2 ADesignofIntelligentControllerforDirectCurrentMotorvol 04 Issueno 04 Page519-536

International Journal of Advances in Engineering and Management (IJAEM)
Volume 4, Issue 4 Apr 2022, pp: 519-536 www.ijaem.net ISSN: 2395-5252
A Design of Intelligent Controller for Direct

Current Motor
Moumita Paul1, Samsul Islam2, Md. Abdul Hannan Mia2 and
Mohammad Zoynal Abedin2
1
Department of Physics, Jahangirnagar University, Savar, Dhaka-1342, Bangladesh.
2
Department of Mechanical Engineering, Dhaka University of Engineering & Technology, Gazipur-1707,
Bangladesh.
---------------------------------------------------------------------------------------------------------------------------------------
Submitted: 01-04-2022 Revised: 12-04-2022 Accepted: 15-04-2022
-----------------------------------------------------------------------------------------------------------------------------------
ABSTRACT movement of motors in elevators where different
A Direct Current motor is commonly employed for types of DC motors are used. The speed of the DC
many industrial applications due to its high torque motor can be adjusted to a great extent to provide
and efficiency. This article expresses an optimally controllability ease and high performance [1, 2].
designed controller of Direct Current motor speed This paper aims to design a position controller of a
control depending on the genetic algorithm (GA). DC motor by selecting PID parameters using a
The optimization method is used for searching for genetic algorithm. The model of a DC motor is
the ideal Proportional Integral Derivative (PID) considered a second-order system. And this paper
factors. The DC motor controller design methods compares two kinds of tuning methods of
include three kinds: trial and error PID design, auto- parameters for PID controllers. One is the controller
tuning PID design, and GA-based controller design. design by the genetic algorithm, second is the
A GA-PID controller is designed to enhance the controller design by the Ziegler and Nichols method.
system performance using GA. PID controller It was found that the proposed PID parameters
coefficients (𝐾𝑝, 𝐾𝑖 𝑎𝑛𝑑 𝐾𝑑) are calculated by GA to adjustment by the genetic algorithm is better than
produce optimal PID as a hybrid PID with a GA the Ziegler & Nichols’ method. The proposed
controller. The suggested controller GA-PID is method could be applied to the higher-order system
planned, modeled and simulated by the MATLAB also. PID Controller was first published officially by
software program. A comparison output system Minor sky in the early 19th century with his theory
performance is monitored for every controller of three-term control. It is the most used controller
scheme. The results display the time characteristics in the industrial and robotics world until now [3, 4].
performance of GA-PID controller-based. Some advantages of PID Controller are easy to
understand, easy to be implemented, and able to
Keywords: Intelligent Controller, Direct Current provide good system stability [5]. Many systems can
(DC), Genetic Algorithm, PID Controller, be stabilized, from non-linear to linear systems [6].
MATLAB. The tuning parameter of PID is important because it
can guarantee controller efficiency [7].
1 INTRODUCTION Due to its excellent speed control
The direct current (DC) motor, converts characteristics, the DC motor has been widely used
electrical energy into mechanical energy. DC motors in industry even though its maintenance costs are
are the simplest type of motor and are used in higher than the induction motor. As a result, DC
household appliances, such as electric razors, and motor has attracted considerable research and
electric windows in cars. Motor speed control keeps several methods have evolved. Proportional Integral
the motor's rotation at the preset speed and drives a Derivative (PID) controllers have been widely used
system at the demanded speed. In a mechanical for speed and position control of DC motors. This
system, speed varies with several tasks so speed paper endeavors to design a system using a Genetic
control is necessary to do mechanical work properly Algorithm. Genetic Algorithm or in short GA is a
and it makes the motor operate easily. The speed stochastic algorithm based on principles of natural
control mechanism is applicable in many cases like selection and genetics. Genetic Algorithms are a
controlling the movement of robotic vehicles, stochastic global search method that mimics the
movement of motors in paper mills, and the process of natural evolution. Using genetic
algorithms to perform the tuning of the controller
DOI: 10.35629/5252-0404519536 Impact Factor value 7.429 | ISO 9001: 2008 Certified Journal Page 519
will result in the optimum controller being evaluated systems can work in highly corrosive environments
for the system every time. The objective of this such as nuclear power stations and highly aggressive
paper is to show that by employing the GA method environments such as corrosive chemicals and
of tuning a system, optimization can be achieved. gases. DC series motor is suitable for both high and
This can be seen by comparing the result of the GA low power drives, for fixed and variable speed
optimized system against the classically tuned electric drives. Because of its high starting torque,
system. this motor uses in cheap toys and automotive
The algorithm itself has many variations applications such as Cranes, Air compressors, Lifts,
due to which natural phenomena in the world. Some Elevators, winching systems, Electric traction, Hair
of them are Genetic Algorithm (GA) [7, 8], Particle drier, Vacuum cleaner and speed regulation
Swarm Optimization (PSO) [9], and Differential application, power tools, Sewing machines, Electric
Evolution (DE) [10, 11]. They are the most footing. The benefits of using a controller are
commonly used algorithm in Intelligent Search electrical protection of the motor and subsequently,
Algorithm. To test the algorithm, it is applied to the the mechanics, maintain a constant speed, even
PID Controller of the DC Motor System. The when loads are changing, dynamic response to
performance of the system along with several changing system demands, even in a braking
iterations and iteration times will be observed and condition with four-quadrant drive, monitoring to
analyzed. DC Motor is chosen due to its being evaluate machine performance, energy-saving and
commonly used in the industrial [10] and robotics accurate speed control.
world. Electric motors were developed in the late
1800s. By 1887, the electric motor was installed on 1.1 Objectives
the first electric trolley system in Richmond, a) To develop a robust control platform with
Virginia. By 1892, the electric elevator and control optimization methods.
system was invented. Thomas Edison promoted DC b) To improve the performance of DC motor speed
power and built electric power transmission systems control/ position control.
to power DC motors. 1 hp to 100,000 hp electric c) To add adaptation/ intelligence in our control
motors are available to meet any industrial need. architecture.
Asynchronous and synchronous motors and
generators are available to suit specialized functions. 1.2 Methodology
The NEMA Frame is especially suited for The step response is the reference signal,
challenging environments. Also, the motors can be DC motor which is a plant, scope where we see the
vertical with high torque availability for additional output response and this is controller where we will
power. Of course, they can operate on AC or DC see the performance by applying different
power. approaches. Firstly, we apply the PID controller.
The work has been done on the analysis of PID controller is a combination of proportional,
genetic algorithm rules and membership function integral and derivative actions. PID Controller is the
parameters [12]. In 1960, Prof. Holland introduced most common control algorithm used in industrial
genetic algorithms [13, 14]. Genetic algorithms are automation. PID controllers show poor control
applied to search for the globally optimal solution to performances for an integrating process and a large
problems [15]. The evolution process of genetic time delay process “Figure 1”. then apply intelligent
algorithms is based on natural selection. Examples controller which is a class of control techniques that
of applications include industrial fans, blowers, use various artificial intelligence computing
pumps, machine tools, power tools, turbines, approaches like neural networks, fuzzy logic,
compressors, alternators, ships, rolling mills, paper machine learning and genetic algorithms.
mills, movers, and other special applications. Some
Figure 1: Methodology of design of intelligent controller for direct current (dc) motor.
2 INTELLIGENT CONTROLLERS derivative terms must be individually adjusted or

