modeling – Free Body Diagram

36
Modeling – Spring-Mass-Damper System

37
Example

Problem: Find the transfer function X(S)/F(S)

38
Example
Solution:

39
 Example Obtain the transfer functions and of the mechanical system
shown in the following figure

 F  ma
 k 2 ( x1  x 2 ) k 2 ( x1  x 2 )
u m1x1  k3 x 2
m x
m1 = m1 m2 m
= 2 2 2

 k1x1 b(x1  x 2 ) b ( x1  x 2 )

u  k1 x1  k2 ( x1  x2 )  b( x1  x2 )  m1 x1 k2 ( x1  x2 )  b( x1  x2 )  k3 x2  m2 x2
m1 x1  bx1  (k1  k2 ) x1  bx2  k2 x2  u m2 x2  bx2  (k2  k3 ) x2  bx1  k2 x1
40
 Assuming zero initial conditions
m1 x1  bx1  (k1  k2 ) x1  bx2  k2 x2  u

m2 x2  bx2  (k2  k3 ) x2  bx1  k2 x1
[m1s 2  bs  (k1  k2 )] X 1 ( s )  [bs  k2 ] X 2 ( s )  U ( s ) (1)

[m2 s  bs  (k2  k3 )] X 2 ( s )  [bs  k2 ] X 1 ( s )
2
(2)
Solving Equation (2) for X2(s) and substituting it into Equation (1), we get

[bs  k2 ]
X 2 (s)  X 1 (s) (3)
[m2 s  bs  (k2  k3 )]
2

[bs  k2 ]
[m1s  bs  (k1  k2 )] X 1 ( s )  [bs  k2 ]
2
X 1 (s)  U (s)
[m2 s  bs  (k2  k3 )]
2

41
and simplifying, we get
[bs  k2 ]
X 2 (s)  X 1 (s) (3)
[m2 s  bs  (k2  k3 )]
2

[bs  k2 ]
[m1s 2  bs  (k1  k2 )] X 1 ( s )  [bs  k2 ] X 1 (s)  U (s)
[m2 s  bs  (k2  k3 )]
2

{[m1s 2  bs  (k1  k2 )][m2 s 2  bs  (k2  k3 )]  [bs  k 2 ]2 } X 1 ( s )  [m2 s 2  bs  (k 2  k3 )]U ( s )

X 1 (s) m2 s 2  bs  k2  k3
 (4)
U ( s ) (m1s 2  bs  k1  k2 )(m2 s 2  bs  k2  k3 )  (bs  k2 ) 2
From Equations (3) and (4) we have
[bs  k2 ] m2 s 2  bs  k2  k3
X 2 (s)   U (s)
[m2 s  bs  (k2  k3 )] (m1s  bs  k1  k2 )(m2 s  bs  k 2  k3 )  (bs  k 2 )
2 2 2 2

X 2 (s) bs  k2

U ( s ) (m1s 2  bs  k1  k2 )(m2 s 2  bs  k 2  k3 )  (bs  k 2 ) 2
42
Inverted pendulum system
An inverted pendulum mounted on a motor-driven cart is shown in the following figure:

Inverted pendulum system Free-body diagram.

Define the angle of the rod from the vertical line as . Define also the (x, y)
coordinates of the center of gravity of the pendulum rod as XG, YG. Then 43
NEWTON’S SECOND LAW FOR ROTATION

τ
The rotational motion of the pendulum
i  Iα rod about its center of gravity
I  Vlsin  Hlcos (1)
where I is the moment of inertia of
xG  x  lsin ;
the rod about its center of gravity
yG  lcos
The horizontal motion of center of
gravity of pendulum rod is given by
d2
mxG  H ; m 2 ( x  lsin )  H ; (2)
dt
The vertical motion of center of gravity of pendulum rod is
d2
myG  V  mg ; m 2 (lcos )  V  mg (3)
dt
The horizontal motion of cart is described by
Mx  u  H (4) 44
Taylor formula f ' (a) f '' (a ) f '''
(a)
f ( x)  f (a)  ( x  a)  ( x  a) 2  ( x  a )3  ...
1! 2! 3!
The Maclaurin expansion (the Taylor expansion about a=0)

cos (0) sin(0) cos (0)

sin( )  sin(0)  (  0)  (  0) 2  (  0)3  ...
1! 2! 3!
1 3 1 5 1 7
sin( )          ...
3 5 7
When θ is at about 0.2441 radians (13.99°) (sin(0.2441) = 0.2417
When θ is at about 0.1441 radians (8.2563°) (sin(0.1441) = 0.1436

One can thus safely approximate

sin   ; cos  1;   0 2

45
Since we must keep the inverted pendulum vertical ( and ’ are small quantities), and
Then, Equations (1) through (3) can be linearized.

