Department of Chemical and Environmental Engineering (ChEE)

Malaysia-Japan International Institute of Technology (MJIIT)

Universiti Teknologi Malaysia (UTM) Kuala Lumpur

Development of computer simulation platform for Rasau water

Title
treatment plant

No. Name Matrix ID

1 TANG KUI EE A21MJ0127

2 TANG YU LYN A21MJ0096

3 TEE ZHEN XIANG A21MJ0098

4 WONG JUN YONG A21MJ0131

 TABLE OF CONTENTS

1.0 Introduction............................................................................................................................1
2.0 Plant Simulation Model......................................................................................................... 4
2.1 Transfer Function...............................................................................................................4
2.2 Ziegler-Nichols Tuning Method.......................................................................................... 5
3.0 Methodology.......................................................................................................................... 7
3.1 Ziegler Nichols Tuning Method.......................................................................................... 7
4.0 Results.................................................................................................................................. 11
4.1 Open Loop....................................................................................................................... 11
4.2 Closed Loop.....................................................................................................................12
4.2.1 P controller....................................................................................................................13
4.2.2 PI controller...................................................................................................................18
4.2.3 PID controller................................................................................................................ 22
4.3 Cascade controller...........................................................................................................28
5.0 Conclusion........................................................................................................................... 33
6.0 Reference............................................................................................................................. 34
1.0 Introduction

Our team is gearing up for a significant project focused on creating a computer

simulation platform tailored for the first stage of the Rasau Water Treatment Plant (WTP). The
primary goal of this process simulator is to facilitate training for operators, providing them with
insights into the intricacies of the various processes within the WTP. Throughout this initiative,
we'll be making use of programmable I/O controllers, specifically incorporating flow rate and
temperature sensors. These sensors will play a crucial role in actively monitoring and controlling
each treatment process, including chemical dosing, filtration, and sterilization.

We have been tasked with designing a control system for the sterilization process. In
typical scenarios, sterilization involves the use of heat, chemicals, or radiation to eliminate
microorganisms. However, at the first stage of the Rasau WTP, the water undergoes a heating
process (following UV radiation) before being supplied to end users. This unique approach
ensures the thorough elimination of all microorganisms. To guide our decision-making in this
project, all the necessary information for developing the process simulator is conveniently
provided in Appendix A. It is crucial for our team to make a thoughtful selection of a single
control system for the sterilization process, aligning with the specific requirements and nuances
of the Rasau WTP. The primary method for deactivating harmful organisms to prevent the
transmission of waterborne diseases to downstream users and the environment is recognized as
disinfection. It is crucial for wastewater to undergo sufficient treatment before disinfection to
ensure the effectiveness of any disinfectant. Various microorganisms commonly present in
domestic waste. Therefore, the ideal lamp wall temperature is between 95°F and 122°F, which is
respectively 35°C and 50°C. 50°C is chosen as our main objective setpoint for the operating
control system in Simulink.

Feedback controllers play a crucial role in water treatment process plants by continuously
monitoring key parameters, including water flow rates and temperature. These controllers
compare the measured values with desired setpoints, identifying any deviations as errors. For
instance, if the temperature falls outside the target range, it signifies an error in the process. In
response to detected errors, feedback controllers take corrective action to bring the system back
to optimal operating conditions. This involves adjusting inputs such as chemical dosages, flow
rates, or treatment processes. PID control algorithms are commonly used, incorporating

1
proportional, integral, and derivative components to ensure a balanced and efficient control
response. Feedback controllers take action when the controlled variable deviates from its set
point, as detected by the nonzero value of the error signal (Hahn, 2003). One key function of
feedback controllers is to prevent oscillations and instabilities in the water treatment process. By
continuously adjusting control inputs based on the feedback loop, these controllers maintain
system stability, avoiding abrupt changes or fluctuations that could impact the treatment
efficiency.

