CHG 802 ASSIGNMENT

BY

BAYODE AJAYI OLADUNNI 209041040

TO

DEPARTMENT OF CHEMICAL AND PETROLEUM ENGINEERING, FACULTY OF

ENGINEERING, UNIVERSITY OF LAGOS, AKOKA, YABA, LAGOS STATE,
NIGERIA

IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE AWARD OF

MASTERS OF SCIENCE (M.SC) DEGREE IN CHEMICAL ENGINEERING

SUBMITTED TO DR. ALMORUF O.F. WILLIAMS

JUNE, 2023
TABLE OF CONTENTS

TITLE PAGE i

EXECUTIVE SUMMARY ii

INTRODUCTION iii

PROBLEM 1 1-5

PROBLEM 2 6-12

PROBLEM 3 12-18

PROBLEM 4 19-22

PROBLEM 5 23-26

PROBLEM 6 27-30
EXECUTIVE SUMMARY

This report covers the overview of process control and how it is practiced in the chemical

process industry. It entails and lays the foundation for study of process dynamics, Modeling

and control.

The work analysed in this study involves the modeling of the mixing tank where both the tank

level and the outlet tank temperature are measured coupled with the modeling of non-linear

behavior involving two gravity tanks in series.

A system with a noticeable delay otherwise known as time-delay system is observed and the

corresponding behavior is analyzed using bode plots.

A closed loop system is also demonstrated with the use of block diagrams and transfer

function where block diagrams is used to analyze the dynamic behavior of simple closed loop

system and model based control system such as Internal model control. Ziegler Nichol

settings are adopted to determine the gain and time constant of different controller.

Scilab software is also used to make bode plot and also to simulate PI and PID Controller.

This software show the comparison between the two controller.

INTRODUCTION/RELEVANCE
Most process systems in reality do not behave in a totally orderly manner. However, even in

the form of an incomplete picture of reality; we may be able to learn several things. Models

are frequently used in science and technology. Applications of models in engineering can be

found in the following areas namely; Research and Development, Process Design, Planning

and Scheduling, Process Optimization, Prediction and Control, etc.

Modeling is the procedure to formulate the steady state or dynamic effects of the system that

will be considered into mathematical equations. The modeling procedure can be summarized

in the following steps: Expression of the problem in mathematical language (based on

assumptions about the process), model solution (which involves the application of the

appropriate mathematical tool for obtaining the general solution of the formulated model),

interpretation of the model solution and general model evaluation and finally, drawing

inferences from the model.

. In their early training, engineering students learn a great deal about steady-state physical

systems, which is natural, because steady state systems are somewhat easier to understand

and provide appropriate learning examples. However, the practicing engineer should have a

mastery of dynamic physical systems as well. This book provides the basic information and

engineering methods needed to analyze and design plants that function well in a dynamic

world. Control engineering is an engineering science that is used in many engineering

disciplines—for example, chemical, electrical, and mechanical engineering—and it is applied

to a wide range of physical systems from electrical circuits to guided missiles to robots. The

field of process control encompasses the basic principles most useful when applied to the

physicochemical systems often encountered by chemical engineers, such as chemical

reactors, heat exchangers, and mass transfer equipment.

PROBLEM 1
Background of Problem
Gravity Tanks Process is comprised of two tanks stacked one above the other. They are

essentially two drums or barrels with holes punched in their bottoms and a variable

position control valve manipulates the inlet flow rate feeding the upper tank. The liquid

drains freely out through the hole in the bottom of the upper tank to feed the lower tank.

From there, the liquid exits either by an outlet drain (another free-draining hole) or by

pumped flow stream.

Description of Problem

This problem is related to two gravity tanks in series. The question involves a operation of

two gravity tanks in series. The fluid enters tank 1 with a flow rate F0 (m3/s). The fluid height

in tank 1 is h1 (m). The stream going out of tank 1 (outlet stream) enters tank 2 at a flow rate

F1. The height of fluid in tank 2 is denoted by h2 (m). The fluid leaves tank 2 at a flow rate F2

(m3/sec). The inlet stream to tank 1 is equipped with a pump.