 
 I  Vlsin  Hlcos (1)  I  Vl  Hl
   I  Vl  Hl
 d2  d 
 m 2 ( x  lsin )  H (2)   m( x  (lcos ))  H   m( x  l ( sin 2   cos )  H 
 dt  dt  ml (cos 2   sin  )  V  mg
 d2  d 
m 2 (lcos )  V  mg (3)  ml (sin  )  V  mg
 dt  dt
 I  Vl  Hl  I  Vl  Hl (5)
 
 m ( x  l (  2
  )  H   m( x  l )  H (6)
 ml ( 2   )  V  mg  0  V  mg (7)
  From Equations (4) and (6), we obtain

 Mx  u  H (4)
 m( x  l )  u  Mx ( M  m) x  ml  u (8)
m( x  l )  H (4)
46
From Equations (5), (6), and (7), we have  I  Vl  Hl (5)

 m( x  l )  H (6) I  Vl  l (mx  ml )
 0  V  mg (7)

Equations (8) and (8) describe the motion of the inverted-pendulum-on-the-cart system.
( M  m) x  ml  u (8)

( I  ml )  mlx  mgl
2
(9)
Since in this system the mass is concentrated at the top of the
rod, the center of gravity is the center of the pendulum ball.
For this case, the moment of inertia of the pendulum about its
center of gravity is small, and we assume I=0 in Equation (9).
Then the mathematical model for this system becomes as
follows:

( M  m) x  ml  u (10) ml   mlx  mgl

2
(11)
47
 By eliminating x’’ from Equations (10) and (11)
( M  m) x  ml  u (10) ( M  m) x  ml  u ( M  m) x  ml  u
 2  
 ml   mlx  mgl (11)  l  x  g  x  g  l
( M  m)( g  l )  ml  u  ( M  m) g  ( M  m)l  ml  u  ( M  m) g  Ml  u

Ml  ( M  m) g  u (12) By eliminating ’’ from Equations (10) and (11)

( M  m) x  ml  u (10) ( M  m) x  ml  u

 2   Mx  u  mg (13)
 ml   mlx  mgl (11)  ml  mx  mg
From Equation (12) we obtain the plant transfer function to be
Mls 2 ( s )  ( M  m) g ( s )  U ( s )  [ Mls 2  ( M  m) g ]( s )  U ( s )
( s ) 1

U (s) Mls 2  ( M  m) g
48
Modeling of Electromechanical Systems

D.C Drives

• Variable Voltage can be applied to the armature terminals of the DC

motor .
• Another method is to vary the flux per pole of the motor.
• The first method involve adjusting the motor’s armature while the
latter method involves adjusting the motor field. These methods are
49
referred to as “armature control” and “field control.”
 Example: Armature Controlled D.C Motor
Ra La
Input: voltage u B
ia
Output: Angular velocity  u eb T J

Electrical Subsystem (loop method):

dia
u  Ra ia  La  eb , where eb  back-emf voltage
dt
Mechanical Subsystem

Tmotor  Jω  Bω

50
 Example: Armature Controlled D.C Motor
Ra La
Power Transformation: B
ia
u eb T J
Torque-Current: Tmotor  K t ia 
Voltage-Speed: eb  K b ω

• Combing previous equations results in the following mathematical

model:

 dia
 La  Ra ia  K b ω  u
 dt
 Jω
   B-K t ia  0
51
 Example: Armature Controlled D.C Motor

Taking Laplace transform of the system’s differential equations with

zero initial conditions gives:

La s  Ra I a(s)  K b Ω(s)  U(s)


Js  B Ω(s)-Kt I a(s)  0

Eliminating Ia yields the input-output transfer function

Ω(s) Kt

U(s) La Js 2  JRa  BLa s  BRa  K t K b

52
 Example: Armature Controlled D.C Motor
Reduced Order Model
Assuming small inductance, La 0

Ω(s) Kt

U(s) JRa s  ( BRa  K t K b )

53
 Example: Armature Controlled D.C Motor

If output of the D.C motor is angular position θ then we know

Ra La
d B
 or ( s )  s ( s ) ia
dt u eb T J
θ

Which yields following transfer function

(s) Kt

U(s) s[ JRa s  ( BRa  K t K b )]

