CBE 346 Spring 2013

Princeton University

Brief Overview of Process Control

Guest Lecturer: Dr. Tomas Co1

Department of Chemical Engineering, Michigan Technological University

Email: tbco@mtu.edu

2013 by Tomas Co

Page 1

Brief Overview of Process Control

1.
2.
3.
4.
5.
6.

2013 by Tomas Co

Elements of Process Control

Feedback Control
Dynamic Modeling
PID Controller Tuning
Analysis
Other Control Issues

Page 2

Process Control :

a statistics and engineering discipline that deals with

architectures, mechanisms and algorithms for maintaining the
output of a specific process within a desired range.
- definition from wikipedia.org

2013 by Tomas Co

Page 3

Elements of Process Control

1. Control objectives
Setpoints (targets), constraints, specifications

2. Input variables
manipulated variable vs. disturbance variable

3. Output variables
controlled variable vs. uncontrolled variable
measured variable vs. unmeasured variable

4. Control strategy
Structure : feedback, feedforward
Control algorithms : On/Off, PID

2013 by Tomas Co

Page 4

Signal Flow Diagram

disturbance
inputs

setpoints
& parameters

Controller

manipulated
variables

Process

process
outputs

2013 by Tomas Co

Page 5

Remarks:
1. Some control problems can be improved/simplified with
design retrofits.
2. Input or Output refers to information flow - not
material flow.
- Two types of diagrams used in control design:
a. Signal block diagrams
b. Piping and instrumentation diagram (P&ID)

2013 by Tomas Co

Page 6

Example 1: Level Control

FT

Fin
LT

h
FI

TT

Fout

Objective: Control the liquid level of a surge tank,

where outlet is under gravity flow.
(Sensors: FT is flow transmitter, FI is flow indicator, LT is level
transmitter, TT is temperature transmitter.)
(see http://www.chem.mtu.edu/~tbco/cm416/pidiag.html)
2013 by Tomas Co

Page 7

Questions/Discussion:
1. Identify and classify the different variables.

2. Propose strategies to control the liquid level.

3. How tightly should the level be controlled?

4. Which control valves should be manipulated?

2013 by Tomas Co

Page 8

Example 2: Heat Exchanger

Steam
PT
TT

TT

FT

TT

Feed

2013 by Tomas Co

Page 9

Questions/Discussion:
1. What is the control objective?

2. Identify and classify the different variables.

3. Propose control strategies.

4. If the product cannot exceed a maximum temperature, how

does this affect the control strategy?

2013 by Tomas Co

Page 10

Feedforward Control
Use input variables (e.g. disturbance measurements) to
determine value of manipulated variable.

Feedback Control
Use output variables (e.g. controlled variable) to determine
value of manipulated variable.

2013 by Tomas Co

Page 11

Case 1:
h setpoint
Controller

Fin
LT

h
FT

Fout

Question: Feedback or feedforward?

2013 by Tomas Co

Page 12

Case 2:
h setpoint
Controller

Fin
LT

h
FT

Fout

Questions: Feedback or feedforward?

2013 by Tomas Co

Page 13

Simple Feedback Control Structure:

(signal block diagram)

disturbance

setpoint

manipulated
input

error

Controller

Process

measurement

2013 by Tomas Co

Sensor

controlled
output

controlled output

Page 14

General Roles of Feedback Control:

- Setpoint (target) tracking
- Disturbance rejection

Relay (On/Off) Control: (ex.: home furnace, refrigerators)

=

if  > 

if  


- Easiest (often cheapest) to implement

- Results in limit cycle response (often complemented with
hyteresis to reduce erratic behavior due to measurement
noise).
- Often:  =
 + ; =
 ( ,
 can be <= 0 or 0)
2013 by Tomas Co

Page 15

Exercise 1a: Manual Control and Relay Control

(http://www.chem.mtu.edu/~tbco/cm416/newpida.html)

1. Move setpoint to 0.6 then vary input (u) to get output (y) to
approach setpoint value.
2. Switch to relay control and try with
 = 0 and = 0.1.
(This mean  = 0.1 and = 0.1)

3. Try to improve the process with different values of
 and .

4. Move the setpoint to -1.

2013 by Tomas Co

Page 16

OBSERVATIONS:
1.
2.
3.
4.