Assumptions

 The density is Constant in all streams

 The outlet flowrates, F1. and F2. are constant and assumed to be in a steady

state.

Methodology
1. The inputs and outputs of the system are to be identified. Inputs are
classified as disturbances or manipulated variables.

2. A dynamic model is to be developed that describes the relationship

between the inputs and outputs.
 3. For your specific choice of inputs and outputs, is the dynamical control
system linear or nonlinear?

4. Assume that the inlet flowrate is Fo = 0.05(m3/sec). Compute the

corresponding equilibrium for this system.

5. If necessary, linearize this system about the steady-state.

Develop a transfer function model relating the inlet flowrate, F0, and the height of

liquid in tank 2, h2

Figure 1: Two tank system

Analysis and Computation

1. Input is the inlet flowrate F0. It is a manipulated variable. Outputs are the levels h1 and h2.
2. The overall mass balance on Tank 1 gives:
dh1 F 0 β √ h1
= −
dt A A

and on tank 2:
dh2 β √ h1 β √ h2
= −
dt A A
3. This system is nonlinear.
4. At steady-state:

From tank 1:

F 0 β √ h1 s
− =0
A A
Or

( )
2
F0
h1 s = =1 m.
β

From tank 2:

β √ h1s β √ h2s
− =0
A A

.
And therefore, h2s = h1s = 1 m.

5. We linearize the system about the steady state as follows:

For Tank 1:

dh1 F 0 β √ h1 s 1 1 β
≈ − + ( F 0−F 0s )− ( h −h )
dt A A A 2 A √ h1 s 1 1 s

1 (0.05)
¿
3.5
( F 0−F 0s ) − ( h −h )
2(3.5) 1 1 s

and for tank 2:

dh2 β √ h1 s β √ h2 s 1 β 1 β
dt
≈
A
−
A
+ ( h −h )− ( h −h )
2 A √ h1 s 1 1 s 2 A √ h2 s 2 2 s

(0.05) (0.05)
¿
2(3.5)
( h1−h1 s )− ( h −h )
2(3.5) 2 2 s

In deviation form, one obtains:

dh ' 1
=0.2857 F ' 0−0.0071 43 h ' 1
dt
And
 dh ' 2
=0.007143 h ' 1−0.007143 h' 2
dt

.
6. Taking the Laplace transform of both linearized ODEs: For tank 1:

sH1(s) = 0.2857F0(s) – 0.007143H1(s)

or

0.2857
H 1 ( s )= F (s ).
s +0.007143 0

For tank 2:
sH2(s) = 0.007143H1(s) – 0.007143H2(s)

or
−3
0.0071 0.0071 0.2857 2.0285 x 10
H 2 ( s )= H 1 ( s )= × F0 ( s )= 2 −5
F 0 ( s) .
s +0.0071 s +0.0071 s+ 0.0071 s +0.014286 s+5.041 x 10
Results and Discussion

The transfer function model that relate the inlet flowrate, F0, and the height of
liquid in tank 2, h2 is

−3
( ) 0.0071 ( ) 0.0071 0.2857 ( ) 2.0285 x 10
H2 s = H1 s = × F0 s = 2 −5
F 0 ( s) .
s +0.0071 s +0.0071 s+ 0.0071 s +0.014286 s+5.041 x 10

PROBLEM 2
Background of Problem

Mixing tanks have been in existence for years and they have found there place in the

chemical industry . the problem gives an illustration of of two streams been mixed,

with the temperature and liquid height been of interest.

Description of Problem

This problem involves a system of two gravity tanks in series. The fluid enters tank 1

at a flow rate F0 (m3/s). The height of fluid in tank is h1 (m). The outlet stream from

tank 1 enters tank 2 at a flow rate F1. The height of fluid in tank 2 is denoted by h2

(m). The fluid leaves tank 2 at a flow rate F2 (m3/sec). The inlet stream to tank 1 is

equipped with a pump.

Assumptions

 Constant density in all streams

 The outlet flow rates, F1 and F2 are constant (not changing with time). Therefore

it is steady state process.

