Indonesian Journal of Electrical Engineering and Computer Science

Vol. 16, No. 3, December 2019, pp. 1221~1228

ISSN: 2502-4752, DOI: 10.11591/ijeecs.v16.i3.pp1221-1228  1221

Speed control of DC motor using conventional and adaptive

PID controllers

Sarah N. Al-Bargothi1, Ghazi M. Qaryouti2, Qazem M. Jaber3

1
Department of Mechatronics Engineering/ Faculty of Engineering, Hashemite University, Jordan
2,3
Department of Mechatronics Engineering/ Faculty of Engineering Technology,
Al Balqa’ Applied University, Jordan

Article Info ABSTRACT

Article history: Proportional Integral Derivative (PID) controllers are extensively used in
practical industries to control the speed of DC Motors. The single weakness
Received Jan 26, 2019 of PID controllers is their sensitivity to variation in parameters and operating
Revised May 5, 2019 conditions; thus, tuning the controller gains to adapt with these variations
Accepted Jun 17, 2019 presents a practical challenge. In this paper, an adaptive mechanism that
utilizes a Recursive Least Square (RLS) algorithm, with rate limiters, is
implemented to perform an online self-adjusting of each of the PID gains in
Keywords: order to achieve Adaptive PID (APID) controller that will accommodate to
system variations. MATLAB/ Simulink software is used to implement and
Adaptive PID (APID) controller simulate APID control of a Chopper-Fed DC motor. A conventional PID
recursive least square (RLS) control system is also designed and simulated to obtain results that can be
DC motor used to judge the performance of the APID controller. Results proved that the
PID controllers APID controller forced the motor speed to track the reference input with
insignificant tracking error, and also managed to attain the motor speed at its
desired value, regardless of the load changes inflected on the motor. This
enhances both transient and steady-state speed responses.
Copyright © 2019 Institute of Advanced Engineering and Science.
All rights reserved.

Corresponding Author:
Ghazi M. Qaryouti,
Department of Mechatronics Engineering,
Faculty of Engineering Technology.
Al Balqa’ Applied University, Amman, Jordan.
Email: [email protected]

1. INTRODUCTION
Many applications use dc motor to benefit from their simple, wide and precise control
characteristics. Such applications include, for example robotic manipulators, steel rolling mills, electric
trains, cranes, electric cars, etc [1-2]. Even the Brushless DC (BLDC) motor has been developed with higher
efficiency in operation than classic DC motor [3]. The most flexible control is obtained by means of
Separately Excited DC Motor (SEDM). The best quality of this motor is that it provides high torque load
sustainable property, and it can be used with batteries and solar cells [4]. DC motors are a good field to study
advanced control algorithms, due to the fact that its theory can be projected on other types of motors [5].
The speed control of dc motor with power electronic systems is obtained generally by changing its
terminal voltage. A PID controller is a good candidate for speed control of dc motors. It is the most common
controller used in industry due to its simplicity and ease of implementation [6]. In addition, the PID
controller is used for controlling the brushless dc motor by designing two controller types Fuzzy logic and PI
controllers [7]. The unknown dc motor parameters could be estimated by experimental data onto armature
current and speed response, or by adapting an adaptive model with reference model created based on
experimental data [8]. In some cases the system parameters are changing during operation, and the PID
controller cannot adjust its own gains to cope with these changes, which will emanate the need to online re-
tune the PID gains, aka adaptive PID [9]. Parameters tuning of the dc motor has been used by different

Journal homepage: http://iaescore.com/journals/index.php/ijeecs

1222  ISSN: 2502-4752

procedure as an example Fractional Order Proportional Integral Derivative [10]. An adaptive controller for a
dc motor is also designed by utilizing Lyapunov-like function methods based on Digital Signal Processing
platform [11].
Proportional Integral Derivative (PID) controllers acquire more than 95% of the controllers in the
industrial process control applications; this is accredited to their robust performance, ease of implementation,
and functional simplicity. The major flaw of the PID controller is its high sensitivity to variation in the motor
parameters and load disturbance. Another disadvantage of such controller is that, it is difficult to tune
PID gains [4, 6].
Adaptive control techniques can be employed to overcome these deficiencies of conventional PID
controller. Adaptive Proportional Integral Derivative (APID) control provides fast speed response and
parameter insensitivity [12]. An adaptation mechanism is combined with the conventional PID control to
auto-tune the controller gain during system operation. The adaptation mechanism adopted in this thesis is
Recursive Least Square (RLS) adaptation algorithm.

