Project: Complex Problem Solving (CPS) on Industrial Process

Control System A report submitted in partial fulfilment of the

requirement for the course of Control Engineering (BDA30703) SEM I
2021/2022
Faculty of Mechanical and Manufacturing Engineering
NAME MATRIC NO.
1)PUTRI FARRAH AIN BINTI AFFANDY AD190078
MEMBER

2)NURUL NATASYA BINTI IDIRS AD190147

3)NUR AMIRA DAYANA BINTI MOHD AFFENDY AD190046
4)NUR KHUZAIMAH BINTI MAT DAH AD190050
5)MUHAMMAD AFHAM BIN LOTPI AMIR DD190024
PROJECT TITLE: PID CONTROLLER IN SHELL AND TUBE HEAT EXCHANGER CONTROL SYSTEM
Weightage Rank Marks Remarks
in analysis (/5)
(WP1) Originality

Review on industrial process

0.5
Chapter 2: control
Literature
review
Review on control system
0.5
components
(WP1) Formulate
suitable solution (/5)

Mathematical model 0.25

Chapter 3:
Theoretical
calculation and Transfer function 0.25
analytical
LO5 - INDIVIDUAL & TEAMWORK & LO11 - PROBLEM

procedure
Block diagram 0.5

System construction in
(WP3) Research

Chapter 3: 0.3
based (/5)

MATLAB Simulink
Theoretical
calculation and Open-loop system analysis
0.3
analytical method and procedure
procedure Closed-loop system analysis
0.4
ANALYSIS

method and procedure

(WP3) Fundamental Professional
based & principle discipline

References Use proper format 0.5

(/5)

Report writing Well-presented and organized 0.5

analytical approach

Analysis on time response 0.4

Chapter 4:
Analysis on frequency
(/5)

Results and 0.3

response
discussion

Analysis on stability 0.3

Chapter 1:
(WP4) Familiarity

Objectives and scope 0.3

of issue (/5)

Introduction

Chapter 5: Conclusions 0.4

Conclusions and
recommendation
Recommendations 0.3

Proper attire 0.2

presentation
Group
LO5

Fluency 0.3

Communication and contents 0.5

Total (/ 35)
Table of Contents
Chapter 1: Introduction ....................................................................................................................... 2
1.1 Project Background ................................................................................................................. 2
1.2 Objectives Of the Project ........................................................................................................ 2
1.3 Scope Of Works ...................................................................................................................... 2
Chapter 2: Literature Review .............................................................................................................. 3
2.1 Review On Industrial Process Control .................................................................................... 3
2.2 Review On Control System Components ............................................................................... 3
Chapter 3: Theoretical Calculation and Analytical Procedure ........................................................ 5
3.1 Mathematical Model ............................................................................................................... 5
3.2 Block Diagram ........................................................................................................................ 6
3.3 PID Controller ......................................................................................................................... 7
3.4 Transfer Function .................................................................................................................... 7
3.5 System Construction in MATLAB/ SIMULINK .................................................................. 10
3.5.1 Starting Up SIMULINK................................................................................................ 10
3.5.2 Basic Elements .............................................................................................................. 11
3.5.3 Building A SIMULINK Model ..................................................................................... 12
3.6 Open-Loop System Analysis Method and Procedure ........................................................... 13
3.7 Closed-Loop System Analysis Method and Procedure ......................................................... 14
3.7.1 Before Inserting PID Controller .................................................................................... 14
3.7.2 After Inserting the PID Controller ................................................................................ 16
Chapter 4: Results and Discussions................................................................................................... 19
4.1 Analysis On Time Response ................................................................................................. 19
4.2 Analysis On Frequency Response......................................................................................... 22
4.2.1 Bode Plot Analysis ........................................................................................................ 22
4.3 Analysis On Stability ............................................................................................................ 24
Chapter 5: Conclusions and Recommendations .............................................................................. 26
5.1 Conclusions ........................................................................................................................... 26
5.2 Recommendations ................................................................................................................. 26
References ............................................................................................................................................ 27