2013 by Tomas Co

Page 17

Proportional Control Law:

=  (
 )
Where,  is known as the Proportional Control Gain.

(Note: often the algorithm includes a
 term, called the bias.

For simplicity, we will assume
 = 0.)

2013 by Tomas Co

Page 18

Exercise 1b: Proportional Control

(http://www.chem.mtu.edu/~tbco/cm416/newpida.html)

1. Change setpoint to -0.2 then switch to Proportional control

(PID mode with I and D mode switched off).
2. Try different values of proportional gain.

2013 by Tomas Co

Page 19

OBSERVATIONS:
1.
2.
3.
4.

2013 by Tomas Co

Page 20

Proportional-Integral Control ( to remove offsets )

=   (
 ) +

1
(
 ) ! "


where,  = integral time constant, aka reset time.

- Simplified interpretation of  : projected average time

for removing offset.
- Larger value of  reduces effects of integral mode.
- Smaller value of  likely to introduce overshoot.

2013 by Tomas Co

Page 21

Exercise 1c: PI Control

(http://www.chem.mtu.edu/~tbco/cm416/newpida.html)

1. Move setpoint to 0.6 then set  = 0.2 and

 = 30.
overshoot
y

+
- 5%
setpoint

response time
time

2. Try other values to reduce overshoot.

3. Try other values to improve response time.
2013 by Tomas Co

Page 22

OBSERVATIONS:
1.
2.
3.
4.

2013 by Tomas Co

Page 23

Proportional-Integral-Derivative Control ( to reduce

oscillation/overshoot effects of integral mode )
1
(
 ) !

(
 )
'
!

=  % (
 ) +
+ &

where, & =derivative time (aka rate coefficient)

- Can improve (decrease) response time.
- Large value of & amplifies noise effects.

2013 by Tomas Co

Page 24

Exercise 1d: PID Control

(http://www.chem.mtu.edu/~tbco/cm416/newpida.html)

1. Set  = 0.3,  = 25 and & = 8.

2. Try & = 50.

2013 by Tomas Co

Page 25

OBSERVATIONS:
1.
2.
3.
4.

2013 by Tomas Co

Page 26

Remarks:
1. P control is the simplest often used for systems where
offset is not a problem.
Example: Level control of surge tanks
2. PI control is used where offset is undesirable, yet responses
to manipulated variables are fast.
Example: Flow control
3. PID control is used where offset is undesirable but responses
are slow.
Example: Temperature control
4. Controller Tuning Problem: determining appropriate values
of  ,  and & .
2013 by Tomas Co

Page 27

RECAP #1
1. Four main elements of control :
i.
ii.
iii.
iv.

Control objective
Input variables
Output variables
Control strategy

2. Two main roles of control:

i.
ii.

2013 by Tomas Co

Setpoint tracking
Disturbance rejection

Page 28

3. Three modes of PID Control:

=  % (
 ) +

(
 )
1
(
 ) ! + &
'

!

i. Proportional Control Gain :


ii. Integral-Time (Reset) :

iii. Derivative-Time (Rate coefficient):

&

2013 by Tomas Co

Page 29

Brief Overview of Process Control

1.
2.
3.
4.
5.

Elements of Process Control

Feedback Control
Dynamic Modeling
PID Controller Tuning
Analysis

6. Other Control Issues

2013 by Tomas Co

Page 30

Dynamic Process Models

- Models used to
i) describe and simulate transient process behavior
ii) predict responses to different conditions
iii) explore effects of redesign/retrofits and/or control
strategies
- Mathematical models: standard formulation involves
differential equations based on time derivatives.

Example: heat exchanger

*
= + ,*, *./ , 0, *
12 , 3
12 , 0
12 ; 5, 6, 78 , 9 :
!