2. SEPARATELY EXCITED DC MOTOR AND DRIVE

Voltage controlled speed controller of dc motors was introduced for the first time by Ward Leonard
in 1981 [13] and the field has witnessed a great advancement since. Choppers are used to obtain a controlled
dc voltage from a fixed dc source. The speed of the separately excited DC motor can be controlled by the
armature voltage, 𝑉𝑎 , known as voltage control.

2.1. Separately Excited DC Motor (SEDM)

The equivalent circuit of the motor is presented in Figure 1 [14]. Armature and field winding are
supplied separately, which makes 𝑖𝑓 independent of 𝑖𝑎 . The interaction of field flux and armature current in
the rotor produces torque [4]. When a SEDM is exhilarated by a field current and armature current 𝑖𝑓 and 𝑖𝑎 ,
the motor induces a back emf and a torque to balance the load torque at a particular speed. Motor parameters
and calculations are shown in Tables 1 and 2.

Figure 1. Equivalent Circuit Diagram of SEDM

Table.1 Motor Parameters

Parameter Value Unit
Ra 0.78 Ω
La 0.016 H
Rf 150 Ω
Lf 112.5 H
Laf 1.234 H
Bm 0.01 N.m.s
J 0.05 kg.m2

Table 2. Motor Calculated Parameters

Parameter Value Unit
Ia_rated 13.66 A
Ia_start 307.69 A
If_rated 1 A
Te_rated 20.37 N.m
kt 1.234 1.234 N.m/A
kb V.s/rad

Indonesian J Elec Eng & Comp Sci, Vol. 16, No. 3, December 2019 : 1221 - 1228
Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752  1223

The back electromotive force 𝑒𝑎 is proportional to the motor speed, and the electromagnetic torque
developed in the motor is proportional to the armature current as presented by (1) and (2).

𝑒𝑎 = 𝑘𝑏 𝜔 (1)

𝑇𝑒 = 𝑘𝑡 𝑖𝑎 (2)

Applying Kirchhoff’s Voltage Law (KVL) to the armature circuit:

𝑑𝑖𝑎
𝑣𝑎 = 𝑖𝑎 𝑅𝑎 + 𝐿𝑎 𝑑𝑡
+ 𝑒𝑎 (3)

Applying Newton’s second law to the mechanical shaft:

𝑑𝜔
𝐽 = 𝑇𝑒 − 𝑇𝐿 − 𝐵𝑚 𝜔 − 𝑇𝑓 (4)
𝑑𝑡

Rearranging equations the previous equations and taking the Laplace transformation will yield to the
SEDM block diagram shown in Figure 2.

Figure 2. Block Diagram of SEDM

2.2. Chopper Drive of DC Motors

Chopper is a static power electronic device. It acts as a high speed on/off switch; that connects or
disconnects the load from the dc source 𝑉𝑖𝑛 , which creates a chopped dc voltage at the load terminal 𝑉𝑎 .
Chopper drives are widely used in speed control of SEDM, and can achieve speed ranges above or below the
rated speed of the motor. In this thesis buck chopper as illustrated in the circuit of Figure 3 [15] is used to
control the motor terminal voltage 𝑉𝑎 .

Figure3. Schematic Circuit of Class A Buck Chopper

The chopper DC voltage transfer function, defined as the ratio of the output voltage to the input
voltage, is:
𝑉𝑖𝑛
𝑉𝑎
=𝐷 (5)

This ratio D is called the Duty Cycle of the chopper.

Speed control of DC motor using conventional and adaptive PID controllers (Sarah N. Al-Bargothi)
1224  ISSN: 2502-4752

3. CONTROLLER DESIGN
PID controllers are the most commonly used controller in industrial practice for more than 60 years
[6], they compose 90% of controllers on process control fields [16]. A conventional closed loop PID control
system block diagram is illustrated by Figure 4.

Figure 4. Closed Loop PID Control System Block Diagram

As shown in (6) represents a PID controller where 𝑘𝑝 , 𝑘𝐼 and 𝑘𝑑 are the controller gains,U(s) and
𝐸(𝑠) are the control signal and error signal, respectively.