1
 Chapter 1: Introduction
1.1 Project Background
Heat transfer using a heat exchanger system is a common technology development in
process industries. As it can withstand a wide range of temperature and pressure, heat
exchanger systems are commonly used in chemical plants. Since this primary objective of a
heat exchanger system is to transfer heat from a hot fluid to a cooler fluid, temperature
control of the exit fluid is important. A traditional PID controller may be used to manage the
temperature of the heat exchanger system's output fluid.
This paper describes a study that takes a heat exchanger and creates a single-input
single-output model of the system using experimental data. According to a process
requirement, the heat exchanger system's output temperature must be kept at a desired set
point.
1.2 Objectives Of the Project
This project's objectives include:
a) Identifying the system's input and output.
b) Identifying the transfer function of the considered system.
c) Using suitable methods, analyze the stability, steady-state errors, and transient
responsiveness of the controlled system.
d) Use MATLAB to perform problem-solving strategies and test the ideas presented in
basic control system by hand calculation and simulation.
e) Study the impact of proportional control.

1.3 Scope Of Works

The scope of this project is to focus on the derivation of transfer function based on the
considered system and analyze the transient respond when the system is subjected to a unit
step input. Transient response is analyzed by constructing the root locus of the system and
verification of calculation using MATLAB software. The stability of the frequency response
can be determined by applying Nyquist and bode plot based on the system.

2
 Chapter 2: Literature Review
2.1 Review On Industrial Process Control
Every chemical reaction involves the production or absorption of energy in the form of
heat. In a chemical process, a heat exchanger is typically used to transfer heat from a hot fluid
to a cooler fluid via a solid wall. There are several types of heat exchangers used in the
industry, but the shell and tube heat exchanger system is the most common.
The most popular type of heat exchanger is the shell and tube heat exchanger, which
can operate at a broad range of temperatures and pressures. It offers a higher heat transfer
surface-to-volume ratio than other heat exchangers and is simple to construct in a wide range
of sizes and configurations. The shell and tube heat exchanger can withstand high pressures,
and its design allows for easy disassembly for frequent maintenance and cleaning. A shell-
and-tube heat exchanger is a variation on the double-pipe design. A shell-and-tube heat
exchanger, as opposed to a single pipe within a bigger pipe, is made up of a bundle of pipes
or tubes contained within a cylindrical shell. One fluid passes through the tubes of a shell and
tube heat exchanger, and a second fluid flows within the gap between the tubes and the shell.

Figure 2.1 Shell and Tube Heat Exchanger

2.2 Review On Control System Components

A chemical reactor and a shell and tube heat exchanger system are common
components of an interacting chemical process for heating. The boiler produces superheated
steam, which passes through the tubes. The process fluid, on the other hand, runs through the
shells of the shell and tube heat exchanger system. In this report, many assumptions have
been considered. The first assumption is that the fluid intake and outflow rates are the same.
The second assumption is that the insulating wall's heat storage capacity is negligible.
A thermocouple is used as the sensing element in the control architecture's feedback
path. The thermocouple measures the temperature of the outgoing fluid, and the

3
thermocouple output is delivered to the transmitter unit, which finally transforms the
thermocouple output to a standardized signal in the 4mA -20mA range. The transmitter unit's
output is sent to the controller unit. The controller executes the control algorithm, compares
the output to the set point, and then sends the appropriate commands to the final control
element through the actuator unit. The valve opens and closes in accordance with the
controller's instructions. The actuator is a current-to-pressure converter, and the final control
unit is an air-to-open (fail-close) valve. The actuator unit converts the controller output in the
4-20 mA range into a standardized pressure signal in the 3-15 PSI Gauge unit range. The
fundamental feedback control strategy employed in a heat exchanger system is visualized in
Figure 2.2.

Figure 2.2 Schematic Diagram of Temperature Control of Heat Exchanger.

4
 Chapter 3: Theoretical Calculation and Analytical Procedure

3.1 Mathematical Model

In this section, the valve, process, and thermocouple sensor are mathematically
modelled using the available experimental data. The experimental process data are
summarized as in table below.