Intelligent control is a class of control "tuned." Based on the difference between these
techniques that use various artificial intelligence values a correction factor is calculated and applied
computing approaches like neural networks, to the input. For example, if an oven is cooler than
Bayesian probability, fuzzy logic, machine learning, required, the heat will be increased “Figure 2”. Here
reinforcement learning, evolutionary computation are the three steps:
and genetic algorithms. Proportional tuning (Kp) involves correcting a
target proportional to the difference. Thus, the target
2.1 Proportional Integral Derivative (PID) value is never achieved because as the difference
Controller approaches zero, so too does the applied correction.
Intelligent PID controllers, are PID Integral tuning (Ki) attempts to remedy this by
controllers where the unknown parts of the plant, effectively cumulating the error result from the "P"
which might be highly nonlinear and/or time- action to increase the correction factor. For example,
varying, are taken into account without any if the oven remained below temperature, “I” would
modeling procedure. Our main tool is an online act to increase the head delivered. However, rather
numerical differentiator, which is based on easily than stop heating when the target is reached, "I"
implementable fast estimation and identification attempts to drive the cumulative error to zero,
techniques. Several numerical experiments resulting in an overshoot.
demonstrate the efficiency of our method when Derivative tuning (Kd) attempts to minimize this
compared to more classic PID regulators. overshoot by slowing the correction factor applied
The working principle behind a PID as the target is approached.
controller is that the proportional, integral and
Figure 2: Proportional integral derivative (PID) controller system.
The controller attempts to minimize the 2.2 Genetic Algorithm (GA) Controller
error by adjusting the process control input. The PID A genetic algorithm is an evolutionary
controller calculation (algorithm) involves three optimization technique. This optimization is
constant parameters called the proportional (P), inspired by genetic evolution in the animal world. It
integral (I) and derivative (D) values, these values can minimize or maximize an objective function
can be interpreted in terms of time. P depends on the given “Figure 3”.
present error, I on the accumulation of past error, It has four steps in the process:
and D is a prediction of future error, based on the (1) Selection.
current rate of change. The weighted sum of these (2) Crossover.
three actions is used to adjust the process via a (3) Mutation.
control element such as the position of a control (4) Replace population with better fitness.
valve, or power supplied to a heating element. .
Figure 3: Genetic algorithm (GA) controller flow chart.
3 DIRECT CURRENT (DC) MOTOR In a DC motor, the input electrical energy is the
A DC motor is an electrical machine that direct current which is transformed into the
converts electrical energy into mechanical energy. mechanical rotation “Figure 4”.
Figure 4: Parts of a direct current (DC) motor [16].
3.1 Types of Direct Current (DC) Motor (3) Compound wound type.
There are different types of DC motors The compound type can be classified into
available. They are listed below: two types namely long shunt as well as the short
(1) PMDC (Permanent Magnet DC Motor) shunt. The long shunt is further classified as
type. cumulative and differential types. Similarly, the
(2) Separately excited. short shunt types are further classified as cumulative
(3) Self-excited. and differential types. The separately excited and
The classification of self-excited can be done like series types are commonly preferred for industrial
the following. purposes “Figure 5”. The DC motors classification
(1) Shunt-wound. figure is shown below.
(2) Series wound.
Figure 5: Types of direct current (DC) motor.
3.2 Direct Current (DC) Motor Controller 4 CONTROLLER DESIGN