𝑘𝑖
𝑈(𝑠) = (𝑘𝑝 + 𝑠
+𝑘𝑑 𝑠) 𝐸(𝑠) (6)

With all its praises, conventional PID controllers have some defects such as the difficulty of tuning
the controller gains, and the poor self-adaptability, both of which justifies the need of adaptive control [17].
The difference between a conventional controller and an adaptive controller is that the parameters of the
later, noted as θ, are time variant.
Model reference adaptive control (MRAD) has been a well-developed approach of the adaptive
control [17] (Tao 2003). Its objective is to force the plant to track the response of some given reference
model [18, 6]. A MRAC controller block diagram is presented in Figure 5, it consists of a reference model
Gm, a controller, and an adaptive mechanism. An error is generated whenever the actual output of the system
fails to track the reference model output, this error is called the tracking error et, and it is the difference
between the actual output of the system y(t) and reference model output ym (t):

et (t) = y(t) − ym (t) (7)

Figure 5. MRAC with PID Control Law of Chopper-Fed DC Motor Schematic Diagram

The adaptation mechanism provide the controller with parameters θ(t) at each sampling time,
depending on the values of tet (t), u(t), and desired input r(t). The purpose of the Adaptive algorithm is to
find the controller gains such that 𝑒𝑡 ≈ 0. According to [5] the output 𝑦(𝑡) is almost equal to the reference
model output 𝑦𝑚 (𝑡), that is:

𝑦𝑚 (𝑡) ≈ 𝑦(𝑡) (8)

Which is rewritten as:

Indonesian J Elec Eng & Comp Sci, Vol. 16, No. 3, December 2019 : 1221 - 1228
Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752  1225

𝐺𝑚 (𝑠)𝑟(𝑡) ≈ 𝑟(𝑡) − 𝑒(𝑡) (9)

The main purpose of this algorithm is to estimate the new parameters vector θ(t k ) at time instant t k
by adding a correction vector to the previous parameters estimation vector θ(t k−1 ) at time instant t k−1 [19].
The estimation error in (9) is to be minimized using RLS algorithm, It, recursively and online, estimates θ(t)
by applying the following equations [5]:

𝐾(𝑡𝑘 ) = 𝑃(𝑡𝑘−1 )𝜑(𝑡𝑘 ) × [𝐼 + 𝜑𝑇 (𝑡𝑘 )𝑃(𝑡𝑘−1 )𝜑(𝑡𝑘 )]−1 (10)

𝑃(𝑡𝑘 ) = 𝑃(𝑡𝑘−1 ) − 𝐾(𝑡𝑘 )𝜑𝑇 (𝑡𝑘 )𝑃(𝑡𝑘−1 ) (11)

𝜃(𝑡𝑘 ) = 𝜃(𝑡𝑘−1 ) + 𝐾(𝑡𝑘 )[𝑢(𝑡𝑘 ) − 𝜑𝑇 (𝑡𝑘 )𝜃(𝑡𝑘−1 )] (12)

Where 𝐾(𝑡) is the adaptation gain and 𝑃(𝑡) is the covariant matrix, which is a 3 by 3 matrix given as:

𝑃(𝑡) = 𝛼𝐼 (13)

And 𝛼 is the learning rate of the algorithm.

4. SIMULATION THE CONTROL SYSTEM USING MATALB/ SIMULINK AND RESULTS

In this paper work an APID controller is designed and simulated to control the speed of a chopper-
fed dc motor when the motor load is varied according to a fan load. In order to have a better judgment of the
system performance, the motor with the same loading condition will be controlled with conventional PID
controller. Both PID and APID speed control system are modeled using MATLAB/Simulink as presented in
Figures 6 and 7 respectively.
The MATLAB/Simulink models consist of the following blocks:
a. Chopper-Fed Separately excited dc motor;
b. PID controller;
c. Reference model;
d. RLS adaptation algorithm

Figure 6. MATLAB/Simulink Model of Closed Loop PID Control of Chopper-Fed DC Motor

The motor is subjected to a step input of amplitude 1750 rpm with a load that is proportional to the
square of the motor speed. The speed responses of the motor are illustrated in Figure 8.

Speed control of DC motor using conventional and adaptive PID controllers (Sarah N. Al-Bargothi)
1226  ISSN: 2502-4752

Figure 7. MATLAB/Simulink Model of APID Control of Chopper-Fed DC Motor

The similarity of the APID control system speed and the reference model's is very obvious at first
sight. The APID controller achieves excellent tracking, with only et of 27.9362 rpm. The PID controller, on
the other hand had a higher et of 40.7299 rpm, the superiority of the APID controller tracking is illustrated
by Figure 9.