Table 1 Experimental Data

Experimental Parameters Experimental Data

Exchanger response to the steam flow gain 50°𝐶/(𝑘𝑔/𝑠𝑒𝑐)
Time constant 30 𝑠𝑒𝑐𝑜𝑛𝑑𝑠
Exchanger response to variation of process 1°𝐶/(𝑘𝑔/𝑠𝑒𝑐)
fluid gain
Exchanger response to variation of process 3°𝐶
temperature gain
Control valve capacity for steam 1.6 (𝑘𝑔/𝑠𝑒𝑐)
Time constant of control valve 3 𝑠𝑒𝑐𝑜𝑛𝑑𝑠
The range of thermocouple sensor 50°𝐶 𝑡𝑜 150°𝐶
Time constant of temperature sensor 10 𝑠𝑒𝑐𝑜𝑛𝑑𝑠
Gain of valve 0.1
Gain of thermocouple sensor 0.16

From the experimental data, we can find the transfer function of the valve, process, and
thermocouple sensor.

By using the standard form of 1𝑠𝑡 order differential equation is;

𝑘
𝑇(𝑠) =
𝑇𝑠 + 1

Where;

𝑘 = 𝐺𝑎𝑖𝑛

𝑇 = 𝑇𝑖𝑚𝑒 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡

5
 i. The process transfer function
𝐸𝑥𝑐ℎ𝑎𝑛𝑔𝑒𝑟 𝑟𝑒𝑠𝑝𝑜𝑛𝑠𝑒 𝑡𝑜 𝑠𝑡𝑒𝑎𝑚 𝑓𝑙𝑜𝑤 𝑔𝑎𝑖𝑛 = 50°𝐶/(𝑘𝑔/𝑠𝑒𝑐)

𝑇𝑖𝑚𝑒 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 = 30 𝑠𝑒𝑐𝑜𝑛𝑑𝑠

𝑘 50
∴ 𝑇(𝑠) = =
𝑇𝑠 + 1 30𝑠 + 1

ii. The valve transfer function

𝐺𝑎𝑖𝑛 𝑜𝑓 𝑣𝑎𝑙𝑣𝑒 = 0.1
𝑇𝑖𝑚𝑒 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 = 3 𝑠𝑒𝑐𝑜𝑛𝑑𝑠
𝑘 0.1
∴ 𝑇(𝑠) = =
𝑇𝑠 + 1 3𝑠 + 1
iii. The thermocouple transfer function
𝐺𝑎𝑖𝑛 𝑜𝑓 𝑡ℎ𝑒𝑟𝑚𝑜𝑐𝑜𝑢𝑝𝑙𝑒 = 0.16
𝑇𝑖𝑚𝑒 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 = 10 𝑠𝑒𝑐𝑜𝑛𝑑𝑠
𝑘 0.16
∴ 𝑇(𝑠) = =
𝑇𝑠 + 1 10𝑠 + 1

3.2 Block Diagram

From the obtained data, the heat exchanger temperature control loop diagram is obtained as
follows:

𝐺(𝑠)

+
𝑅(𝑠) 𝐶(𝑠)
0.1 50
𝐾
3𝑠 + 1 30𝑠 + 1
− PID Process
Valve
Controller

0.16
10𝑠 + 1

𝐻(𝑠)