DC motor controller manipulates the 4.1 Direct Current (DC) Motor: Proportional
position, speed, or torque of a DC-powered motor Integral Derivative (PID)
and easily reverses, so the DC runs in the opposite Direct-current motors find applications in
direction. Enjoy higher starting torque, quick electrical equipment, computer peripherals, robotic
starting and stopping, reversing, variable speeds manipulators etc. due to their excellent speed control
with voltage input and more. characteristics. Therefore, the speed control of DC
There are two major types of DC motor - the motor attracts the attention of researchers to date as
common DC motor with brushes, and the brushless a notable task [17]. For many years, the
DC motor. Both types have a non-moving source of conventional proportional-integral-derivative (PID)
magnetic fields, known as a stator, and a rotating controllers had been used as a control strategy for
source of magnetic fields, known as the rotor. various industrial processes and motor control
The intended use of a motor controller is to manage applications. The steps for obtaining the optimized
the performance of an electrical motor. Irrespective PID gains using the LQR technique are presented in
of the motor type, this electronic device can fulfill [18]. Long time and effort are required to tune
the following functions: controller parameters using the Ziegler Nichols
(1) Start/stop the motor. frequency response method. Ziegler Nichols
(2) Change the rotation direction. considers the system in an oscillation mode to
(3) Control the speed and torque. realize the tuning procedure, which is not physically
(4) Provide overload protection. realizable [19].
(5) Prevent electrical faults.
Figure 6: Conventional proportional integral derivative (PID)control structure.
Ki The choice of the gains of PID is a challenge that has

