Fuzzy-PID vs.

PID Controllers: A Comparative Study on

DC Motor Speed Control

Mr. Tanakorn Prasoetphon Mr. Kasit Selamassakul

Department of Production and Robotics Engineering Department of Production and Robotics Engineering
King Mongkut's University of Technology North Bangkok King Mongkut's University of Technology North Bangkok
Bangkok, Thailand Bangkok, Thailand
[email protected] [email protected]

Amornphun Phunopas Wisanu Jitviriya

Department of Production and Robotics Engineering Department of Production and Robotics Engineering
King Mongkut's University of Technology North Bangkok King Mongkut's University of Technology North Bangkok
Bangkok, Thailand Bangkok, Thailand
[email protected] [email protected]

Abstract - In this study, an educational experimental kit is

introduced to provide a comprehensive comparison of PID and
Fuzzy-PID controllers in the context of DC motor speed control.
The experimental setup incorporates an incremental encoder to
accurately measure motor speed, while a user-friendly graphical
user interface (GUI) enables seamless interaction with the
system. The performance of both controllers is thoroughly
evaluated based on four critical parameters: rise time, overshoot,
steady-state error, and settling time. Additionally, the theoretical
underpinnings of each controller type are discussed, illustrating
the distinct advantages and drawbacks associated with each
approach. The project aims to offer a hands-on learning
experience, deepening students' understanding of these two
prevalent control strategies and empowering them to make
informed decisions when selecting the most suitable controller for Fig. 1. DC motor model
various applications.
Keywords – DC motor speed control, PID controller, Fuzzy- controller, which is designed to operate effectively with
PID controller, Graphical user interface (GU I) nonlinear systems [5]. By employing fuzzy logic, the Fuzzy-
PID controller can determine error and change in error at each
I. INTRODUCTION point in time, dynamically adjusting the Kp, Ki, and Kd values
for improved DC motor control [6]. This adaptive control
Over the past decades, The PID (Proportional-Integral- strategy enables the Fuzzy-PID controller to overcome the
Derivative) Control System, also known as the Linear Control performance limitations of conventional PID controllers when
System, has gained widespread adoption in various industries dealing with a broader range of setpoint values and nonlinear
such as machinery, automotive, and aerospace, where high systems [7].
stability and endurance are critical for system performance [1]. Over the past few years, educational institutions have
This popularity can be attributed to the PID controller's ability utilized PID simulation kits to teach students about closed-
to provide precise and efficient control in linear systems and loop and open-loop control of DC motors [8]. Although these
within specific setpoint value ranges [2]. However, the PID simulation kits offer valuable hands-on experience for learning
control system's effectiveness diminishes when faced with PID control principles, they are limited by their inability to
nonlinear systems and wider setpoint ranges, often resulting in demonstrate the functionality of Fuzzy-PID controllers [9].
overshoot and compromised control performance [3]. This This restriction prevents students from fully understanding the
limitation becomes particularly concerning in applications like capabilities and advantages of Fuzzy-PID controllers
elevator control systems, where overshoot can lead to compared to traditional PID controllers.
passenger injuries or system failures [4]. To address these In light of these limitations, the current project seeks to
shortcomings, researchers developed the Fuzzy-PID develop an experimental kit that enables a comprehensive
comparison of PID and Fuzzy-PID controllers. The proposed
kit will illustrate the distinctions between the two control
strategies and demonstrate how to tune P, I, and D parameters
using a graphical user interface (GUI) for a more in-depth
understanding [10]. By integrating both controller types into
the experimental setup, students will gain a broader
perspective on their respective strengths and weaknesses,
equipping them with the knowledge to select the most
appropriate controller for various real-world applications [11].

II. METHODOLOGY
A. DC Motor Modeling Fig. 2. PID controller block diagram
To design the DC motor model, Figure 1 is taken into
consideration. From Eq.(1), it can be observed that the torque
of the motor (T) is directly related to the current (i) through
the use of a torque constant (K):

T =Ki (1)