Figure 3.2 Block Diagram

6
3.3 PID Controller
Because current control systems are automated, it is preferable to have auto-control
rather than manual control in most situations. It is a feedback control system featuring
Proportional, Integral, and Derivative control.
The proportional control has a crucial element known as the proportional gain 𝐾𝑝 ,
which amplifies the output proportionally to the received 'error' signal. The 'error' in integral
control is collected and multiplied by the integral gain 𝐾𝑖 . The output of derivative control
varies with the rate of change of the 'error' concerning time multiplied by the derivative gain
𝐾𝑑 .
In case of combined PID control the relation is as follows:
𝑡
𝜕𝑒(𝑡)
𝑢(𝑡) = 𝐾𝑝 𝑒(𝑡) + 𝐾𝑖 ∫ 𝑒(𝑡)𝑑𝑡 + 𝐾𝑑
0 𝜕𝑡
𝑈(𝑠) 1
= 𝐾𝑝 (1 + + 𝑇𝑑 𝑠) − − − − − (1)
𝐸(𝑠) 𝑇𝑖 𝑠
𝐾𝑝
𝑤ℎ𝑒𝑟𝑒, 𝐾𝑖 = , 𝐾𝑑 = 𝐾𝑝 𝑇𝑑
𝑇𝑖
From (1)
𝑈(𝑠) (𝑇𝑑 𝑇𝑖 𝑠 2 + 𝑇𝑖 𝑠 + 1) 𝐾𝑝
= 𝐾𝑝 − − − − − (2)
𝐸(𝑠) 𝑇𝑖 𝑠

3.4 Transfer Function

We utilise Ziegler-Nichols rules and Routh's criterion to determine the transfer function
of a PID controller. So, as to determine the values of 𝐾𝑝 , 𝑇𝑖 , 𝑇𝑑 and, we will use Ziegler-
Nichols' second method, in which we first set 𝑇𝑖 = ∞ and 𝑇𝑑 = 0, then use only proportional
control 𝐾𝑝 by increasing it from 0 to a critical value 𝐾𝑐𝑟 at which we get sustained
oscillations to get the corresponding period 𝑃𝑐𝑟 . Then we calculate the values as follows:

Table 2 Ziegler-Nichols rules

Type of Controller 𝐾𝑝 𝑇𝑖 𝑇𝑑
PID 0.6 𝐾𝑐𝑟 0.5 𝑃𝑐𝑟 0.125 𝑃𝑐𝑟

7
So, based on our block diagram shown before, which is:

Figure 3.4 Block Diagram

𝐶(𝑠) 5𝐾(10𝑠 + 1)
=
𝑅(𝑠) 900𝑠 + 420𝑠 2 + 43𝑠 + 1 + 0.8𝐾
3

Routh-Hurwitz stability criterion may be used to determine the value of 𝐾𝑝 that will
render the system moderately stable during prolonged oscillations. The near loop system's
characteristic equation is as follows:
Take 𝑑𝑒𝑛𝑜𝑚𝑖𝑛𝑎𝑡𝑜𝑟 = 0
900𝑠 3 + 420𝑠 2 + 43𝑠 + 1 + 0.8𝐾

Table 3 Determination of stability using Routh-Hurwitz stability criterion

𝑠3 900 43
𝑠2 420 1 + 0.8𝐾
𝑠1 420(43) − 900(1 + 0.8𝐾) −
420
𝑠0 1 + 0.8𝐾 −

To reach the stability of the system, the 1𝑠𝑡 column element must be positive.
420(43) − 900(1 + 0.8𝐾)
>0
420

8
𝐾 < 23.833
By examining the first column, we observed that sustained oscillations will occur if
𝐾𝑝 = 23.833, thus the critical gain 𝐾𝑐𝑟 = 23.833.
Then,
420𝑠 2 + 1 + 0.8𝐾 = 0
420𝑠 2 = −20.0664
𝑠 2 = −0.0478
𝑠 = ±0.2186
Thus, frequency sustained oscillations, 𝜔 will be 0.2186.
𝜔 = 0.2186
2𝜋 2𝜋
𝑃𝑐𝑟 = = = 28.7428 𝑠
𝜔 0.2186
By referring to Table 2
𝐾𝑐𝑟 = 23.833
𝑃𝑐𝑟 = 28.7428
𝐾𝑝 = 0.6 𝐾𝑐𝑟 = 14.2998
𝑇𝑖 = 0.5 𝑃𝑐𝑟 = 14.3714
𝑇𝑑 = 0.125 𝑃𝑐𝑟 = 3.5929
By substituting the value of 𝐾𝑝 , 𝑇𝑖 and 𝑇𝑑 into the equation (2), we get the transfer function
for PID as follows:
𝑈(𝑠) 51.38𝑠 2 + 14.30𝑠 + 0.995𝑠
= − − − − − (3)
𝐸(𝑠) 𝑠