C(s) = Kp + + Kd × s (i)
s
been solved using different tuning rules for PID
PID is a very simple but very effective controller.
control.
It has three gains 𝐾𝑃 proportional gain, 𝐾𝑖integral
The gain parameters can be defined by conventional
gain, 𝐾𝑑derivative gain.
tuning rules e.g.
• Ziegler Nichols (ZN).
• Modified ZN (MZN). system, they select the value of kp, ki, kd. If the
• Cohen Coon (CC). model of the system is wrong the value of k p, ki, kd
• Chien Hrones Reswick (CHR) etc. will be wrong “Figure 6”. Therefore, these systems
are not always effective.
There are also many more tuning rules to Without depending on the reaction curve of the
determine the Kp, Ki, Kd values of PID. But these system and use modern optimization techniques that
rules have some limitations. They assumed the will automatically work with system errors and
system to be a first-order pulse dead time model. automatically update the gain. Then this problem
Depending on the reaction curve of the open-loop can be fixed.
Figure 7: Genetic algorithm based on proportional integral derivative (PID).
In this block here added optimization block cause minimizes errors. Which is on the basis of
which is an additional block. A genetic algorithm objective function (TTSE).
can optimize a given error function to find the best First order plus Dead time (FOLDT) model:
suitable sets of gains for the PID controller. This Km e−STm
Gm (s) = (ii)
1+sTm
Table1: Conventional tuning rule.

Rule Name Kp Ti Td
ZN T 2L 0.5L
1.2
L
Modified ZN T 1.4T 0.47T
0.95
L
4.2 Proportional Integral Derivative (PID) performance table is the lowest. Integral squared
+Genetic Algorithm (GA) error (ISE) and integral absolute.
Genetic algorithm is the major category of Genetic algorithms can optimize a given error
evolutionary algorithms (EA) that creates function to find the best suitable sets of gains for the
resolutions to optimization difficulties by retaining PID controller. Objective Functions/ performance
methods stimulated by usual developments such as; indices used here is:
selection, inheritance, mutation and crossing. GA is Integral of Time times the Squared Error, ITSE =
∞
operated to decide on optimizing the rates of PID ∫0 te(t)2 dt. (iii)
controller gains for the selected BLDC motor It gives robustness to the controller as it does not
system. The controller is proposed to reduce the depend on the system model. Genetic algorithm is
error decrease of the output response with regard to an evolutionary optimization technique. This
the orientation signal. The error is computed equally optimization is inspired by genetic evolution in
in the performance table, achieved by reducing the animal world. It can minimize or maximize any
error among the unit step of input response. The objective function given.
physiognomies for the wanted performance are
typically stated in relation to time field amounts. The 4.3 Motor Models
performance of the control system is measured Here DC motor model transfer function
equally to the optimum “Figure 7”. If the parameters that has been collected from various references.
values of the PID controller are selected the
0.01
Motor Model 1:t ∫ 1 (s) =
0.005s2 +0.06S+0.1001 0.067
0.00113 s2 +0.0078854 S+0.0171
0.02
0.008s2 +0.12S+0.004
These are the transfer function of our four motor
Motor Model 3: t ∫ 3 (s) =
0.972 models.
0.001488 s2 +0.2832 S+0.78341
4.4 Step Response and Comparison between ZN vs GA+PID
Figure 8: Comparison between ZN vs GA+PID for motor model-1.
Figure 10: Comparison between ZN vs. GA+PID for motor model-3.
Figure 11: Comparison between ZN vs. GA+PID for motor model-4.
5 RESULTS AND DISCUSSION of the wellness of the best arrangement with era,
5.1 Proportional Integral Derivative (PID)+ where best arrangement is characterized as the
Ziegler Nichols (ZN) Vs. Proportional Integral one which gives least rise time, settling time,
Derivative (PID) + Genetic Algorithm (GA) zero overshoot and almost zero consistent zero
• The extent of Kp, Ki and Kd are picked between steady state error.
(0-100) separately.
0.01
• Estimations of Kp, Ki and Kd are plotted through For Motor Model 1: t ∫ 1 (s) =
0.005s2 +0.06S+0.1001
the objective function demonstrates the variety
Figure 12: Open loop step response for motor model-1.
5.2 Comparison between PID+ZN vs. GA+PID
Table2: Step response for motor model-1.