Figure 8. Speed Responses with Fan-Load

The similarity of the APID control system speed and the reference model's is very obvious at first
sight. The APID controller achieves excellent tracking, with only et of 27.9362 rpm. The PID controller, on
the other hand had a higher et of 40.7299 rpm, the superiority of the APID controller tracking is illustrated
by Figure 9.

Figure 9. Tracking Error

Table 3 presents both control systems response criteria as well as that of the reference model. The
APID scheme outperforms the PID scheme in every aspect. It accomplished the fastest rise and settling times,
the smallest tracking and steady state errors, and a negligible percentage overshoot. The percentage overshoot
of the APID controller is drastically reduced; it is even within the tolerated ess .

Indonesian J Elec Eng & Comp Sci, Vol. 16, No. 3, December 2019 : 1221 - 1228
Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752  1227

Table 3. Comparison of Different Systems with Fan-Load

Time response criteria Control Configuration
Reference Model PID Control APID control
𝑇𝑟 (s) 0.0586 0.0817 0.0555
𝑇𝑠 (s) 0.0913 0.1577 0.1174
𝑃𝑂% 0.7948 0.4034 0.2048
𝑒𝑠𝑠 (rpm) 0 0.4697 0.4697
𝑒𝑡 (rms) --- 40.7299 27.9362

This improvement in performance is due to the fact that the APID controller gains, unlike the PID
controller gains, are not constant, but they change according to the RLS adaptation algorithm to achieve
perfect tracking of the reference model. The change in the adaptive controller gain is captured in Figure 10. It
is clear the controller gains differ from their initial values, each to reach a value that best suit the application.

Figure 10. APID Controller Gains of Chopper-Fed Motor System with Fan-load

5. CONCLUSION
In this paper a model reference APID controller was designed to control the speed of chopper-fed
SEDM, a RLS algorithm with rate limiters was implemented that separately adjust each of the
controller gains. The APID control performance was outstanding. It kept the 𝑃𝑂% of the transient response
less than 0.2048 %, it accomplished fast settling of no more than 0.1577 s, and kept the final value of the
speed within 4 rpm off the desired reference speed. Its ability of tracking the reference speed was
phenomenal.
The adaptation algorithm played a key role in the performance of the controller. The RLS algorithm
updated the values of the controller gain at each time instant, which allowed the controller to adopt different
values of PID gains that adapted the system to changes in the load. It is an online tuning of the controller
more or less, and with a rate limiter at the output of each controller gain, the rate of change in the controller
gains was limited to prevent any sudden change in the controller gains which ensured system stability.
Based on the result of the work of this paper, the APID controller proved its superiority over the
conventional PID controlled systems. The PID controller whilst may perform well enough under constant
loading conditions, did not accomplish as good tracking in the case of variable load as did the
APID controller.

REFERENCES
[1] George, M. (2008). “Speed control of separately excited DC motor.” American journal of applied sciences,
5(3), 227-233.
[2] Pavankumar, S., Krishnaveni, S., Ven, gopal, Y. B., & Babu, Y. K. (2010, November). “A neuro-fuzzy based speed
control of separately excited DC motor.” In Computational Intelligence and Communication Networks (CICN),
2010 International Conference on pp. 93-98. IEEE.
[3] G. G. Raja Sekhar, & Basavaraja Banakar, “ Solar PV fed non-isolated DC-DC converter for BLDC motor drive
with speed control” Indonesian Journal of Electrical Engineering and Computer Science (IJEECS), Vol. 13, No. 1,
Speed control of DC motor using conventional and adaptive PID controllers (Sarah N. Al-Bargothi)
1228  ISSN: 2502-4752

pp. 313-323, January 2019.