Methodology/Approach

For this process, we are asked to perform the following:

 Identify the inputs and outputs of the system. Classify inputs as

disturbances or manipulated variables.

 Develop a dynamic model that describes the relationship

between the inputs and outputs.

  For your specific choice of inputs and outputs, is the dynamical

control system linear or nonlinear

Analysis and Solution

CLASSIFICATION OF VARIABLES

VARIABLES

States L, T

Input F2 T 2

Disturbances F2 T 2

Output (Measurements) L, T

Assumptions

1. Perfect mixing
2. Constant densities
Mass Balance

d ( ρv )
=ρF 1+ ρF 2− ρF 3
dt

V = AL

dL
ρA =ρ(F 1 + F2−F 3 )
dt

dL
A =F1 + F 2−F3 … … … … … … … … … … … … … … …(1)
dt

Energy Balance

MCt

ρVCt

ρALCt

d
dt
[ LAρ C P T ]=F1 ρ C P ( T 1−T ref ) + F 2 ρ C P ( T 2−T ref ) −F3 ρ C P ( T −T ref )

T ref =0

d
Aρ C P [ ¿ ] =ρ C P [ F1 T 1+ F 2 T 2−F 3 T ]
dt

d
A [ ¿ ] =F 1 T 1+ F 2 T 2−F 3 T … … … … … … … … … … …(2)
dt
 [
A L
dT
dt
+T
dL
dt ]
=F 1 T 1+ F 2 T 2−F 3 T

dT dL
AL +TA =F1 T 1 + F 2 T 2−F 3 T … … … … … … … … … (3)
dt dt

Putting (1) in (2)

dT
AL +T [ F 1 + F2−F 3 ]=F 1 T 1 + F2 T 2 −F 3 T
dt

dT
AL =F 1 T 1 + F 2 T 2−F3 T −F1 T −F2 T + F 3 T … … … … … … … … …(3)
dt

dT (T 1−T ) (T 2−T )
=F 1 + F2
dt AL AL
PROBLEM 3
Background of Problem
A composition analyzer is used to measure the concentration of a pollutant in a

wastewater system.. Feedback control system is based on measuring fractionator

performance by automatic analysis of product streams. Combined predictive and

and feedback controls permit precise meeting of product specifications for both

products streams and maximization of fractionation profit.

Description of Problem

This problem presents us with a system built to analyze and control the concentration of

pollutants in a waste stream. This is a first order plus time delay system

Assumptions
 Is a steady state process
 .

Methodology/Approach
 Its understood that at steady state , the measured concentration of

thepollutant is equal to the actual concentration of the pollutants.

 The process transfer function is converted to time domain by finding

the laplace inverse of the transfer function

 The time is calculated.

 Open loop Bode plot is drawn for the system using scilab.
Analysis and Solution

C ' m(S) e−θS

=
C '( S) τs+1

Given θ=2 , τ=4 min , C=5 ppm

C ' m(S) e−2 S

G ( s )= =
C ' (S) 4 s+1
'
C ( t )=5+2 t

C ' ( t )=C−C

¿ 5+2 t−5
'
C ( t )=2 t

Taking Laplace

' 2
C ( s )= 2
s

At steady State C m=C=5 ppm

C m=measured concentration

' 2
Since C ( s )= 2
s

C ' m(S) e−2 S

=
2 4 s +1
2
s

[ ]
−2 S
' 2 e
C m ( S )=¿ 2
s 4 s+1

{ [ ]}
−2 S
−1 ' −1 2 e
L C m ( S ) =L 2
s 4 s+1

First the Laplace of L−1 ¿

2
A B C
s 2(¿4 s+1) = + 2+ ¿
S S 4 s +1

Let S =0
B=2

−1
Let S =
4

1
C= 2
16

C= 32

2 = 4A S2 + AS+4 BS+ B+C S 2 … … … … … … … … … … … … … … .(1)