Step response Motor 1
ZN+PID GA+ PID
Settling Time 0.8520 0.323
Overshoot 43.55% 0.00 %
In this graph, the system is getting stable after almost 2.5s and it takes 2.5s to settle. In this comparison
if added a PID controller structure with this, then see a close loop response comparison.
Here blue curve is PID controller and red curve is GA+PID controller (proposed controller).
0.02
For Motor Model 2: t ∫ 2 (s) = 2
0.008s +0.12S+0.004
Table 3: Step response for motor model-2.

ZN+PID GA+ PID
Rise Time 0.0396 0.0185
Overshoot 48.9026 17.603
In this graph, the system is unstable and if added a PID controller structure with this, then see a close loop response
comparison.
0.972
0.001488 s +0.2832 S+0.78341

ZN+PID GA+ PID
Over shoot 41.7377 0.0693
S-S Error 24.8% 0.00%
Bandwidth 2.8004 Hz 129.176 Hz
In this graph, getting output is 1 but the model is giving 1.2 which means there is an error in the system and it
takes time to settle 1.5s.
0.067
0.00113 s +0.0078854 S+0.0171

ZN + PID GA + PID
Rise Time 0.0965 0.0060
Overshoot 44.9641 0.00
S-S Error 78.4% 0.00%
Bandwidth 2.9107 Hz 362.905 Hz
In this graph, the systems getting output is 1 but model is giving almost 4 which mean there is a huge error in this
system and it takes time to settle almost 1.4s.
5.5 Bode Diagram for Motor Model One
Figure 20: Bode diagram for motor model-1.
5.6 Bode Diagram for Motor Model Two
5.7 Bode Diagram for Motor Model Three

5.8 Bode Diagram for Motor Model Four
5.9 Step Response for Motor Model One, Two, Three and Four
Table 6: Step response for motor model 1, 2, 3 and 4.

Step Motor 1 Motor 2 Motor 3 Motor 4
response
ZN+ GA+PID ZN+ PID GA+PID ZN+ GA+PID ZN+ GA+P
PID PID PID ID
Rise 0.0679 0.207 0.0396 0.0185 0.0055 0.0166 0.0965 0.0060
Time
Settling 0.8520 0.323 0.3315 0.4952 0.0572 0.0267 1.4338 0.0108
Time
Overshoot 43.558 0.00 48.9026 17.603 41.738 0.0693 44.965 0.00
S-S Error 0.00% 0.00% 0.00% 0.00% 24.8% 0.00% 78.4% 0.00%
Bandwidth 25.304 2.911 Hz 116.253 116.253 2.800 129.176 2.9107 362.90
Hz Hz Hz Hz Hz Hz 6
Hz
NB: Green color numbers indicates improvement in system response.
There are four motor models used in this close loop response comparison. For motor model
study and they are motor models one, two, three and three, getting output is 1 but the model is giving 1.2
four. For motor model one, the system is getting which means there is an error in the system and it
stable after almost 2.5s and it takes 2.5s to settle. In takes time to settle to 1.5s. For motor model four,
this comparison, if added a PID controller structure the system getting output is 1 but the model is giving
with this, then see a close loop response comparison. almost 4 which means there is a huge error in this
For motor model two, the system is unstable and if system and it takes time to settle to almost 1.4s.
added a PID controller structure with this, then see a
Table 7: Observation of previous experimental work from the research paper.

References Working on Observation
[6] PID Controller The author gets the best value of PID Controller for controlling the
speed of DC motor is proportional gain 𝐾𝑝 = 20, integral gain 𝐾𝑖 =
2.57, and derivative gain 𝐾𝑑 = 0.0235 with steady-state error =
0.53017, rise time= 0.0527, settling time = 0.097, overshoot = 0 and
the modification in the objective function will obtain the best and
consistent solution.
[17] Intelligent Control The author’s work on the subject highlights the idea, creation,
development and implementation of intelligent control and finally,
the results considering the application and development for the
purpose are presented exposed:
• To within the controller, a value is entered within a range of
1600 RPM to start control of engine speed.
• Achieving values of error of 3%, which is acceptable in a
practical manner.
[20] Intelligent Control The author’s work on the subject highlights the idea, creation,
development and implementation of intelligent control and finally,
the results considering the application and development for the
purpose are presented exposed:
• To the speed of the motor is slowed down only for about 270
rpm (9%) in 980 milliseconds under the effect of full load.
• The motor speed is hunting about 200 rpm (6.66%) in 900
milliseconds on unloading conditions.
[21] Genetic Algorithm The results obtained from the simulations clearly show the genetic
algorithm is used for integral absolute error, integral time-weighted
absolute error, integral square error, and integral time-weighted
square error. It is ascertained that; the proposed controller works very
well in all operating conditions than the other considered controllers.
[22] PID Controller The author gets the best value of PID Controller for controlling the
speed of DC motor is proportional gain 𝐾𝑝 = 5.0196, integral gain
𝐾𝑖 =80.8051, and derivative gain 𝐾𝑑 =0.0549 with steady-state error
=0.0763, rise time= 0.0527, settling time =0.116, overshoot = 0 and
the modification in the objective function will obtain the best and
consistent solution.
[24] DC Motors The results of experiments prove that the approach has lots of good
performances in response speed, control accuracy, adaptability and
robustness. The modification in the objective function will obtain the
best and most consistent solution.
The future works of this control system can be 6 CONCLUSIONS