Flow Process Control System 1

Industrial control systems (ICS), also known as process control systems (PCS), serve as
devices within manufacturing that assess the production process through various methods and
supply data for monitoring and addressing issues. Industrial control system (ICS) is a collection
of control systems and associated tools that encompass the instruments, networks, systems, and
controls utilized in the operation and/or automation of industrial processes(Arora, 2022). There
exists a range of process control systems, such as distributed control systems (DCS) or
programmable logic controllers (PLC), which gather and transmit data obtained during the
manufacturing process.

A relatively straightforward PCS includes the primary transducer, which is a sensor

accepting input, a controller processing this input, and a receiver handling the output. The
objective of process control systems is to maintain process conditions close to their design values
to prevent the occurrence of abnormal operational situations. In these investigations, controllers
of the proportional (P), proportional integral (PI), and proportional integral derivative (PID)
types are employed.

In feedback control, the objective is to reduce the error signal to zero where,
e(t) = 𝑦𝑠𝑝(𝑡) − 𝑦𝑚(𝑡). Whereas e(t) indicates error signal, 𝑦𝑠𝑝(𝑡) indicates set point, and 𝑦𝑚(𝑡)

indicates the measured value of the controlled variable (Seborg, 2016). In our project, we
consider servo which may apply to the cleaning process of heat exchanger. RYDLYME should
be used as directed and at any temperature within the operating limits of between 0°F / -18°C
and 180°F / 82°C (Apex,2022). Moreover, we consider disturbance during a real life situation,

2
which could be a leakage of a piping system which may affect the flow rate of the system which
eventually affects the process desired.

P controller is a type of control system that calculates the input point based on the error
between the desired output and the current output. The control input, proportional to the error, is
then provided to the system to reduce the error and bring the output closer to the intended value.
P controllers find frequent use in industrial control systems and other applications requiring swift
response and robust stability.

A PI controller, on the other hand, determines the control input by considering both the
error and the integral of the error. Similar to the P controller, the control input is applied to
reduce the error and bring the output closer to the desired value. The integral component in a PI
controller helps diminish the potential for steady-state error associated with a P controller.

A PID controller, a feedback-based control loop mechanism, is commonly utilized in

industrial control systems and applications requiring continuous modulated control. This
controller continuously calculates an error value (PV) when there is a difference between a
desired setpoint (SP) and a measured process variable. Practically, the PID algorithm
automatically adjusts the control function in a precise and timely manner, minimizing delays and
overshoots by incrementally adjusting the engine's power output to return the measured speed to
the target speed.

Control Objective

A control system is employed to regulate the output temperature by adjusting the rate of
steam flow. This means that the system actively manages the flow of steam to ensure precise
control over the temperature of the output.

Process Control Variables

The variables in this context can be categorized into control variables, manipulated
variables, and disturbances. Control variables, including level, flow, temperature, and pressure,

3
are those that require precise regulation. In the case of this reactor, the temperature is identified
as the control variable that needs to be closely monitored and controlled. Manipulated variables,
on the other hand, are those that need to be adjusted or altered to meet the set point requirement,
also known as the plant objective. In this reactor, the flow rate of steam output is identified as the
manipulated variable requiring adjustments to achieve the desired set point.

2.0 Plant Simulation Model

Figure 2.1 Conventional P&ID for CSTR Figure 2.2 Cascade control P&ID for CSTR

Controlled Variable Temperature

Manipulated Variable Flow Rate

Control Strategy Conventional and Cascade control

Disturbance Temperature at 100s

Setpoint Temperature of water at 50oC and 80oC

Measured Variable Temperature

Table 2.1 Process Variables for Continuous Reactor

2.1 Transfer Function

For this operation, it is decided that we will be changing the temperature of the reactor

4
(CV) by manipulating the flow rate of steam when heating. Thus, we will use this transfer
function in order to achieve the plant objective.

Table 2.2.1 Transfer function

Component Transfer function

Process (Temperature-flowrate) −𝑠
1.4 𝑒
2𝑠+1.2

Actuator 1.25
3𝑠+2

Sensor 0.16
𝑠+2

Disturbance 4
4𝑠+5

Cascade inner process (flowrate-flowrate) 𝑒

−𝑠

2𝑠+2

2.2 Ziegler-Nichols Tuning Method

Ziegler and Nichols introduced PID controller tuning methods in 1942, which have since
been extensively employed in their original or modified forms even until nowadays. The
Ziegler-Nichols step response method is the classical tuning method for PID controllers. The
primary benefit of the proposed approach is its frequent application in situations where the
mathematical model of the system is unavailable. The Ziegler-Nichols method is applicable to
both closed and open-loop systems. The subsequent sections provide detailed discussions on
various types of PID controller tuning methods.