54
 Example: Field Controlled D.C Motor
Rf Ra La

if
ef Lf Tm J ea
B ω

Applying KVL at field circuit

di f
ef  if Rf  Lf
dt
Mechanical Subsystem
Tm  Jω  Bω

55
 Example: Field Controlled D.C Motor

Power Transformation:

Torque-Current: Tm  K f i f

where Kf: torque constant

Combing previous equations and taking Laplace transform (considering

initial conditions to zero) results in the following mathematical model:


E f ( s )  R f I f ( s )  sL f I f ( s )

 Js( s )  B( s )  K f I f ( s )

56
 Example: Field Controlled D.C Motor
Eliminating If(S) yields

Ω(s) Kf

E f (s) Js  B ( L f s  R f )

If angular position θ is output of the motor

Rf Ra La

if
ef Lf Tm J ea
B θ

(s) Kf

E f (s) sJs  B ( L f s  R f )
57
Example: Angular Position Control System

Ra La N1
+
JM
+ + + BM
ia BL
kp
e ea e b T θ JL
r c
_ _ _
N2
-
if = Constant

58
 Numerical Values for System constants
r = angular displacement of the reference input shaft
c = angular displacement of the output shaft
θ = angular displacement of the motor shaft
K1 = gain of the potentiometer shaft = 24/π
Kp = amplifier gain = 10
ea = armature voltage
eb = back emf
Ra = armature winding resistance = 0.2 Ω
La = armature winding inductance = negligible
ia = armature winding current
Kb = back emf constant = 5.5x10-2 volt-sec/rad
Kt = motor torque constant = 6x10-5 N-m/ampere
Jm = moment of inertia of the motor = 1x10-5 kg-m2
Bm=viscous-friction coefficients of the motor = negligible
JL = moment of inertia of the load = 4.4x10-3 kgm2
BL = viscous friction coefficient of the load = 4x10-2 N-m/rad/sec
n= gear ratio = N1/N2 = 1/10 59
 Example: Angular Position Control System
• Transfer function of the armature controlled D.C motor with load connected to it
is given by

𝜃(𝑠) 𝐾𝑡
=
𝐸𝑎 (𝑠) 𝑠 𝐽𝑒𝑞 𝑅𝑎 + 𝐵𝑒𝑞 𝐿𝑎 𝑠 + 𝐵𝑒𝑞 𝑅𝑎 + 𝐾𝑡 𝐾𝑏

• Where 2 2
𝑁1 1
𝐽𝑒𝑞 = 𝐽𝑚 + 𝐽𝐿 = 1 × 10−5 + × 4.4 × 10−3 = 5.4 × 10−5
𝑁2 10
2 2
𝑁1 1
𝐵𝑒𝑞 = 𝐵𝑚 + 𝐵𝐿 = × 4 × 10−2 = 4 × 10−4
𝑁2 10
Kinetic Energy
1 1 1 N 1
T J m m2  J L L2  [ J m  J L ( 1 ) 2 ] m2  J eq m2
2 2 2 N2 2

𝜃(𝑠) 6
=
𝐸𝑎 (𝑠) 𝑠(1.08𝑠 + 8.33)
60
 Example: Angular Position Control System
• Error is difference between reference input 𝒓 𝒕 and out calculated 𝒄 𝒕 and can be
calculated as
𝑒 𝑡 = 𝐾1 𝑟 𝑡 − 𝑐(𝑡)
24
E 𝑠 = 𝑅 𝑆 − 𝐶(𝑆) = 7.64 𝑅 𝑆 − 𝐶(𝑆) (a)
𝜋

• Output of amplifier is
𝐸𝑎 𝑠 = 𝐾𝑝 𝐸 𝑠 = 10𝐸(𝑠) (b)
• Merging eq (a) and eq (b) yields
𝐸𝑎 𝑠 = 76.4 𝑅 𝑆 − 𝐶(𝑆)
• Relation between angular position of motor 𝜽 and angular position of load 𝒄 is given as
1
𝐶 𝑠 = 𝜃(𝑠)
• or 10
10𝐶 𝑠 = 𝜃(𝑠)

61
 Example: Angular Position Control System
𝜃(𝑠) 6
=
𝐸𝑎 (𝑠) 𝑠(1.08𝑠 + 8.33)

• Final Closed loop transfer function of the system can now be written as

10𝐶(𝑠) 6
=
76.4 𝑅 𝑆 − 𝐶(𝑆) 𝑠(1.08𝑆 + 8.33)

𝐶(𝑠) 42.3
= 2
𝑅 𝑆 𝑠 + 7.69𝑠 + 42.3

62