From the obtained data, the system PID control loop diagram is obtained as follows:

Figure 3.5 Block diagram with PID controller designed

9
So, the Closed Loop Transfer Function (CLTF) of the system as follows:
𝐶(𝑠) 2569𝑠 3 + 958.4𝑠 2 + 119.9𝑠 + 4.975
=
𝑅(𝑠) 900𝑠 4 + 420𝑠 3 + 84.104𝑠 2 + 12.224𝑠 + 0.796

3.5 System Construction in MATLAB/ SIMULINK

While the conventional MATLAB package can be used to analyse linear systems,
SIMULINK is significantly more effective for control system simulation. SIMULINK is a
graphical interface for modelling, simulation, and analysis of dynamic systems where it is a
MATLAB add-on product. It allows for the rapid construction of virtual prototypes that may
be used to explore design concepts at any level of detail with minimal effort. SIMULINK has
a graphical user interface (GUI) for creating models in the form of block diagrams. It comes
with a large library of predefined blocks that may be used to create graphical representations
of systems by dragging and dropping them. The user can create a "up-and-running" model
that would otherwise take hours to construct in a laboratory setting. It can represent linear and
nonlinear systems in continuous time, sampled time, or a combination of both. Since students
learn more effectively when they receive frequent feedback, the interactive nature of
SIMULINK encourages you to try things out as you can change settings and observe what
occurs right away. Therefore, below we show how to construct a system or a block diagram
in SIMULINK.
3.5.1 Starting Up SIMULINK
1) First, we must start MATLAB to use SIMULINK.

2) Click the SIMULINK icon on the MATLAB toolbar.

3) In response, SIMULINK start page will appear and click on the blank model to
create a new model of block diagram.

10
 4) A blank window then appears on the screen and then, in this mode window,
models are drawn and edited mainly by mouse drive commands.

3.5.2 Basic Elements

There are two major categories of elements in SIMULINK:
⚫ Blocks

⚫ Line

Blocks are used to generate, modify, combine, output and display signals while lines
are used to transfer signals from one block to another. In the SIMULINK library
browser, there are several general classes of blocks, some of which we are used to
construct our block diagram are:
➢ Sources: Step. In SIMULINK the symbol is known as step. In a block diagram
it is an input or reference.

➢ Sinks: Scope. In SIMULINK the symbol is known as scope but in a block

diagram it is an output.

➢ Continuous: Transfer function and PID controller.

➢ Math operations: Gain and sum.

➢ Signal routing: Mux.

Function of lines is to transmit signals in the direction indicated by the arrow from the
output terminal of one block to the input terminal of another block.

11
3.5.3 Building A SIMULINK Model
1) Remember that we are already have a blank window to create a block diagram.
Then, click on the library browser icon and a few options are listed as shown in
figure below.

2) On the SIMULINK library browser, click on the source’s icon. Sources are used
to generate signals. Once the sources window opened find for the step blocks and
drag it into the left side of the model window.

3) Click the sink icon and drag the scope blocks into the right side of the model
window as it shows the output of our system.

4) Click the continuous icon and drag the PID controller and transfer function blocks
in between the step and scope.

5) Click the math operations icon and drag the sum blocks in between the step and
PID controller.

6) Click the signal routing icon and drag the mux blocks in between the transfer
function and scope.

7) Connect all the blocks with a line and form a block diagram as below.

12
3.6 Open-Loop System Analysis Method and Procedure
Just now, we have enough information on how to generate a block diagram.
Therefore, by applying the same step as stated above, we need to create an open-loop
system as below.

The s-polynomials in the transfer function process were entered by double-clicking on

the transfer function blocks and entering the coefficients for the numerator and
denominator polynomials. The default values of default step have been changed by
double-clicking the step icon. We change the initial value to 0 and final value to 10. after
the value has changed, click Apply button and click OK.