From (1) 4A + C =0

−C −32
A= = = -8
4 4

2
−8 2 32
s 2(¿4 s+1) = + 2+ ¿
S S 4 s +1

{ [ ]}
−2 S 1
−1 2 e 4
( t −2)
L =−8+2 ( t−2 ) +8 e . U (t−2)
s 4 s +1
2

−1 −td
L f (s)e =f ( t−td ) .U (t−td )

t ≥ td

U ( t−2 )=1 t ≥ td

U ( t−2 )=0 t ⩼2
1
( t −2)
' 4
C m ( t )=−8+2 ( t−2 )+ 8 e
1
Cm−C m=−8+2 (t−2 )+ 8 e 4 (t −2)
1
25 ppm−5 ppm=−8+2 (t −2 )+ 8 e 4 (t −2)
1
( t −2)
20=−8+2 ( t−2 ) +8 e 4

t=15.8753 min

Results and Discussion

The warning light will turn on at 15.8753min since the warning light on the analyzer turn
whenever the measured concentration exceeds 25ppm.

The open loop Bode plot is in figure 4

Figure 4: open loop Bode plot
PROBLEM 4

Background of Problem

The IMC method is based on assumed process model and leads to analytical

expressions for the controller settings. Internal model control has the advantage that

it allows model uncertainty and trade-off between performance and robustness to be

considered in a more systematic fashion.

Description of Problem

This problem depict an internal model control whose closed loop transfer functions

for both the servo(tracking) and regulator problem is determined. The IMC method is
~
based on simplified block diagram below in figure 5. A process model GP and the
~
controller output U are used to calculate model response Y , the model response is
~
subtracted from the actual response Y and the difference Y −Y is used as input signal

to the controller GC .

Methodology/Approach

For this process, we are asked to perform the following:

 Use block diagram algebra to obtain the overall transfer function of

the system

 Develop a transfer function for servo proble

  Develop a transfer function for regulator problem.

 Obtain the actual response used as input signal

Figure 5;Internal model control

Analysis and Solution

From the diagram

UG P + DG d=Y … … … … … … … … … ..(1)

−(−U G¿¿ P+ Y )+Y Sp=E ¿

U GP −Y +Y Sp =E … … … … … … … … ..(2)

E Gc =U … … … … … … … … … … … … … … . ( 3 )

From eqn ( 3 )

U
E= … … … … … … … … …(4)
Gc

Substituting eqn ( 4 ) into eqn(2) to eliminate E

 U
U GP −Y +Y Sp =
Gc

Multiply through by G c

U GP Gc −Y Gc +Y Sp Gc =U

Collecting like terms and factorising to make U subject of formular

U GP Gc −U =Y Gc −Y Sp Gc

U (G ¿ ¿P Gc −1)=Y Gc −Y Sp Gc ¿ ¿

Y Gc −Y Sp G c
U= … … … … … … … … … … .(5)
( G p Gc −1 )
Substituting eqn ( 5 ) into eqn(1) to eliminate U

GP
[ Y Gc −Y Sp Gc
( G p Gc −1 ) ]
+ DG d =Y

G P Y Gc −G P Y Sp Gc + DG d ( G p G c −1 )=Y ( G p Gc −1 )

G P Y Gc −G P Y Sp Gc + DG d G p G c−DG d =Y G p Gc −Y

Collecting like terms

G P Y Gc −Y G p Gc +Y =GP Y Sp Gc + D ( Gd−G p G p Gc )

Y ( G p Gc −G p G c +1 ) =Y sp G p Gc + D ( Gd −G p G p Gc )

Y sp G p Gc D ( G d−G p G p G c )
Y= + … … … … … … … … … … … … … …(6)
( G p Gc −G p Gc + 1 ) ( G p Gc −G p Gc +1 )
To find Y/Y sp we equate D to zero (D=0)

Y G p Gc
= … … … … … … … … … … … … … ..( A)
Y sp ( G p Gc −G p Gc +1 )

Y
To find let Y sp=0∈ eqn (6 )
D

Y ( Gd−G p G p Gc )
= … … … … … … … … … … … … … .. ( B )
D ( G p Gc −G p Gc +1 )
 (B)

D
GP

TR E P + Y
+- +- GC + G

~ ~ ~ ~
Y1 - Y2 G∗(1−e−θ s )