concluded: None of the four models open-loop
• Apply GA with other controller structures such as responses is good. So, if don’t use the proposed
MPC, LQR, LQG, and different non-linear controller but will not be able to give good
controllers. performance. The designed PID with GA has a much
• Apply another optimization method such as faster response than the response of the classical
Particle swarm optimization (PSO), gray wolf method. The classical method is good for giving us
optimization (GWO) etc. the starting point of what are the PID values.
• To work on different systems other than motor However, the GA-designed PID is much better in
such as different power electronics circuit control terms of the rise time and the settling time than the
such as buck-boost converter, PWM inverter, conventional method. Finally, the genetic algorithm
microgrid etc. provides much better results compared to the
conventional methods. And also, the error
associated with the genetic-based PID is much lesser [6]. Ma’arif, A., Nabila, H., & Wahyunggoro, O.
than the error calculated in the conventional scheme. (2019, November). Application of Intelligent
There are four motor models used in the study. For Search Algorithms in Proportional-Integral-
model one, the system is getting stable after almost Derivative Control of Direct-Current Motor
2.5s and it takes 2.5s to settle. In this comparison, if System. In Journal of Physics: Conference
added a PID controller structure with this, then see a Series (Vol. 1373, No. 1, p. 012039). IOP
close loop response comparison. For model two, the Publishing.
system is unstable and if added a PID controller [7]. Vishal V, Kumar V, Rana K.P.S and Mishra
structure with this, then see a close loop response P. 2014 IEEE International Advance
comparison. For model three, getting output is 1 but Computing Conference (IACC) 1342–1347
the model is giving 1.2 which means there is an error [8]. Wati D.A.R and Hidayat R. 2013
in the system and it takes time to settle to 1.5s. For International Conference on Robotics,
model four, the system getting output is 1 but the Biomimetics, Intelligent Computational
model is giving almost 4 which means there is a Systems (Indonesia) 30–34.
huge error in this system and it takes time to settle [9]. Kumar P, Chatterjee S, Shah D, Saha U. K,
to almost 1.4s. In this paper, the implementation of and Chatterjee S. 2017 Cogent Engineering.
the genetic algorithm-based PID controller for the 4 1-20.
DC motor position control system is covered. [10]. Wati D.A.R 2013 Proceeding of IEEE
Adding a genetic algorithm (GA) with PID has International Conference on Computational
improved system performance significantly more Intelligence and Cybernetics
than the conventional tuning rules of PID. It has (CYBERNETICSCOM) (Indonesia) 40–44.
added robustness to the system model. It has added [11]. Idir A, Kidouche M, Bensafia Y, Khettab K,
adaptation and intelligence to the system as the and Tadjer S. A 2018 International Journal of
controller itself can decide which gains values can Intelligent Engineering and Systems 20 241-
act best for system performance. The controller is 249.
highly customizable as the objective function can be [12]. Yen J., Langari R., Fuzzy logic: Intelligence,
programmed/ set from the user end. Control, and Information, Prentice-Hall,
1999.
ACKNOWLEDGMENT [13]. Swarup K.S., Yamashiro S., Unit
The authors would like to express their commitment solution methodology using
gratitude to Md. Hassanul Karim Roni, Lecturer, genetic algorithm, IEEE Trans: Power
Department of EEE, BAUST, Saidpur, Bangladesh System, vol. 17, pp. 87–91, 2002.
for his help. This study was funded by Hazi Md. [14]. Gen M., Cheng R., Genetic Algorithms and
Shajahan Bhuiyan Foundation (HMSBF). Engineering Design, John Wiley and Sons,
INC, 1997.
REFERENCES [15]. Allaoua, Boumediene, Abdellah Laoufi,
[1]. Hénao H., Capolino G.A., Méthodologie et Brahim Gasbaoui, Abdelfatah Nasri, and
application du diagnostic pours les systems Abdessalam Abderrahmani. "Intelligent