In Eq. (2), the voltage generated by the torque (Ea) is

proportional to the angular velocity ( m) and is related to the
constant (K), here θ represents the angular position of the
motor.
Fig.3. Fuzzy PID controller block diagram
dθ
e a=K ❑m =K (2) appropriate control strategies to optimize the motor's
dt
performance.
Here, m represents the angular velocity, and θ represents
the angular position of the motor. Based on Newton's law and B. PID Controller
considering Fig. 1, the equations for the DC motor are The PID (Proportional-Integral-Derivative) controller is a
combined with Kirchhoff's Law, resulting in the following closed-loop control system widely employed in various
expressions: applications due to its versatility and effectiveness. It uses
feedback from a sensor to regulate the actual process variable,
2 ensuring it remains as close as possible to the predetermined
d θ dθ setpoint.
J 2
+b =Ki (3)
dt dt 1) The Proportional controller: The first element of the PID
controller, generates a control signal proportional to the
instantaneous error in the system. This error represents the
di dθ difference between the setpoint and the actual process
L + Ri=V −K (4)
dt dt variable. The Proportional controller amplifies or attenuates
the control signal according to the magnitude of the error, with
In these equations, J is the moment of inertia, b is the a larger error causing a more significant adjustment.
viscous friction coefficient, R is the armature resistance, and L 2) The Integral controller: The second element, works to
is the armature inductance. Eq. (3) represents the mechanical eliminate the steady-state offset between the setpoint and the
part of the system, while Eq. (4) represents the electrical part actual process variable. This controller accumulates the error
of the system. Using Eq. (3) and Eq. (4), the transfer function over time and adjusts the control signal accordingly. By
from the input voltage, V(s), to the output angle, θ, can be continuously correcting for the accumulated error, the Integral
derived. This transfer function, G(s), characterizes the controller helps bring the system to a stable state at the desired
dynamic behavior of the DC motor: setpoint.
3) The Derivative controller: The final element, focuses on
θ(s) K the rate of change of the error to improve the system's
G ( s )= = (5) response to abrupt changes and reduce overshoot. By
V (s) [s { ( R+ Ls ) ( Js +b )+ K 2 }] considering the error's derivative, this controller anticipates
future system behavior and applies the necessary adjustments
In this transfer function, s is the Laplace operator. The preemptively. The Derivative controller provides a damping
transfer function allows for the analysis of the DC motor's
response to various inputs and facilitates the design of
effect, which contributes to overall system stability by
mitigating oscillations and reducing overshoot.
In summary, the PID controller combines the actions of
the Proportional, Integral, and Derivative controllers to create
a robust control strategy. By addressing the instantaneous
error, accumulated error, and rate of change of the error, the
PID controller effectively adapts to various system conditions,
ensuring optimal performance and stability.
C. Fuzzy-PID Controller
The Fuzzy-PID controller is an advanced control strategy
that builds upon the conventional PID controller to better
handle strongly nonlinear systems. The term "Fuzzy" refers to
Fig. 4. Fuzzification membership function
the inherent ambiguity or vagueness present in certain real-life
situations where problems or statements cannot be strictly
classified as true or false. Fuzzy logic provides a spectrum of
values between true and false, allowing for increased
flexibility in determining the most suitable solution to a given
problem.
In this work, the Fuzzy logic controller is employed to
determine two essential parameters: error and the rate of
change in error. By evaluating these parameters, the controller
determines appropriate gain values (Kp, Ki, and Kd) for the
PID controller, enabling effective speed control of the DC
motor. Fuzzy logic controllers consist of four main operations,
each playing a crucial role in the decision-making process [6]:
1) Fuzzification: This initial step involves converting
crisp inputs, which are precise numerical values, into fuzzy
sets that represent ranges of values with varying degrees of Fig.5. Rule base
membership. Fuzzification allows for the incorporation of
imprecise or uncertain information into the control process, as
shown in Fig. 4.
2) Rule base: The rule base is a collection of predefined
rules and IF-THEN conditions, as depicted in Fig. 5. These
rules establish the relationships between input and output
fuzzy sets and serve as the foundation for the controller's
decision-making process. The rules are often derived from
expert knowledge or system observations and are used to Fig.6 Defuzzification membership function
guide the controller's actions.
3) Inference engine: The inference engine plays a critical
role in determining the appropriate fuzzy output sets based on Fuzzy logic to determine error and rate of change in error, the
the input fuzzy sets and the rule base. It evaluates the rules and controller can effectively adjust the PID gains, resulting in
combines the fuzzy inputs accordingly, generating fuzzy improved performance and adaptability for a wide range of
output sets that represent the controller's actions. The nonlinear systems.
inference engine's ability to weigh and aggregate fuzzy rules III. RESULTS
allow for a more nuanced control strategy, catering to complex A. Simulation
or nonlinear systems. In the simulation, MATLAB programming was utilized to
4) Defuzzification: The final operation in the Fuzzy logic generate control systems and subsequently display the output
controller's process is defuzzification, which converts the in time-domain graphs for comparison between PID and
fuzzy output sets obtained from the inference engine into a Fuzzy-PID controllers. The variations in proportional, integral,
crisp value, as illustrated in Figure 6. This crisp value and derivative gains were received by the PID controller from
represents a specific control action, directly applicable to the the Fuzzy logic controller. After simulating the output, both
system under control. Defuzzification methods, such as the controllers were compared based on their desired fundamental
centroid or weighted average methods, enable the translation parameters in the time-domain (time-domain specifications).
of fuzzy output sets into actionable control signals. The simulation results revealed that both controllers exhibited
In summary, the Fuzzy-PID controller leverages the a rapid response without overshooting. The PID controller
strengths of both Fuzzy logic and PID control to deliver a demonstrated slightly faster rise and settling times compared
more versatile and adaptive control strategy. By employing to the Fuzzy-PID controller, while the steady-state error of the
Fuzzy-PID controller was marginally lower than that of the
PID controller as shown in Fig.7.