Figure 7.0: A reduced version

P Gc
=G!c = ~ ~
E 1+ Gc G∙ (1−e−θ s )

In the outer loop, we have

Y GdG
=
D 1+G!c G

Substitute for G!c

Y GdG
=
D Gc G
1+ ~∙ ~
−θ s
1+Gc G (1−e )

~∙ ~
Y Gd G (1+G c G (1−e ) )
−θ s

=
D 1+ Gc ~
~
G∙ ( 1−e−θ s ) +Gc G

Conclusion and Results

For servo problem where the set point changes, it is assumed that disturbances is zero while
for regulatory problem, the set point remains zero.
PROBLEM 5

Background of Problem

Frequency response concept and technique play important role in control system

design and analysis. In particular they are very useful in control system design,

stability analysis and robustness analysis. Bode stability criteria are generally

applicable and provides exact result for system with time delay. Both bode and

nyquist criteria provides measures of relative stability and the two metrics for

relative stability are gain and phase margins.

Description of Problem

This problem involves the dynamic behavior of heat exchanger whose its relative

stability is to be determined, and also determine the control system design of heat

exchanger using Ziegler Nichols settings for a PI controller. It is important we

compare the open and closed loop bode plots of the system

Assumptions

 Bode stability criteria is assumed

Methodology/Approach

For this process, we are asked to perform the following:

 Obtain the block representation of the system.

 Write the closed loop transfer function of the system.

 Identify the open loop transfer function and write it out

  Derive the ultimate gain and ultimate period of the system.

 Apply the Ziegler Nichol settings for PI controller

 Obtain the corresponding gain and phase margins

 Draw both open and closed bode plots

Analysis and solution

1+GOL =0

1+GV GM G P GC =0

1+
5.264
( 0.12
0.083 s +1 0.024 s+ 1 )( 0.4322 s+ 1 )( 0.0171 s +1 ) K =0
c
 ( 0.083 s +1 )( 0.024 s +1 ) ( 0.432 s+1 )( 0.017 s+1 ) +1.2634 K c
=0
( 0.083 s+1 ) ( 0.024 s+1 )( 0.432 s +1 ) ( 0.017 s +1 )

−5 4 −4 3 2
1.4629 e s +16.8022 e s +0.057379 s + 0.556 s +1.2634 K c =0

Setting s = jw

1.4629 e−5 ( jw 4 ) +16.8022 e−4 ( jw ) + 0.0574 ( jw ) +0.0556 ( jw ) +1.2634 K c =0

3 2

−5 4 −4 3 2
1.4629 e w −16.8022 e jw −0.0574 w +0.556 jw+ 1.2634 K c =0

Real: 1.4629 e−5 w4 −0.057379 w 2+1.2634 K c =0

Imaginary:−16.8022 e−4 jw 3 +0.556 jw=0 2

From equation 2, -16.8022e−4 jw3 =0.556jw

2
0.556
w= −4
16.8022e

W=18.19rad/min

From equation , putting w=18.19rad/min

−5 4 2
1.4629 e w −0.0574 w +1.2634 k c =0

1.4629 e−5 ( 18.194 ) −0.0574 ( 18.19 ) +1.2634 K C =0

2

1.6015-0.057379( 18.19 )2=-1.2634 K C

K C =K CU =13.76

2π
For PI controller, PU =
W

2π
PU = =0.3454min
18.19

Using Ziegler Nichol PI Settings,

K CU 13.76
K C= = =6.25
2.2 2.2

PU
τ= =0.288 min
1.2

(b)
G C ( s )=6.25 1+( 1
0.288 s )
Since a PI controller has the transfer form

(
GC ( s )=K C 1+
1
τs )
Setting s=jw

√ ( )
2
1
∥ GC ( jw ) ∥=6.25 1+
0.288 w

ø C =tan−1 ( 0.288
−1
w)