électriques, Article invité dans Revue de controller design for dc motor speed control
l'Electricité et de l'Electronique (REE), n°6, based on fuzzy logic-genetic algorithms
pp.79-86. (Text in french), June 2002. optimization." Leonardo Journal of Sciences
[2]. Raghavan S., Digital control for speed and 13 (2008): 90-102.
position of a DC motor, MS Thesis, Texas [16]. Byjus.com/physics/dc-motor/ [Accessed on
A&M University, Kingsville, August 2005. 05 March 2022]
[3]. Gunawan S.A, Yuwono Y.C.H, Pratama [17]. Pal, P., Dey, R., Biswas, R. K., & Bhakta, S.
G.N.P, Cahyadi A.I and Winduratna B. 2018 (2015). Optimal PID controller design for
4th International Conference on Science and speed control of a separately excited DC
Technology (ICST) (Indonesia) 1–6. motor: A fireflybased optimization approach.
[4]. Sirisantisamrid K, Wongvanich N, Gulpanich International Journal of Soft Computing,
S, and Tammarugwattana N, 2018 Mathematics and Control (IJSCMC), 4(4).
Proceedings of the International Multi [18]. Anderson, B.D.O. & Moore, J.B (1989)
Conference of Engineers and Computer “Optimal control: Linear quadratic methods”,
Scientists (Hong Kong) 14-16. Prentice –Hall International, Inc., Englewood
[5]. El-Deen A.T, Mahmoud A.A.H, and El-Sawi Cliffs, NJ.
A.R. 2015 Int. Rev. Autom. Control 8 81–82. [19]. Soni, Ritu. Singh, D.B.V. Pandey, Pramod. &
Sharma, Priyanka (2013) “Simulation of
optimal speed control for a DC motor using Information Technology (FIT) (pp. 19-23).
conventional PID controller and Fuzzy logic IEEE.
controller”, International Journal of [29]. Khather, S. I., Almaged, M., & Abdullah, A.
information and computation Technology, I. (2018). Fractional order based on genetic
Vol. 3, No. 3, pp 181-188. algorithm PID controller for controlling the
[20]. Khan, M. R., Khan, A. A., & Ghazali, U. speed of DC motors. International Journal of
(2015). Speed control of DC motor under Engineering & Technology, 7(4), 5386-5392.
varying load using PID controller. [30]. Yusoff, T. A. F. K., Atan, M. F., Rahman, N.
International Journal of Engineering (IJE), A., Salleh, S. F., & Wahab, N. A. (2015).
9(3), 38-48. Optimization of PID tuning using genetic
[21]. Perng, J. W., & Hsieh, S. C. (2019). Design algorithm. Journal of Applied Science &
of digital PID control systems based on Process Engineering, 2(2), 97-106.
sensitivity analysis and genetic algorithms.
International Journal of Control, Automation
and Systems, 17(7), 1838-1846.
[22]. Suman, S. K., & Giri, V. K. (2016, March).
Speed control of DC motor using
optimization techniques based PID
Controller. In 2016 IEEE International
Conference on Engineering and Technology
(ICETECH) (pp. 581-587). IEEE.
[23]. Xia, C., Guo, P., Shi, T., & Wang, M. (2004,
August). Speed control of brushless DC
motor using genetic algorithm-based fuzzy
controller. In Proceeding of the 2004
International Conference on Intelligent
Mechatronics and Automation, Chengdu,
China, 3rd edn. A Treatise on Electricity and
Magnetism (Vol. 2, pp. 68-73).
[24]. Wang, Y., Xia, C., Zhang, M., & Liu, D.
(2007, May). Adaptive speed control for
brushless DC motors based on genetic
algorithm and RBF neural network. In 2007
IEEE International Conference on Control
and Automation (pp. 1219-1222). IEEE.
[25]. Jaradat, M. A., Awad, M. I., & El-
Khasawneh, B. S. (2012, April). Genetic-
fuzzy sliding mode controller for a DC
servomotor system. In 2012 8th International
Symposium on Mechatronics and its
Applications (pp. 1-6). IEEE.
[26]. Flores-Morán, E., Yánez-Pazmiño, W.,
&Barzola-Monteses, J. (2018, May). Genetic
algorithm and fuzzy self-tuning PID for DC
motor position controllers. In 2018 19th
International Carpathian Control Conference
(ICCC) (pp. 162-168). IEEE.
[27]. Abdulameer, Ashwaq, et al. "Tuning methods
of PID controller for DC motor speed control.
"Indonesian Journal of Electrical Engineering
and Computer Science 3.2 (2016): 343-349
[28]. Ulasyar, A., Zad, H. S., & Zohaib, A. (2018,
December). Intelligent speed controller
design for brushless DC motor. In 2018
International Conference on Frontiers of