2013 by Tomas Co

Page 31

Praters Principle of Optimum Sloppiness

- There is an optimum level of model detail to yield maximum
engineering utility based on the proposed objectives of the
model (balanced among accuracy, cost, flexibility, etc.)
model
utility

model detail

( but the optimum may change depending on availability and

cost of new technologies.)
2013 by Tomas Co

Page 32

-Types of models:
a) Phenomenological (based on first principles)
b) Empirical
c) Mixed

Typical empirical models used in process control design:

a) First order and first order plus time delay (FOPTD)
b) Second order underdamped models
c) Inverse-response models

2013 by Tomas Co

Page 33

First Order Process:

; 1
= (
;)
!  8

Example: Temperature in Continuous-Stirred Tank

F
Tin
V,T

Fout
T

<9=7> ?* *@A BC = 907> ?*./ *@A B 90DE 7> ?* *@A B

Assume 0 = 0DE and =, 9, 7> constant:

2013 by Tomas Co

* 0
= (* *)
! = ./

Page 34

Solution: (use variation-of-parameters)

; (!) = ; G

H/J

 
(K)
8

+

LMHN
G J

Special Case:
=
new (constant), 8
 = ;

;(!) = ; G

!/

R
new


!/

new

 GK/ K

= ; G!/ + R
new (1 G
= ; + SR

0
!/ )

;0 T (1 G!/ )

= ;0 + 8 L
new
 N<1 G H/J C

;(!) ;
= 8 <1 G H/J C

new

2013 by Tomas Co

Page 35

REMARKS:

1. As ! , ; ;/V , a new steady state.

2. Then 8 = ;

/

, known as the Process gain,

where ;

= ;/V ; and

=
/V
 .
3. When ! = ,
;() ;
= (1 G HX ) = 0.632
;/V ;
Thus,  is known as the time constant of the process
( or the 0.632 point).
Question: How does half-life compare to timeconstant?

2013 by Tomas Co

Page 36

Step-response experiment:
1. Fix manipulated input
variable to
D and wait until
output settles to steady
state D .
2. Introduce a step change in
input to
/V . (Note the
time when the step was
implemented.)
3. Record the response of
output  unit it reaches a
new equilibrium /V .
Process Gain<R C =

aa

aa

Input

unew
Duss

uo
Time

t step
Output

ynew
Dyss

yo
Time

t step

2013 by Tomas Co

Page 37

First Order Plus Time-Delay (FOPTD) Model

&b
&

X
J

= L8
<HJcdefg C N

where,
8 = process gain
 = time constant
&h1b = time delay

Input

unew
Duss

uo
Time

t step
Output

ynew
Dyss
0.632

yo

t step

Dyss
Time

tdelay

2013 by Tomas Co

Page 38

Analytical solution of FOPTD model subject to step test:

D

(!) = 
D + ?1 G Hi() B>

j (!) =

if ! < !
8 + &h1b
if ! > !
8 + &h1b

! !
8 &h1b
;
=
/V
D


2013 by Tomas Co

Page 39

Estimation of k, lm and knopqr

Method 1:

8 =
(by setting ! 

Let !1 be the time such that j!1  = 1/3, then

!1  = D + 8

<1 G HX/w C
= D + 0.283

Let !x be the time such that j!x  = 1, then

!x  = D + 8

1 G HX 
= D + 0.632

2013 by Tomas Co

Page 40

From the experimental output, determine !

8 , !1 , !x .
1
!1 !
8 &h1b = 
3
!x !
8 &h1b = 
3
 = (!x !1 ) ; &h1b = !x !
8 
2
Method 2: use computers (e.g. MS Excel)

2013 by Tomas Co

Page 41

Exercise 2: Parameter Estimation of FOPTD

(http://www.chem.mtu.edu/~tbco/cm416/newpida.html)

1. Implement a step test.

2. Collect a range of data that contains initial steady state and
final steady state. Estimate the model parameters using
method 1.
3. Use MS Excel to estimate model parameters using the
analytical solution of FOPTD.

2013 by Tomas Co

Page 42

Cohen-Coon PID Tuning Rules:

Based on FOPTD, obtain 8 ,  and &h1b .