13
 Then, click a run icon to run the simulation and the reader should generate the
simulations and observe the behaviour of the output with respect to a step input change.
The output should appear as an overdamped graph like the picture below for open-loop
system.

In open-loop system it is differ with closed loop system since it is not involving a PID
controller also it does not have feedback. It is a direct system starting from the input whereby
the input is generated and end with output.

3.7 Closed-Loop System Analysis Method and Procedure

For closed-loop system, we have brought into condition, first, before inserting the PID
controller and second, after inserting the PID controller.

3.7.1 Before Inserting PID Controller

By using block diagram of the open-loop system, we are required to add the sum
block, gain block and mux block. To insert the sum, gain and mux block, follow the
method below:
1) Click on the library browser icon.

2) Then, click on the math operations icon to find the sum and gain block and drag it
into the model window.

3) Place the sum block next to input and the gain block in between the sum block
and transfer function block.

4) Connect all the blocks added to the existing block diagram.

14
5) Make sure the initial and final value of input as below by double-clicking on the
input block.

6) Double-clicking on the gain block to change the value of K to 0.2102. Click

Apply button and click OK button.

15
 7) Click run icon to run the simulation.

8) Double-clicking on the scope to get the result of the system.

Based on the graph that we obtained, the line form as an overdamped graph where the
percentage of overshoot is 0.505%, the rise time is 360.822 s, and the settling time is 208.167
s.
3.7.2 After Inserting the PID Controller
Applying the same block diagram as the block diagram before PID controller has been
inserted but the difference is we must change the gain block to with the PID controller
block by following the step below:
1) Click on the gain block and tap on the cut icon.

2) Click on the library browser icon and click the continuous icon

3) Find and drag the PID controller blocks and placed it in between the sum and
scope blocks.

4) Connect the PID controller to the existing block diagram.

16
5) Make sure the initial and final value of input as below.

6) Change the controller turning parameters of the proportional (P) = 0.25, the
integral (I) = 0.002, the derivative (D) = 6 and the filter coefficient (N) = 10 as
below.

17
7) Then, click Apply button and OK button.

8) Click the run icon to run the simulation.

9) Double-clicking the output block to get the results of the system as below.

Based on the graph that we obtained, the line form as an overdamped graph where the
percentage of overshoot is 0.505%, the rise time is 360.822 s and the settling time is
208.167 s.

18
 Chapter 4: Results and Discussions
4.1 Analysis On Time Response
Transient response and steady state response are two kinds of time response. Transient
response is which goes from initial to final state and steady state response is refers to
behavior of output of system from 𝑡 → ∞.

Transient Response
A transient response is a system's out characteristic. It is usually exhibiting damped
oscillations before achieving steady state, as expressed in a transient response after
parameters are considered.

• Delay Time (𝑡𝑑 ): It is the time required for the response to reach half the final value
for first time.
• Rise Time (𝑡𝑟 ): It is the time required by the response to reach to its final or 90%
value from: 0 to 10%.
• Peak Time (𝑡𝑝 ): It is the time required by the response to reach to first peak of the
overshoot.
• Maximum overshoot (𝑀𝑝 ): It is the maximum peak value reached reference to the
unity value.
• Settling Time (𝑡𝑠 ): It is the time required by the response to reach to become stable
within an allowable range.

Figure 4.1 Unit step response represented in equation of transfer function

19
Then we will perform transient analysis. From the Figure, first we will find the damping
ratio,𝜁 and natural frequency, 𝜔𝑛 . We found that:

Mp1 = 1.56
Mp2 = 1.04
t p1 = 6.23
t p2 = 18.8
t d = 2.5
y∞ = 1.002

Mp2 − y∞
d=
Mp1 − y∞
1.04 − 1.002
d=
1.56 − 1.002
d = 0.07

Tp = t p2 − t p1
Tp = 18.8 − 6.23
Tp = 12.57

So, we can find the damping ratio,ζ and natural frequency, ωn :