(a) Open-loop system

The step response approach relies on conducting an open-loop step response test for the
process, necessitating process stability. The unit step response is defined by two parameters,
denoted as L and T, which are determined by drawing a tangent line at the inflection
point—where the step response slope is at its maximum. The points where the tangent intersects
with the coordinate axes provide the process parameters, as illustrated in Figure 3.1. These
parameters are then employed in the computation of controller parameters.

5
 Figure 2.1 Shaped response curve for Ziegler-Nichols Tuning Algorithm

(b) Closed-loop system

In this procedure, the initial step involves selecting proportional control (Kp) alone. The
sustained oscillation method or known as closed-loop method for Ziegler-Nichols tuning is
shown in Figure 3.2. The subsequent phase entails incrementally raising the proportional gain
until the system reaches the verge of instability, characterized by sustained oscillations, thereby
determining the critical gain value (Kcr). Following this, the oscillation period is measured to
derive the critical time constant (Pcr). Once both Kcr and Pcr values are established, the PID
parameters can be calculated, adhering to the design specifications outlined. This systematic
approach ensures the effective tuning of PID controllers for optimal system performance.

Figure 2.2 Sustained oscillation method/ closed-loop method for Ziegler-Nichols tuning

6
3.0 Methodology

3.1 Ziegler Nichols Tuning Method

3.1.1 Open Loop (Manual Mode) Tuning

Figure 3.1.1: Initial Gain Value and Temperature (SP and PV) Display for Manual Controller.

Step 1: Figure 3.1.1.1 shows the initial value of the gain controller in the open loop system.
According to Vivek and Ashih (2016), to tune the open loop system, we have to change the gain
value manually of the controller to achieve the setpoint.

Figure 3.1.1.2: Gain Value After Manual Tuning.

Figure 3.1.1.2 shows the optimum gain value for the system to achieve the setpoint.

7
Step 2: Measure the time delay, L and the time constant, T as shown in Figure 3.1.1.3.

Figure 3.1.1.3: T Value and L Value Obtained.

𝑇
Step 3: Find the KP. The new gain controller value can be calculated using KP = 𝐿
, which is

17.125 in this case. Then, enter the new gain value to the controller.

8
3.1.2 Closed Loop (Control Mode) Tuning

Step 1: According to Higher Meditation (2021), to tune our feedback controller (P, PI and PID),
the controller is set to Proportional Mode first.

Step 2: We adjust the gain value of the controller until obtaining a sustained oscillation as shown
in Figure 4.1.2.1. A gain value that results in a sustained oscillation is called ultimate gain, Kcu.

Figure 3.1.2.1: Kcu Value to Result in a Sustained Oscillation.

Step 3: Measure the ultimate period, Pu between two peaks.

Figure 3.1.2.2: The Value of Two Peak (Ultimate Period, Pu)

9
Step 4: Using Table 3.1.2 to obtain the KP , 𝛕I and 𝛕D based on the desired controller type.

Table 3.1.2.3: Controller Setting Based on the Continuous Cycling Method of Ziegler and
Nichols (1942)

Note that the value given by the table is followed by the controller transfer function,
1
𝐺𝐶(𝑠) = 𝐾𝑃 + 𝐾𝑃 · 𝛕𝐼𝑠
+ 𝐾𝑃· 𝛕𝑑𝑠

1 𝑁
However, the transfer function given in the simulink is 𝐺𝐶(𝑠) = 𝑃 + 𝐼 𝑠
+𝐷 1 , which N
1+𝑁 𝑠

is the filter coefficient.

Step 5: Obtain value of P, I and D.