Let y =
P
PI
PID

Jcdefg
J



X
(1
z{ @

+ )

X
~
,
z{ @ w

+ :

X
,0.9
z{ @

@
w

@
:
X}

@
~


&h1b

30 + 3y
9 + 20y

32 + 6y
&h1b
13 + 8y

&

4
&h1b
11 + 2y

2013 by Tomas Co

Page 43

Exercise 3: Cohen-Coon Tuning

(http://www.chem.mtu.edu/~tbco/cm416/newpida.html)

1. Use FOPTD parameters to find PID parameters.

2. Implement PID.

2013 by Tomas Co

Page 44

Closed-loop Modeling for Ziegler-Nichols Tuning

Output

@ Kc<Ku

1. Implement P Control.
2. Obtain ultimate gain, E
(the critical value of 
where the process is about
to be unstable.)

Time

Output

Pu
@ Kc=Ku

3. At  = E , measure the

ultimate period 3E (the

time from one peak to the
next).

Time

Output

@ Kc>Ku

Time

2013 by Tomas Co

Page 45

Ziegler-Nichols PID Tuning Rules:

Using E and 3E ,
P
PI
PID

2013 by Tomas Co





E /2.2

3E /1.2

E /2

E /1.7

3E /2

&

3E /8

Page 46

Tyreus-Luyben PID Tuning Rules:

Using E and 3E ,
P
PI
PID





E /3.2

2.23E

E /2

E /2.2

2.23E

&

3E /6.3

2013 by Tomas Co

Page 47

Exercise 4: Ziegler-Nichols Tuning

(http://www.chem.mtu.edu/~tbco/cm416/newpida.html)

1. Find E and 3E .

2. Evaluate PID parameters based on Ziegler-Nichols rules.

3. Implement PID.

4. Repeat with Tyreus-Luyben.

2013 by Tomas Co

Page 48

Second-order Underdamped Processes:

& b
/}
&

+ 2/

&b
&

+  = 8

where, =damping coefficient

/ =natural teim constant

8 =process gain=

/

= exp

K=

XH

}J

XH

Input

unew
Du

uo
Time

t step

Output

a
ynew
Dy

yo
Time

t step

2013 by Tomas Co

Page 49

Inverse Response Processes:

Input

& b
}
&

&b
+ X +  
&
&E
8 (jX +
)
&

-needs numerical methods to

estimate parameters

unew
Du

uo
Time

t step
Output

ynew
Dy

yo

t step

2013 by Tomas Co

Time

Page 50

General 2nd Order Linear Model:

}

!}

+ X




+   = X
+ 

!
!

Is equivalent to

X
= X X + } + X

!
}
=  X + 

!
 = X

2013 by Tomas Co

Page 51

General nth Order Linear Model:

/

!/

+ + X

/HX



+   = /HX /HX + + 

!
!

Is equivalent to

X
= /HX X + } + /HX

!

/HX
= X X + / + X

!
/
=  X + 

!
 = X

2013 by Tomas Co

Page 52

Computer Simulation to Estimate Parameters

Euler Method:

X 
= +(,
, !)
= +( ,
, ! )
!
!
X =  + ! +( ,
, ! )

So for  order process,

(X )X = (X ) + !L/HX (X ) + (} ) + /HX (
) N

(/ )X = (/ ) + !L (X ) +  (
) N
X = (X )X

2013 by Tomas Co

Page 53

Q: What about initial conditions?

A: For convenience, it would be helpful if the initial conditions
were all zero.
This can be accomplished if:
a) The process is initially at equilibrium
all time derivatives
are zero
b) The variables are replaced by deviation variables
& =  D and
& =

D
Note: Using these tricks will also help later when building
transfer functions.
2013 by Tomas Co

Page 54

For PID, let G = 

  ,

X

!
&
=  G + G. + LG GHX N

!
.

HX

!
&

=  GHX + G. + LGHX GH} N

!
.

After subtraction, we get the discrete PID form:

X =

+  G GHX +

!
&
G + LG 2GHX + GH} N

!