[4] Khanke, M. P., & Jain, A. P. S. (2015). “Speed control of Separately Excited DC Motor using various Conventional
Controllers”. International Journal of Engineering Research and Applications, 5(4), 67-72.
[5] Qadir, A. S. (2013). “Electro-mechanical modeling of SEDM (separately excited DC motor) & performance
improvement using different industrial controllers”. Lulu. com.
[6] Fahmy, R. A., Badr, R. I., & Rahman, F. A. (2014). “Adaptive PID controller using RLS for SISO stable and
unstable systems”. Advances in Power Electronics, 2014.
[7] Muhammed A.Ibrahim, et al. “Optimal PID controller of a brushless DC motor using genetic algorithm”
International Journal of Power Electronics and Drive System (IJPEDS), Vol. 10, No. 2, pp. 822-830, June 2019.
[8] Byamakesh Nayak, , & Sangeeta Sahu, “Parameter estimation of DC motor through whale optimization algorithm”
International Journal of Power Electronics and Drive System (IJPEDS), Vol. 10, No. 1, pp. 83-92 , March 2019.
[9] Mansour, T., Konno, A., & Uchiyama, M. (2011). “Neural Network Based Tuning Algorithm for MPID Control.”
In PID Control, Implementation and Tuning. InTech.
[10] Badriyah Ahmed Obaid1, et al. “Resolving of optimal fractional PID controller for DC motor drive based on anti -
windup by invasive weed optimization technique”, Indonesian Journal of Electrical Engineering and Computer
Science (IJEECS), Vol. 15, No. 1, pp. 95-103, July 2019.
[11] Alejandro. Rinc´on , et al. “Controlling a DC Motor through Lypaunov-like Functions and SAB Technique” ,
International Journal of Electrical and Computer Engineering (IJECE), Vol. 8, No. 4, pp. 2180-2198,
August 2018.
[12] Hameed, W. I., & Mohamad, K. A. (2012). “Speed control of separately excited dc motor using fuzzy neural model
reference controller.” International Journal of Instrumentation and Control Systems (IJICS) Vol, 2, 27-39.
[13] Chan, C.C. (1987), Low-cost electronic-controlled variable-speed reluctance motors. IEEE transactions on
industrial electronics, (1), 95-100.
[14] Sattigeri, S., Patil, M., Huded, S., and Malashetti, M., (2016). “Speed Control of DC Motor using Chopper
(MATLAB Simulation)”. (Vol. 3) International Journal on Recent Technologies in Mechanical and Electrical
Engineering (IJRMEE).
[15] Nagarajan, D. R., Sathishkumar, S., Balasubramani, K., Boobalan, C., Naveen, S., & Sridhar, N. (2016). “Chopper
Fed Speed Control of DC Motor Using PI Controller. IOSR-Journal of Electrical and Electronics Engineering
(IOSR-JEEE), 11(3), 65-69.
[16] Kanai, M. M., Nderu, J. N., & Hinga, P. K. (2011). “Adaptive PID DC motor speed controller with parameters
optimized with hybrid optimization strategy.” In Second International Conference on Advances in Engineering and
Technology, pp. 121-127.
[17] Chen, C. (2015). “Self-tuning PID Control of Induction Motor Speed Control System Based on Diagonal Recurrent
Neural Network.” International Journal of Control and Automation, 8(10), 321-334.
[18] Zhu, L. (2008). “Adaptive control of sinusoidal brushless DC motor actuators.”
Michigan Technological University.
[19] Sahputro, S. D., Fadilah, F., Wicaksono, N. A., & Yusivar, F. (2017, July). “Design and implementation of adaptive
PID controller for speed control of DC motor”. In Quality in Research (QiR): International Symposium on
Electrical and Computer Engineering, 2017 15th International Conference on, pp. 179-183. IEEE.

BIOGRAPHIES OF AUTHORS

Sarah Nabhan Al-Bargothi received the B.Eng. degree from The Hashemite University, Zarqa,
Jordan, in 2009, and the M.Sc. degree from Al-Balqa Applied University, Salt, Jordan, in 2018.
She is currently a laboratory supervisor in the Department of Mechatronics Engineering, Faculty
of Engineering, Hashemite University. Email: [email protected]

Ghazi M. Qaryouti received his B.Sc. in electrical Engineering from Damascus university in
1982, and M.S in Mechatronics Engineering from Al-Balqa’ Applied University Jordan in 2006,
and Ph.D. degrees in Mechatronics Engineering from De Montfort University in Leicester, UK
in 2015 His research interests are in Power Electronics and Electric Machine Drives, CNC
Machine, 3D Printer, Energy Efficiency and Management, and Renewable System.
Email : [email protected]

Qazem Jaber received his M.S. and Ph.D. degrees in Electrical Drive and Automation
Engineering from Moscow Power Institute in 1982 and 1986 respectively. Currently Dr. Qazem
Jaber an Professor at Department of Mechatronics Engineering, Al-Balqa’ Applied University
Jordan. His research interests are in Power Electronics and Electric Drives, Energy Efficiency
and Management, and Renewable System. Email : [email protected]

Indonesian J Elec Eng & Comp Sci, Vol. 16, No. 3, December 2019 : 1221 - 1228