B. Implement with hardware

In the actual performance, step input was employed to
demonstrate the performance of PID and Fuzzy-PID
controllers at each setpoint speed, as depicted. The results
(Fig. 8 and Fig. 9) showed that both controllers exhibited a
rapid response to control signals. However, some overshooting
occurred with the PID controller at significantly different
setpoint speeds from the tuned PID parameter setpoint speed.
The settling time of the PID controller was marginally faster
than that of the Fuzzy-PID controller, while the steady-state Fig. 7. Comparision of PID and Fuzzy-PID controller in simulation
error of the PID controller was slightly lower than the Fuzzy-
PID controller.

IV. DISCUSSIONS & CONCLUSIONS

. In this paper presented the development and
implementation of a comprehensive experimental kit designed
to teach and demonstrate closed-loop control using PID and
Fuzzy-PID controllers for DC motor speed control. The
equipment effectively showcases the differences between the
two control strategies and provides a hands-on learning
experience for students. With the inclusion of open-loop, P,
and PI controllers, the experimental kit offers a holistic Fig. 8. PID controller result
approach to understanding various control systems. The
simulation results and actual performance tests highlight the
strengths and limitations of both PID and Fuzzy-PID
controllers. Students can use the provided GUI to experiment
with adjusting controller parameters and analyze the system
response using time-domain specifications. This interactive
learning experience facilitates a deeper understanding of
control systems and aids students in selecting the most suitable
controller for a given application. Overall, the PID and Fuzzy-
PID experimental kit is a valuable educational tool for
engineering students, promoting a comprehensive
understanding of closed-loop control systems, and fostering Fig. 9. Fuzzy-PID controller result
the ability to make informed decisions regarding controller
selection for practical applications. Future work may involve
expanding the kit to include additional control strategies or equipment and ensuring the safe and efficient operation of the
exploring the use of different types of motors and actuators to experimental kit.
enrich the learning experience further.
REFERENCES
ACKNOWLEDGMENT [1] Z. Li, Q. Li, and C. Li, "A survey on the development of PID control
technology," IEEE Access, vol. 8, pp. 153456-153468, 2020.
The authors would like to express our sincere gratitude to [2] J. G. Ziegler and N. B. Nichols, "Optimum settings for automatic
our advisors and faculty members for their invaluable controllers," in Trans. ASME, vol. 64, pp. 759-768, 1942.
[3] K. J. Åström and T. Hägglund, "Advanced PID control," ISA-The
guidance, support, and encouragement throughout the course Instrumentation, Systems and Automation Society, 2006.
of this project. Their expertise and insights greatly contributed [4] M. Hasan, M. A. Khan, and A. Z. Khan, "Performance evaluation of
to the success of this work. We are also grateful to our peers elevator control using conventional PID and fuzzy PID controllers," in
and fellow students who provided constructive feedback and 2017 IEEE Calcutta Conference (CALCON), pp. 262-267, IEEE, 2017.
[5] H. X. Li, K. W. E. Cheng, and S. L. Ho, "Fuzzy-PID control systems and
suggestions during the development of the experimental kit. their applications," in 2010 5th IEEE Conference on Industrial Electronics
Their input helped to refine the project and enhance its and Applications, pp. 501-506, IEEE, 2010.
educational value. Additionally, we would like to thank the [6] H. Ying, "Fuzzy control of linear systems: A characterization of the
technical staff at the Faculty of Engineering, KMUTNB stabilization domain," IEEE Transactions on Fuzzy Systems, vol. 6, no. 2,
pp. 219-235, 1998.
University, for their assistance in setting up the laboratory
[7] [7] S. Mondal, S. Das, and A. R. Choudhury, "A comparative study of
fuzzy-PID and conventional PID controllers for speed control of DC
motor drives," in 2016 IEEE 1st International Conference on Power
Electronics, Intelligent Control and Energy Systems (ICPEICES), pp. 1-6,
IEEE, 2016.
[8] D. Wang, Y. Huang, and Q. Wang, "A novel PID simulation kit for DC
motor control education," in 2015 10th Asian Control Conference
(ASCC), pp. 1-6, IEEE, 2015.
[9] R. E. Precup, S. Preitl, and P. Korondi, "Fuzzy controllers with maximum
sensitivity for servosystems," IEEE Transactions on Industrial
Electronics, vol. 48, no. 4, pp. 824-831, 2001.
[10]M. A. Bajodah and A. A. Alsayed, "Design and implementation of a GUI-
based DC motor control system," in 2009 IEEE International Conference
on Control and Automation, pp. 21-26, IEEE, 2009.
[11]T. Özdemir, S. Kılıç, and E. Zergeroğlu, "Teaching fuzzy logic control
using MATLAB for engineering education," in 2011 IEEE Global
Engineering Education Conference (EDUCON), pp. 336-341, IEEE,
2011.