J Electr Eng Technol.

2018; 13(1): 460-467 ISSN(Print) 1975-0102

http://doi.org/10.5370/JEET.2018.13.1.460 ISSN(Online) 2093-7423

Optimal PID Controller Design for DC Motor Speed Control System

with Tracking and Regulating Constrained Optimization
via Cuckoo Search

Deacha Puangdownreong†

Abstract – Metaheuristic optimization approach has become the new framework for control synthesis.
The main purposes of the control design are command (input) tracking and load (disturbance)
regulating. This article proposes an optimal proportional-integral-derivative (PID) controller design for
the DC motor speed control system with tracking and regulating constrained optimization by using the
cuckoo search (CS), one of the most efficient population-based metaheuristic optimization techniques.
The sum-squared error between the referent input and the controlled output is set as the objective
function to be minimized. The rise time, the maximum overshoot, settling time and steady-state error
are set as inequality constraints for tracking purpose, while the regulating time and the maximum
overshoot of load regulation are set as inequality constraints for regulating purpose. Results obtained
by the CS will be compared with those obtained by the conventional design method named Ziegler-
Nichols (Z-N) tuning rules. From simulation results, it was found that the Z-N provides an impractical
PID controller with very high gains, whereas the CS gives an optimal PID controller for DC motor
speed control system satisfying the preset tracking and regulating constraints. In addition, the
simulation results are confirmed by the experimental ones from the DC motor speed control system
developed by analog technology.

Keywords: PID controller, Tracking and regulating, Cuckoo search, Metaheuristics

1. Introduction make those design methods and tuning rules non-general.

Moving toward new era, the intelligent control design
In control context, there are two main purposes of has been transferred from the conventional framework to
control system design [1-7]. The first one is called new paradigm based on modern optimization [14-15].
command tracking (or input following), while the second Finding optimal parameters of PID controller can be
one is called load regulating (or disturbance rejection). In considered as the optimization problem. Modern optimi-
industrial applications, using the proportional-integral- zation using the selected powerful metaheuristic techniques
derivative (PID) controllers for industrial applications was as an optimizer (or solver) has been accepted and applied
first introduced by Minorsky in 1922 [8]. To-date, the PID to PID design optimization, for example of some popular
controller has been a worldwide solution for an effective metaheuristic applications, designing of the PID controller
control and more than half of industrial automatic by genetic algorithm (GA) [16-23], designing of the PID
controllers are of PID type. Not only ease of use and controller by particle swarm optimization (PSO) [24-32]
simple realization, but the PID also can achieve two main and designing of the PID controller by ant colony
purposes of control as mentioned above. Tuning parameters optimization (ACO) [33-36]. However, almost all research
of PID controller is a challenging work. Regarding to works considered only tracking purpose of control system
conventional design method for obtaining appropriate design.
parameters of PID controller, one can proceed with In 2009, the cuckoo search (CS) was firstly proposed by
available analytical design methods or tuning rules. Mostly Yang and Deb [37] as one of the most efficient population-
the analytical design methods assume known plant models based metaheuristic optimization techniques. The CS
[1-4, 8-11] while the tuning rules assume known process algorithm is based on the obligate brood parasitic behavior
responses [9, 12-13]. Those analytical design methods and of some cuckoo species in combination with the Lévy
tuning rules, however, have some particular conditions flight behaviour of some birds and fruit flies. The global
concerning the plant models, such as dead time or transport convergent property of the CS algorithms has been proved
lag, fast and slow poles, real and complex conjugated zeros and reported [38]. By literatures, the performance of the
and poles, as well as unstable poles, etc. These conditions CS was evaluated against many benchmark problems
† Corresponding Author: Dept. of Electrical Engineering, Graduate including unconstraint, constraint, deterministic, stochastic,
School, Southeast Asia University, Thailand. ([email protected]) single-objective and multi-objective [37-40]. It was found
Received: June 3, 2017; Accepted: August 14, 2017

460
Copyright ⓒ The Korean Institute of Electrical Engineers
This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/
licenses/by-nc/3.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
 Deacha Puangdownreong