By comparing both transfer functions,
P = KP .
1
𝐼 = 𝐾𝑃 · 𝛕𝐼
.

𝑁
𝐷 1 = 𝐾𝑃· 𝛕𝑑𝑠 ,
1+𝑁 𝑠

𝑁𝑠
𝐷 𝑠+𝑁
= 𝐾𝑃· 𝛕𝑑𝑠 ,
(∞)
If N is big enough, lim 𝑠+∞
𝑠 = 𝑠 .Thus we set N = 100.
𝑁→∞

Then equation becomes 𝐷𝑠= 𝐾𝑃· 𝛕𝑑𝑠.

𝐷= 𝐾𝑃· 𝛕𝑑

After obtaining P, I and D values, enter the values into the controller.

10
4.0 Results

4.1 Open Loop

Figure 4.1.1: Open Loop Configuration and Performance

In the open loop which uses a manual controller, the PV from the sensor is manually tuned by
Ziegler Nichols method up to two different set points, 50oC and 80oC. From the graph, it requires
more than 10 seconds for PV to reach the set point if it is tuned manually. Besides, it takes a
longer response time because it requires human’s eyes to observe the PV differs from the set
point, then adjust the gain value manually to achieve the setpoint. In this open loop, the final gain
value obtained is 17.125. Hence, we will not compare the open loop system with other closed
loop systems.

11
4.2 Closed Loop

Closed loop
Based on the Figure 3.1.2.1, Figure 3.1.2.2 and Table 3.1.2

Kcu
Pu

78.2 5.720
Table 4.2.3: Kcu & Pu value took from Figure 3.1.2.1 and Figure 3.1.2.2

Type of controller 𝐾𝑃 𝛕𝐼 𝛕𝑑

P 39.1

PI 35.19 4.766667

PID 46.92 2.86 0.715

Table 4.2.2: Controller Setting Based on the Continuous Cycling Method of Ziegler and Nichols
(1942)

Type of controller 𝑃 𝐼 𝐷

P 39.1

PI 35.19 7.382517

PID 46.92 16.40559 33.5478

Table 4.2.3: Controller Setting in Simulink

12
4.2.1 P controller

(a) Ziegler Nichols Method

(i) Servo Problem:

Figure 4.2.1: Closed Loop P Controller Configuration and P Value with Servo Problem

Figure 4.2.2: Closed Loop P Controller Performance with Servo Problem

13
(ii) Regulatory Problem

Figure 4.2.3: Closed Loop P Controller Configuration and P Value with Regulatory Problem

Figure 4.2.4: Closed Loop P Controller Performance with Regulatory Problem

14
 (b) Autotune
(i) Servo Problem:

Figure 4.2.5: Closed Loop P Controller Configuration and P Value with Servo Problem

Figure 4.2.6: Closed Loop P Controller Performance with Servo Problem

15
(ii) Regulatory Problem

Figure 4.2.7: Closed Loop P Controller Configuration and P Value with Regulatory Problem

Figure 4.2.8: Closed Loop P Controller Performance with Regulatory Problem了

b

16
P controller Ziegler-Nichols tuning Autotuning

Servo When the setpoint is set at 50oC, PV When the setpoint is at 50oC,
shows oscillation and stabilise starting PV also shows oscillation and
from 30s, but does not reach the set stabilise starting from 25s.
point. When the set point is changed to The deviation between PV
80oC, the PV oscillates again but still and setpoint is slightly larger
cannot reach the setpoint. than Ziegler-Nichols tuning
but it is not substantial. When
the set point is changed to
80oC, the PV oscillates again
but still cannot reach the
setpoint.

Regulatory When the setpoint is set at 80oC, a When the setpoint is set at
disturbance is introduced at 100 s, 80oC, a disturbance is
fluctuation of PV occurs. P controllers introduced at 100 s,
are primarily used to lower the fluctuation of PV occurs. The
system's steady state error, with the duration of fluctuation
output of the controllers being directly persists slightly shorter than
proportional to the present error and Ziegler-Nichols tuning but it
has a steady-state error. However, it is not substantial. The
does not meet the setpoint as the P deviation between PV and
controller is unable to eliminate the both setpoints is slightly
error. larger than Ziegler-Nichols
tuning but it is not substantial
also.