2 ADesignofIntelligentControllerforDirectCurrentMotorvol 04 Issueno 04 Page519-536

Uploaded by

Copyright:

Available Formats

2 ADesignofIntelligentControllerforDirectCurrentMotorvol 04 Issueno 04 Page519-536

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2 ADesignofIntelligentControllerforDirectCurrentMotorvol 04 Issueno 04 Page519-536

Uploaded by

Copyright:

Available Formats

International Journal of Advances in Engineering and Management (IJAEM)

Volume 4, Issue 4 Apr 2022, pp: 519-536 www.ijaem.net ISSN: 2395-5252

A Design of Intelligent Controller for Direct

2 INTELLIGENT CONTROLLERS derivative terms must be individually adjusted or

Figure 2: Proportional integral derivative (PID) controller system.

Figure 3: Genetic algorithm (GA) controller flow chart.

Figure 4: Parts of a direct current (DC) motor [16].

Figure 5: Types of direct current (DC) motor.

3.2 Direct Current (DC) Motor Controller 4 CONTROLLER DESIGN

Figure 6: Conventional proportional integral derivative (PID)control structure.

Ki The choice of the gains of PID is a challenge that has

Figure 7: Genetic algorithm based on proportional integral derivative (PID).

Table1: Conventional tuning rule.

4.4 Step Response and Comparison between ZN vs GA+PID

Figure 8: Comparison between ZN vs GA+PID for motor model-1.

Figure 9: Comparison between ZN vs GA+PID for motor model-2.

Figure 10: Comparison between ZN vs. GA+PID for motor model-3.

Figure 11: Comparison between ZN vs. GA+PID for motor model-4.

Figure 12: Open loop step response for motor model-1.

5.2 Comparison between PID+ZN vs. GA+PID

Figure 13: Comparison between ZN vs GA+PID for motor model-1.

Table2: Step response for motor model-1.

Figure 14: Open loop step response for motor model-2.

5.3 Comparison between PID+ZN vs. GA+PID

Figure 15: Comparison between ZN vs GA+PID for motor model-2.

Table 3: Step response for motor model-2.

Figure 16: Open loop step response for motor model-3.

5.4 Comparison between PID+ZN vs. GA+PID

Figure 17: Comparison between ZN vs GA+PID for motor model-3.

Table 4: Step response for motor model-3.

Figure 18: Open loop step response for motor model-4.

Figure 19: Comparison between ZN vs GA+PID for motor model-4.

Table 5: Step response for motor model-4.

5.5 Bode Diagram for Motor Model One

Figure 20: Bode diagram for motor model-1.

5.6 Bode Diagram for Motor Model Two

Figure 21: Bode diagram for motor model-2.

5.7 Bode Diagram for Motor Model Three

Figure 22: Bode diagram for motor model-3.

5.8 Bode Diagram for Motor Model Four

Figure 23: Bode diagram for motor model-4.

Table 6: Step response for motor model 1, 2, 3 and 4.

Table 7: Observation of previous experimental work from the research paper.

The future works of this control system can be 6 CONCLUSIONS

You might also like