1
Damping ratio, ζ =
√1 + ( 2π )²
ln d
1
Damping ratio, ζ =
2
√1 + ( 2π )
ln 0.07
Damping ratio, ζ = 0.3898

2π
Natural frequency, ωn =
Tp √1 − ζ²
2π
Natural frequency, ωn =
12.57√1 − 0.3898 ²
Natural frequency, ωn = 0.5428

Then we find the transfer function associated with the value of 𝜁 and 𝜔𝑛 :

20
C(s) ωn ²
=
R(s) s² + 2ζωn s + ωn ²
C(s) 0.2946
R(s)
=
s²+0.4232s+0.2946
− − − − −(4)

Output response for the above transfer function is:

Figure 4.2 Unit step response represented in equation 4

√1 − ζ²
∅ = tan−1
ζ
√1 − 0.3898²
∅ = tan−1
0.3898
∅ = 1.17 rad

ωd = ωn √1 − ζ²

ωd = 0.5428√1 − 0.3898²
ωd = 0.4999 rad/sec

Thus,

π−∅
Rise Time(t r ) =
ωd
π − 1.17
RiseT ime(t r ) =
0.4999

21
Rise Time(t r ) = 3.94

π
Peak Time (t p ) =
ωd
π
Peak Time (t p ) =
0.4999
Peak Time (t p ) = 6.284

Maximum overshoot (Mp ) = e−ζπ/√1−ζ²

Maximum overshoot (Mp ) = e−(0.3898xπ)/√1−0.3898²

Maximum overshoot (Mp ) = 0.26453

4
Settling Time (t s ) =
ζωn

4
Settling Time (t s ) =
0.3898x0.5428
Settling Time (t s ) = 18.9051

According to the theoretical calculation, the transient response parameters are as above and
according to simulation, t r = 3.89, t p = 6.24, Mp = 0.3214 and t s = 20.
Steady State Response
Now as can observe that control system is type 1 and the steady state error for unit step input
for type 1 is equal to zero. So, 𝑒𝑠𝑠 = 0

4.2 Analysis On Frequency Response

The response of the system to an input sine function is referred to as frequency
response analysis. The ‘Bode Plot’, which is the same as a step response in time domain, and
the ‘Nyquist Plot’, which will provide light on absolute and relative stability analysis, will be
used to show frequency response analysis.
4.2.1 Bode Plot Analysis
In the frequency response, it is a graphical representation of the system transfer
function. There are two graphs in it which are a magnitude plot and a phase plot. For drawing
first, we get the open loop transfer function in frequency domain by replacing s by jω. So, we
have G(jω) from Equation 4 as
0.2946
𝐺(𝑗𝜔) = (𝑗𝜔)2+0.4232𝑗𝜔 − − − − −(5)

22
Magnitude of resonance point (𝑀𝑟 ). It is maximum value dimension in 𝐺(𝑗𝜔). It is related to
dimensions of overshoot and damping ratio, 𝜁.
1
𝑀𝑟 =
2𝜁√1 − 𝜁 2
1
𝑀𝑟 =
2(0.3898)√1 − (0.3898)2
𝑀𝑟 = 1.39

It also expresses relative stability if 1.0 < Mr <1.4 i.e., 0dB – 3dB, it corresponds to
effective damping ratio 0.4<ζ<0.7 for satisfactory performance. If the value is greater than
1.5 then multiple overshoots will be available.
Resonant Frequency (𝜔𝑟 ) it is the frequency at maximum resonant point and is related to
natural frequency(𝜔𝑛 ) as 𝜔𝑟 = 𝜔𝑛 √1 − 2𝜁 2 it indicates the speed of response.

𝜔𝑟 = 0.5428√1 − 2(0.3898)2
𝜔𝑟 = 0.453
Bandwidth (𝐵𝑤) is a measure of how much data can be. It shows how the sine wave
is traced by the system. The higher the bandwidth, the better the high frequency passes. It is
related to rise time and in general is proportional to response time. However, it is suggested
that 𝑀𝑟 be small and 𝜔𝑟 be large.