Summary Autotune is better in terms of adjusting the gain value automatically to

obtain a sustained oscillation compared to Ziegler-Nichols tuning which
requires manual adjustment. Both results are similar with no big
difference.

17
4.2.2 PI controller

(a) Ziegler Nichols Method

(i) Servo Problem:

Figure 4.2.9: Closed Loop PI Controller Configuration and P, I Value with Servo Problem

Figure 4.2.10: Closed Loop PI Controller Performance with Servo Problem

18
(ii) Regulatory Problem

Figure 4.2.11: Closed Loop PI Controller Configuration and P, I Value with Regulatory Problem

Figure 4.2.12: Closed Loop PI Controller Performance with Regulatory Problem

19
 (b) Autotune
(i) Servo Problem

Figure 4.2.13: Closed Loop PI Controller Configuration and P, I Value with Servo Problem

Figure 4.2.14: Closed Loop PI Controller Performance with Regulatory Problem

20
(ii) Regulatory Problem

Figure 4.2.15: Closed Loop PI Controller Configuration and P, I Value with Regulatory Problem

Figure 4.2.16: Closed Loop PI Controller Performance with Regulatory Problem

21
PI controller Ziegler-Nichols tuning Autotuning

Servo When the setpoint is at 50oC, When the setpoint is at 50oC,

PV oscillates with large PV also shows oscillation and
amplitude, and stabilizes at stabilise starting from 26s.
40s. It can reach the setpoint. The time taken for PV to
When the setpoint is changed achieve stability is shorter
to 80oC, PV requires around than Ziegler-Nichols tuning.
30s to stabilize and meet the Besides, the number of
setpoint. oscillations is lesser and
fluctuation has smaller
amplitude than
Ziegler-Nichols tuning.
Same goes to the setpoint of
80oC.

Regulatory When disturbance is When disturbance is

introduced at100s when the introduced at100s when the
setpoint is set at 80oC, the PV setpoint is set at 80oC, the PV
reacts and stabilizes itself reacts and stabilizes itself
within 50s back to the within 25s back to the
setpoint. The PI controller setpoint. It took a shorter time
provides zero control error to regulate back to the
and is insensitive to setpoint.
interference of the
measurement channel.

Summary It requires a longer time to achieve stability compared to the P

controller but can meet the setpoint. Therefore, this controller
is mostly used in the system where the speed of the system is
not an issue. Overall, the PV line is smoother and reacts faster
by using autotune instead of Ziegler-Nichols tuning.

22
4.2.3 PID controller

(a) Ziegler Nichols Method

(i) Servo Problem

Figure 4.2.17: Closed Loop PID Controller Configuration and P, I, D Value with Servo Problem

Figure 4.2.18: Closed Loop PID Controller Configuration with Servo Problem

23
(ii) Regulatory Problem

Figure 4.2.19: Closed Loop PID Controller Configuration and P, I, D Value with Regulatory
Problem

Figure 4.2.20: Closed Loop PID Controller Configuration with Regulatory Problem

24
 (b) Autotune
(i) Servo Problem

Figure 4.2.21: Closed Loop PID Controller Configuration and P, I, D Value with Regulatory
Problem

Figure 4.2.22: Closed Loop PID Controller Configuration with Regulatory Problem

25
(ii) Regulatory Problem

Figure 4.2.23: Closed Loop PID Controller Configuration and P, I, D Value with Regulatory
Problem

Figure 4.2.24: Closed Loop PID Controller Configuration with Regulatory Problem

26
PID controller Ziegler-Nichols tuning Autotuning

Servo When the setpoint is at 50oC, When the setpoint is at 50oC,

PV oscillates with a smaller PV also shows least
amplitude than the PI fluctuation among P, PI and
controller and stabilizes at PID controllers, and already
around 18s. It can reach the stabilise before 10 s. The time
setpoint. When the setpoint is taken for PV to achieve
changed to 80oC, PV requires stability is shorter than
around 25s to stabilize and Ziegler-Nichols tuning.
meet the setpoint. Same goes to the setpoint of
80oC.