−π −1
= + tan ( 0.288 w )
2

Also from A Rol ( w c )=∥ Gc ( jw ) ∥=1

ø c ( w c ) =∠ Gol ( jw c )=−180

Therefore

,
−π −1 −1 −1 −1 −1
−π= + tan (−0.288 wc )−tan ( 0.083 w c )−tan ( 0.432 wc )−tan ( 0.017 wc )−tan ( 0.024 wc )
2

w=w c =15.11rad /min

1
But Gain margin=
ARC

ARC =∥ GC ( jw ) ∥ ∥G v ( jw ) ∥ ∥ G p ( jw ) ∥ ∥ Gm ( jw ) ∥

√ ( ) [ √ (0.024 w ) +1 ][ √ (0.432 w ) +1 ][ √( 0.017 W ) +1 ][ √ (0.083 W ) +1 ]

2
1 0.12 2 1 5.264
ARC =6.25 1+
0.288 w c 2 2 2 2
c c C C

ARC =6.41 X 0.2933 X 0.1128 X 3.2818

1
GM= =1.44
0.6960

AROL=1
W g=11.78 rad /min

The phase margin can be calculated using

PM 180+ ø g

−π −1 −1 −1 −1 −1
ø= + tan (−0.288 wg )−tan ( 0.083 w g )−tan ( 0.432 w g )−tan ( 0.017 wg )−tan ( 0.024 w g )
2

=-166.8

PM =180−166.8=13.2

(C)

5.264
G v ( s )=
0.083 s +1

2
G p ( s )=
( 0.432 s +1 )( 0.017 s+ 1 )

0.12
G m ( s )=
0.024 s+1

G c ( s )=6.25 1+( 1
0.288 s )

Open transfer function GOL =GC GV G P G M

The open loop bode plot is shown below

Figure 7: open loop bode plot

Closed-loop bode plot

GV GC G P
G=
1+GOL
Figure 8.0 closed loop bode plot

PROBLEM 6

Background of Problem
Like a kitchen faucet that allows water to back flow through the pipes. The way that

works is that you turn on the main faucet handle to warm water, then turn off the

water at the faucet and don't use the sprayer.

The liquid drains freely out through the hole in the bottom of the upper tank to feed

the lower tank. From there, the liquid exits either by an outlet drain (another free-

draining hole) or by a pumped flow stream.

Description of Problem

Hot and cold liquids are mixed at the junction of two pipes and the temperature of the

resulting mixture is to be controlled using a control valve on the hot stream. And the

sensor transmitter gain is 6 rnA/rnA. The temperature sensor is located well

downstream of the junction with 8 s time delay occurs.

Assumptions

 The dynamics of the mixing process, control valve, and temperature

sensor/transmitter are negligible

 There are no heat losses/gains for the downstream pipe.

Methodology/Approach

1. Draw a block diagram for the closed-loop system using Microsoft visio.

2. Determine the Ziegler-Nichols settings (continuous cycling method) for

both PI and PID controllers using.

 3. For each controller, simulate the closed-loop responses for a unit step

change in set point using Scilab.

Analysis and Solution

a. Schematic diagram:

Block diagram:

b. GvGpGm = Km = 6 mA/mA

GTL = e-8s

GOL = GvGpGmGTL = 6e-8s

 If GOL = 6e-8s,

| GOL(jω) | = 6

∠ GOL (jω) = -8ω rad

Find ωc: Crossover frequency generates − 180° phase angle = − π radians

-8ωc = -π or ωc = π/8 rad/s

2π 2π
Find Pu: Pu ¿ = =16 s
ω c π /8

1
Find Kcu: Kcu = =0.167
¿¿

Ziegler-Nichols ¼ decay ratio settings:

PI controller:

Kc = 0.45 Kcu = (0.45)(0.167) = 0.075

τI = Pu/1.2 = 16/1.2 = 13.33 sec

PID controller:

Kc = 0.6 Kcu = (0.6)(0.167) = 0.100

τI = Pu/2 = 16/2 = 8
s τD = Pu/8 =
16/8 = 2 s
c.

1.4

1.2

0.8
y
PID control
0.6 PI control

0.4

0.2

0
0 30 60 90 120 150
t

Figure 9. Set-point responses for PI and PID control.

d. Derivative control action reduces the settling time but results in a more oscillatory
response.