2013 by Tomas Co

Page 55

Exercise 5: Optimal Tuning from Simulation

(http://www.chem.mtu.edu/~tbco/cm416/newpidb.html)

1. Obtain step test data.

2. Use MS Excel to approximate the model.
3. Use the model to find optimal tuning.
4. Implement the PID parameters.

2013 by Tomas Co

Page 56

RECAP # 2
1. Models (at various levels of details) are used to help
characterize the dynamics of a process.
2. If FOPTD applies, then Cohen-Coon tuning rules apply.
Alternatively, the Ziegler-Nichols tuning is also often used
for PID tuning.
3. Computer simulation can also be used to estimate the
model and this can be used for optimal tuning of PID
controllers.

2013 by Tomas Co

Page 57

Brief Overview of Process Control

1.
2.
3.
4.
5.
6.

2013 by Tomas Co

Elements of Process Control

Feedback Control
Dynamic Modeling
PID Controller Tuning
Analysis
Other Control Issues

Page 58

Typical Dynamic Elements

1. Exponential Decay or Growth: (!) = 6G x

y

b more positive

b more negative

2. Sinusoidal Response: (!) = G x L6 sin(!) + cos(!)N

y

b negative

ebt
|B|

b positive

ebt

2p
w

2013 by Tomas Co

Page 59

Q: Which functions will match the graphs below?

a) 1 4G H/}

e) G .w cos(3!)
2013 by Tomas Co

b) 4 8G H/} +
4G H}
f) 2G

c) G H.w L3 sin(2!)
2 cos(2!)N
g) G H.} Lsin(20!) +
4 cos(20!)N

d) 2G H.X

h) 0.01G . +
3G H.X
Page 60

Solution of ODE using Laplace Transforms

Definition: Given +(!), then Laplace transform is given by

L+(!)N =  +(!)G H
 ! = +(a)

Example: +(!) = G H1 , where  is a constant.

LG H1 N

=

(G H1 )G H


; G(a) > 0.

! =  G H(
1) !


1
1
G H(
1)
=
a+
a+


Special case:  = 0, L1N = 1/a.

2013 by Tomas Co

Page 61

Simple Laplace Transform table:

+(!)

+(a) = L+N

G H1 cos( !)

(a + )
(a +  + )(a +  )

G H1

G H1 sin( !)

(where = 1)
2013 by Tomas Co

!/

1
a+

(a +  + )(a +  )
!

a /X

Page 62

Laplace transform of derivatives:

+
+
 " =  G H
 !
!
!


Integration by parts:

= G H

=

&A
&

;
= aG H


! ; = +

+
H

 " = +G
+ a  +G H
 !
!



= +(0) + aL+N

Generalizing:


/+

!/

"=

a / L+N

/HX + (0)

/HX

2013 by Tomas Co

Page 63

let  and be constant, then

Linearity Property:

L+(!) + (!)N = L+N + LN

Inverse Laplace Transform:

HX ?+(a)B = HX L L+(!)N N = +(!)

- Often use table of Laplace transforms, if item is available

- If necessary, can use the Bromwich formula (quite rarely)
M
1
% lim  +(a)G
 a'
?+(a)B =
2 M HM
HX

Example:

HX 
2013 by Tomas Co

1
" = G Hw
a+3

Page 64

Example: Obtain the solution of ODE using Laplace transforms

}

!}

+5



+ 6 = 3 ; (0) = 0;
=0
!
! 

Apply Laplace transforms of both sides,

LaN =

a } LN

3
+ 5aLN + 6LN =
a

3
3
6

=
=
+
+
a(a } + 5a + 6) a(a + 3)(a + 2) a a + 3 a + 2
6 = 1/2,

= 1,

= 3/2

1
1
1
" + HX 
"
(!) = 6HX  " + HX 
a
a+3
a+2
= 6 + G Hw + G H}
2013 by Tomas Co

Page 65

Example: Obtain the solution of ODE using Laplace transforms

}

!}

+4



+ 5 = 3 ; (0) = 0;
=0
!
! 