𝐵𝑤 = 𝜔𝑛 √1 − 2𝜁 2 + √2 − 4𝜁 2 + 𝜁 4

𝑥 = √2 − 4(0.3898)2 + (0.3898)4 = 1.18967

𝐵𝑤 = 0.5428√1 − 2(0.3898)2 + 𝑥
𝐵𝑤 = 0.7454

Gain & Phase Margin (Gm & Pm) Gain margin is difference between gain magnitude
and 0dB at phase transit frequency 𝜔𝑛 . Phase margin is difference between the phase and -
180˚ when gain is 0dB at gain transit frequency 𝜔𝑔 .
𝑃𝑚 = 100 𝑥 𝜁 = 100 × 0.3898
𝑃𝑚 = 38.98
If the gain margin is negative below 0dB then it indicates stability. Grater the better.
These margins indicate the margin gain and phase has for system stability.
From simulation the values of 𝑀𝑟 = 1.39, 𝜔𝑟 = 0.453, 𝐵𝑤 = 0.7454, 𝐺𝑚 = ∞, 𝑃𝑚 =

23
38.98, 𝜔𝑝 = ∞ 𝑎𝑛𝑑 𝜔𝑔 = 0.4629 are Bode plot is

Figure 4.3 Bode Plot

4.3 Analysis On Stability

Nyquist stability criterion determines the stability of closed loop system from its open
loop frequency response. It has two conditions for stability:
i) If the Nyquist contour for open loop system does not encircle -1+j0 point, then the
close loop system is stable.
ii) Another condition is, there should be no poles to right of the s-plane

Let us consider the open loop transfer function represented by Equation 5. The
Nyquist plot for the same is as

24
 Figure 4.4 Nyquist plot

The Nyquist stability criterion can also be expressed as Z= N+P where

Z= is number of zeros in right- half plane of s-plane.
N= is number of encircles of -1+j0
P= is number of Poles in right- half plane of s-plane.

As it is clear from the figure 4.4 that Z=0, P=0 & N=0. So, we can conclude that the system is
stable.

25
 Chapter 5: Conclusions and Recommendations
5.1 Conclusions
Temperature control is a critical control component in most industrial applications. In
this study, a feedback controller is constructed to control the output fluid temperature of a
shell and tube heat exchanger. The feedback control system is used to optimise the
performance of the heat exchanger system's anti-disturbance capabilities. The transient
response is used to evaluate the performance of the system controllers. As a result, using
MATLAB Simulink, the whole mathematical model component of the dynamic heat
exchanger system is effectively studied and developed. The simulation results demonstrate
that the Feedback controller, in combination with the PID controllers, produces good results
with minimal overshoot and the shortest settling time.
Eventually, our project's objective has been accomplished. We have successfully
studied and constructed physical structures for real-life applications utilising time and
frequency response methodologies. Furthermore, we have allowed the transfer function to be
created and the procedure to be solved using theoretical calculations and tested using
MATLAB.
5.2 Recommendations
It is suggested that the student learn to use MATLAB properly in terms of drawing and
understanding the graph from this project. Not just that, but it would be extremely beneficial
for the student to learn if the lecturer provided a real-life example of using all the method
used above.

26
References

T N Luan et.al, “Design of Multi-Loop PID Controllers Based on the generalized IMC-PID
Method with Mp Criterion,” International Journal of Control, Automation, and
Systems, vol. 5, no. 2, 2007, pp. 212-217.
P.E.Welstead “Introduction to Physical System Modelling” Electronic Ed., Academic Press
Ltd. 2006, ISBN 0-12-744380-0
I.J.Nagrath, M.Gopal “Control System Engineering” 5th Ed. New Age International
Publishers,2010, ISBN 978-81-224-2008-7
Katsuhiko Ogata “Modern Control Engineering” 5th Ed. Pearson Prentice Hall, 2010, ISBN-
978-81-203-4010-7.

27