Regulatory When disturbance is When disturbance is

introduced at100s when the introduced at100s when the
setpoint is set at 80oC, the PV setpoint is set at 80oC, the PV
reacts and stabilizes itself reacts and stabilizes itself
within 12.5s back to the within 10s back to the
setpoint. The PV line is setpoint. It took a shorter time
illustrated smoothly with less to regulate back to the
fluctuations and oscillations setpoint, which is shorter than
compared to P and PI Ziegler-Nichols tuning.
controllers.

Summary With the help of Derivative mode, fast response for any
changes involved in the input are able to be made. Among P,
PI and PID controllers, the time taken for PV to achieve
stability is the shortest for PID autotuned controllers. Thus,
the PID controller has the highest sensitivity which is able to
detect and return small fluctuations back to its set point.

27
4.3 Cascade controller

4.3.1 Inner loop tuning

In a cascade controller system, the inner loop has to pose a faster response than the outer
loop, and the inner loop controller can be tuned by using Ziegler–Nichols method. To enhance
the performance of the cascade controller system, we use the PI controller in the inner loop and
PID controller in the outer loop. However, due to Ziegler-Nichols method’s first step, which is
use P controller to determine Kcu and Pu, and P controller has faster response than the PI
controller, we cannot apply Ziegler-Nichols method to tune the outer loop controller. Therefore,
only the Autotune was applied to tune the cascade controller, either inner and outer loop. In
section 4.2.3, we summarized that the PID controller than tuned by the Autotune method has the
best performance. Hence, the Autotune Cascade controller will be compared with the Autotune
PID controller.

Figure 4.3.1: Closed Inner Loop PI Controller Configuration and P, I Value tuned by Autotune

The inner loop of the cascade controller system controls the flow rate based on the flow rate
feedback. After tuning the inner controller, the inner loop is installed into the outer loop as the
actuator of the outer loop. The results of the outer loop sensor and setpoint value are shown in
section 4.3.2. We assume the inner and outer sensor has same transfer function.

28
4.3.2 Outer loop
(i) Servo Problem

Figure 4.3.1: Cascade Controller Configuration and both inner and outer controller values
tuned by Autotune with Servo Problem

Figure 4.3.2: Cascade Controller with Servo Problem

29
(ii) Regulatory

Figure 4.3.3: Cascade Controller Configuration and both inner and outer controller values
tuned by Autotune with Regulatory Problem

Figure 4.3.4: Cascade Controller with Regulatory Problem

]

30
PID controller (autotune) Cascade Controller (autotune)

In the previous discussion, we concluded that Cascade controller is one of the advanced
the PID controller using autotune method controllers with the addition of a set of slave
gives the most effective results to regulate the PID parameters (inner loop) to the same
PV to the setpoint. When the setpoint is at master controller (outer loop) that were used
50oC, PV also shows least fluctuation among in temperature process control. For the inner
P, PI and PID controllers, and already stabilise loop it controls only the flow rate of water.
before 10 s. Inner loop dynamics faster than outer loop
therefore tuning is done for inner loop before
When disturbance is introduced at100s when outer loop.
the setpoint is set at 80oC, the PV reacts and
stabilizes itself within 10s back to the In the servo controller, we can see that the PV
setpoint. reacts slower but steadily to achieve setpoint
compared to the PID controller. It requires
PID controllers are widely used due to their almost 25 seconds for the PV to reach and
simplicity and effectiveness in many control a maintain at the setpoint.
Applications. For simpler processes, a
well-tuned PID controller might be sufficient. For the regulatory controller, when
PID controllers are known for their disturbance is introduced at 100s, the PV
adaptability and can often handle changes in takes around 20s to back to the setpoint after
the process dynamics with relatively simple slight fluctuation occurs.
tuning. In summary, PID control is often
preferred in many applications due to its Cascade control comprises two control loops
simplicity, lower implementation costs, and where the setpoint of one controller is
ease of maintenance. regulated by another. Although this
configuration can enhance disturbance
rejection and performance in specific
situations, it introduces complexity. If the
process has significant time delays or
complex dynamics, cascade control can offer

31
better performance by addressing the
interactions between different loops. The
choice between the PID or cascade depends
on the specific requirements of heating
application and the trade-offs between
performance and cost-effectiveness.