Apply Laplace transforms of both sides,

3
a
3
3
LN =
=
a(a } + 2a + 5) a(a + 1 + 2 )(a + 1 2 )

(a + 1)
6
= +
+
a (a + 1 + 2)(a + 1 2) (a + 1 + 2)(a + 1 2)
a } LN + 2aLN + 5LN =

(!) = 6 + G H sin(2!) + G H cos(2!)

where 6 = 3/5, = 3/10, = 3/5

2013 by Tomas Co

Page 66

For the general nth order linear model, assuming zero initial
conditions:
/

/

!/

+ + X

/HX


+ D  = /HX /HX + + D

!
!

Taking the Laplace transforms yields

or

(/ a / + + X a +  ) = ( /HX a /HX + +  )

2013 by Tomas Co

/HX a /HX + + 
 = %
'
= (a)

/ a / + + X a + 

Page 67

Remarks:

1. (a) is the transfer function from
to .

2. The roots of the denominator are known as the poles of the
transfer function, also known as the eigenvalues of the
process.
3. The eigenvalues determine the transient behavior of the
process:
a) If any of the eigenvalues have a positive real part, then the
process will be unstable.
b) The more negative the real part, the faster the dynamics
die out.
c) The imaginary parts of the eigenvalues determine the
frequency of oscillations.
2013 by Tomas Co

Page 68

Main Principle for Linear Control Design

Feedback controllers, compensators and control configurations
are designed to alter the system dynamics by adjusting the
values (i.e. position in complex plane) of the eigenvalues.

faster
response

Im(s)

lower
frequency

x
x

Desirable Region

x
x

Re(s)

unstable eigenvalues
(have to move to the
left side for stability)

2013 by Tomas Co

Page 69

 By considering each block in a signal flow diagram to have

a transfer function, the overall equivalent transfer function
from the setpoint to the output can be found by simple
algebraic manipulation.
 Likewise, the overall equivalent transfer function from
disturbance to the output can also be found by algebraic
manipulations.

2013 by Tomas Co

Page 70

Example: Simple feedback control

~
d
disturbance
dynamics

D(s)
~
yset

+
-

~
e

~
u

C(s)

G(s)
process

controller

M(s)

y~m

+
~
a

~
b

y~

y~

sensor+filters

 =  +
 =

=

= G
G = 
 2
2 = 

 = (
 ) +
(1 + ) = 
 +

 = 
" 
 + 
"
1 +
1 +

2013 by Tomas Co

Page 71

Challenge: Internal model control

~
d
disturbance
dynamics

D(s)
~
yset

+
-

~
e

C(s)
controller

~
u

G(s)
process

+
~
a

H(s)

~
b

~
h +

internal model
y~m

y~
y~
+
y~e

M(s)
sensor+filters

 = ( ? )
 + ( ? )
2013 by Tomas Co

Page 72

Transfer Functions of P, PI and PID Controllers

P
PI
PID



 a + 1
 a
 a + 1 & a + 1


; < 0.05
 a
& a + 1


2013 by Tomas Co

Page 73

Example: Simple Feedback Control (continuation)

Let (a) =  , (a) = 1, (a) =
Then, after substitution,

H}

and (a) =

w
.

H}

3
2
:
a 2  +
a2

 =


3
3
:
:
1 +  ,
1 +  ,
a 2
a 2
3
2

 +

=
a 2 3
a 2 3
 ,

For stabilization, we need:  <

2013 by Tomas Co

}
w

Page 74

Q: What about estimation of error offset ?

A: One can use the final value theorem of Laplace transform.
Final value theorem: Assuming +(!) is stable,
lim +(!) = lim a L+N

Proof: from Laplace transform of derivative

lim a L+N = lim 

 

= +()

A()
+ H

G
! + +(0) = 
+ + +(0)
!
A()

2013 by Tomas Co

Page 75

Example: (continuation from previous example)

 =

Assume 
 = 3 and

3
2

 +

a 2 3
a 2 3

= 2.

3
3
3
2
2
G (a) = 
  =


a
a 2 3 a
a 2 3 a

The offset is then given by

3
3
3
2
2
G() = lim a 

"


a
a 2 3 a
a 2 3 a
=

3(2 3 ) + 9 4

10
=
2 3
3 + 2
}

as  , the smaller the offset.

w

2013 by Tomas Co

Page 76

Q: Laplace transforms and transfer functions are only valid for

linear dynamics. What about nonlinear systems?

A: If process are expected to be operating in a small region

around a set of nominal values, then linearization can be used,
i.e. the eigenvalue analysis will be valid (around the small
region).

Note: one particular feature of nonlinear systems is the possibility

of multiple steady states

2013 by Tomas Co

Page 77

Linearization around an operating point ( , r , n ):

where,


= +(,
, )
!
( D ) + (

D ) + 7(
= + (D ,
D ,

E b ,E ,&

D ) ; 

; 7 =

b b ,E ,&

D)

& b ,E ,&

Common simplification: Use deviation variables, & =

( D ), , and assuming (D ,
D , D ) is at equilibrium,

2013 by Tomas Co

&
& +
& + 7
!

&

Page 78

Example:

*
= * } 80* + 10
} + 250 = +(*,
, )
!

At operating point 1: (*D ,
, ) = (30,10,2)

= 2*D 80 = 20
* (w,EX,&})

+
+

= 200 ;
= 250

(w,EX,&})
(w,EX,&})
*&
= 20*& + 200
& + 250
!

At operating point 2: (*D ,
, ) = (50,10,2)

2013 by Tomas Co

*&
= +20*& + 200
& + 250
!

& (stable)

& (unstable)
Page 79

RECAP #3
1. The eigenvalues are key tools for analysis of the dynamics
with or without controllers.
 If any of the eigenvalues has positive real parts, the
system will be unstable
 The more negative the real parts the faster the
response
 The larger the imaginary parts, the higher the
frequency of oscillation
2. Using Laplace transforms, we can characterize the effects of
inputs to the outputs via transfer functions.

2013 by Tomas Co

Page 80

3. Only algebraic manipulations are needed to obtain the

transfer functions from either setpoint or disturbance to the
process output.
4. Control design, configuration and tuning is focused on how
to move the eigenvalues to locations in the complex plane
that would achieve desired dynamic behavior.
5. If system is nonlinear, linear analysis can be used on
linearized approximate models.

2013 by Tomas Co

Page 81

Other Issues in Classical Process Controls

1. Signal filtering
- Need to smooth out noise without damping crucial dynamic
information

2. Anti-reset windup
- Integral mode accumulate error information even though
valves/control-elements have saturated, causing unnecessary
inertial effects on controller response.

3. Cascade control
- Direct feedback control be become sluggish due to
nonlinearities (e.g. valve stiction).
4. Split-range control
- Control elements are often directional, e.g. cooling and
heating elements have different dynamic effects.
2013 by Tomas Co

Page 82

5. Robustness and Auto-tuning

- Require new controller parameters when set-points or process
dynamics are significantly far from nominal design conditions

6. Multivariable and plant-wide control

- Various control configuration are possible: cascade, multiple
single-input/single-output (SISO) control loops, multiinput/multi-output (SISO) control loops, etc.

2013 by Tomas Co

Page 83

Other Control Strategies:

1.
2.
3.
4.

Cascade Control
Feedforward-Feedback Control
Internal Model Control (special case: Smith predictor)
Model Predictive Control

2013 by Tomas Co

Page 84

Model Predictive Control:

1. Use optimization to evaluate N-steps ahead:
minE ,,E Cost(, 
 ,
)

Subject to:  = +(,
, )

  21 ;


21

2. Implement only one step (or a few steps)

3. Repeat from step 1.
Past

Future
setpoint
apply only the first move

2013 by Tomas Co

Page 85

OVERALL RECAP
1. Introduction to control concepts
- elements and feedback control
2. PID control and tuning rules
- Control law: P, PI and PID
- Ziegler-Nichols and Cohen-Coon
- Optimal tuning approach
3. Process modeling
- FOPTD model
- General linear model
- Simulation and parameter estimation

2013 by Tomas Co

Page 86

4. Analysis
- Using eigenvalues to predict behavior
- Laplace transforms to generate transfer functions
- Analysis and design of feedback system using transfer
function manipulation
- Linearization
5. Other control issues and advanced control configurations.

2013 by Tomas Co

Page 87