32
5.0 Conclusion

In the preceding discourse, it was deduced that employing the autotune method in a PID
controller yields the most efficacious outcomes for maintaining the Process Variable (PV) in
close proximity to the designated setpoint. Specifically, when the setpoint is established at 50°C,
the PV exhibits minimal oscillation when governed by P, PI, and PID controllers, with
stabilization occurring before the 10-second mark.

Upon the introduction of a disturbance at the 100-second mark, with the setpoint
configured at 80°C, the PV promptly responds and re-establishes equilibrium within 10 seconds,
reverting back to the predefined setpoint. Therefore, our process UV radiation and servo can be
achieved by using an auto tuned PID controller.

The benefits of employing feedback control stem from its ability to gather information at
the output of the process. This characteristic enables the control system to consider unforeseen
disruptions like frictional and pressure losses. The architecture of feedback control guarantees
the attainment of desired performance by promptly adjusting inputs as soon as deviations are
detected, irrespective of the origin of the disturbance. Another merit of feedback control lies in
its capacity to stabilize unstable processes through the analysis of system outputs. Feedback
controls do not require detailed knowledge of the system and, in particular, do not require a
mathematical model of the process. Feedback controls can be easily duplicated from one system
to another (Woolf, n.d).

The widespread adoption of PID controllers can be attributed to their simplicity and
effectiveness across various control applications. For less intricate processes, a well-calibrated
PID controller may prove sufficient. Renowned for their adaptability, PID controllers can adeptly
navigate changes in process dynamics through relatively straightforward tuning. In summation,
PID control is often favored in numerous applications owing to its simplicity, lower
implementation costs, and ease of maintenance.

33
6.0 Reference

Arora, V., Singh, A., et al. (2022). Machine learning and its applications. In Smart Electrical and
Mechanical Systems. Retrieved from
https://www.sciencedirect.com/topics/engineering/industrial-control#:~:text=Industrial%20contr
ol%20system%20(ICS)%20is,or%20automation%20of%20industrial%20processes.

Hahn, J., & Edgar, T. F. (2003). Process Control Systems. In Encyclopedia of Physical Science
and Technology (Third Edition). Retrieved from
https://www.sciencedirect.com/topics/engineering/feedback-controller

Apex Team. (2022, February 24). Rydlyme Specifications. Retrieved from

https://www.apexengineeringproducts.com/rydlyme-specifications/

United States Environmental Protection Agency. (1999, September). Wastewater Technology

Fact Sheet: Ultraviolet Disinfection. https://www3.epa.gov/npdes/pubs/uv.pdf

Woolf, P., et al. (n.d.). Feedback control: What is it? When useful? When not? Common usage.
University of Michigan.
https://eng.libretexts.org/Bookshelves/Industrial_and_Systems_Engineering/Chemical_Process_
Dynamics_and_Controls_(Woolf)/11%3A_Control_Architectures/11.01%3A_Feedback_control-
_What_is_it%3F_When_useful%3F_When_not%3F_Common_usage

Patra, A. (2016). Analysis of Power Transformer using fuzzy expert and neural network system.
Retrieved from
https://www.researchgate.net/profile/Ashis-Patra/publication/327816155_Analysis_of_Power_Tr
ansformer_using_fuzzy_expert_and_neural_network_system/links/5ba5f47c299bf13e60436d23/
Analysis-of-Power-Transformer-using-fuzzy-expert-and-neural-network-system.pdf

34
Higher Meditations. (2021, February 23). Ziegler Nichols PID tuning method using Simulink
[Video file]. Retrieved from https://www.youtube.com/watch?v=MRA-yt22j5I

Seborg, D. E., Mellichamp, D. A., Doyle III, F. J., & Edgar, T. F. (2016). Process Dynamics and
Control Techniques (4th ed.). Publisher.

35