that the CS is very promising and could outperform

existing popular population-based metaheuristic algorithms
such as GA and PSO [37-40]. Moreover, the CS has been
successfully applied to several real-world engineering
problems such as spring design optimization [39], welded
beam design [39-40], multiple disc brake [40] and optimal
PID controller design [41-43].
By literatures about applications of metaheuristic
techniques to optimal PID controller design as mentioned
earlier, the objective functions (error between the referent
input and the controlled output) were mostly set to be Fig. 1. Schematic diagram of DC motor
minimized with tracking constraint. By those approaches,
the command tracking and the steady-state responses of the
controlled system would be treated, but the disturbance
rejection response of the controlled system would be
unpredictable. The motivation of this work is to propose
the general design approach for optimal PID controller in
which the command tracking and disturbance rejection
Fig. 2. DC motor plant
responses of the controlled system will be simultaneously
considered. In this article, the CS is applied to optimal PID
controller design problem for DC motor speed control of inertia, B is viscous friction, w(t) is motor speed and q(t)
system based on modern optimization approach with is motor position.
tracking and regulating constraints. The sum-squared error Once if(t) is assumed to be constant, the induced torque
between the referent input and the controlled output will be T(t), the armature voltage ea(t) and the back-emf voltage
set as the objective function to be minimized. The rise time, eb(t) can be expressed in (1), (2) and (3), respectively [4,
the maximum overshoot, the settling time and the steady- 44-45], where Kt is torque constant and Kb is back-emf
state error will be performed as constraints for tracking constant. By taking the Laplace transform of (1), (2) and
purpose, while the regulating time and the maximum (3), the s-domain transfer function of a DC motor can be
overshoot of load regulation will be set as constraints for formulated and written as stated in (4).
regulating purpose. Results obtained by the CS will be
compared with those obtained by one of the conventional d w (t )
T (t ) = K t ia (t ) = J + Bw (t ) (1)
design methods named Ziegler-Nichols (Z-N) tuning rules dt
[12]. This article consists of five sections as follows: di (t )
introduction, problem formulation of the PID controller ea (t ) = Ra ia (t ) + La a + eb (t ) (2)
dt
design, CS algorithms, results and discussions and finally
eb (t ) = K bw (t ) (3)
conclusions as appeared in section 1-4 and 5, respectively.
W( s ) Kt
= (4)
Ea ( s ) JLa s 2 + ( BLa + JRa ) s + ( BRa + K t K b )
2. PID Design Problem Formulation
Commonly, using a DC motor needs a power amplifier
This section begins with the DC motor plant model used as a driver. Due to this scheme, the DC motor plant
in this work. Then, the PID control loop and PID optimi- consists of a driver and a DC motor as shown in Fig. 2. The
zation problem with tracking and regulating constraints driver model is approximated by the first-order transfer
will be followed. function as stated in (5), where KA is driver gain and tA is
driver time constant. Therefore, the DC motor plant model
2.1 DC motor plant model can be rewritten as stated (6).

A DC motor is a useful machine transforming electrical Ea ( s ) KA

energy to mechanical energy. It is a well-known device and = (5)
Vin ( s ) (t A s + 1)
widely used in industrial applications as an actuator. The
W( s ) K A Kt
schematic diagram of an armature-controlled DC motor = 2 (6)
can be represented in Fig. 1[44-45], where ea(t) is armature Vin ( s ) (t A s + 1)[ JLa s + ( BLa + JRa ) s + ( BRa + K t K b )]
voltage, ef(t) is field voltage, Ra is armature resistance, La is
armature inductance, Rf is field resistance, Lf is field As a testing rig, the DC motor plant consists of a DC
inductance, ia(t) is armature current, if(t) is field current, motor (LEYBOLD-DIDACTIC GMBH, Type 731-91, 0.3
eb(t) is back-emf voltage, T(t) is motor torque, J is moment kW, 220 V, 2.2 A, 2000 rpm) and a driver (SCR full-wave

http://www.jeet.or.kr │ 461
 Optimal PID Controller Design for DC Motor Speed Control System with Tracking and Regulating Constrained Optimization via Cuckoo~

controlled rectifier) as shown in Fig. 3. A speed sensor controller is represented in Fig. 5, where Gp(s) and Gc(s)
(tacho-generator LEYBOLD, Type 731-09) and a low-pass are the plant and the PID controller, respectively. The
filter are also conducted. The parameters of this motor model in (7) will be used as the plant Gp(s) in Fig. 5. The
plant have been identified and obtained at 1,000 rpm as PID controller receives error signal E(s) and generates
follows: Ra = 54.7280 W, La = 1.5104 H, J = 36.4277 kg- control signal U(s) to control output C(s) and regulate load
m2, B = 0.0988 N-m-sec/rad, Kt = 2.7761 N-m/A, Kb = disturbance D(s) referring to referent input R(s). The time-
1.6046 V/rpm, KA = 3.4449 and tA = 0.3350 sec. Very good domain and s-domain functions of the PID controller are
agreement between actual speed response and the plant stated in (8) and (9), where Kp, Ki and Kd are the
model can be observed in Fig. 4. The DC motor plant in (6) proportional, integral and derivative gains, respectively.
can be written in (7). The closed loop transfer function with PID controller is
given in (10). The main purposes of control are to make the
2.2 PID controller loop C(s) tracking the R(s) and simultaneously regulating the
C(s) whenever the D(s) is applied into the control loop at
A feedback loop of DC motor speed control with PID Tdist as virtualized in Fig. 6.

W( s ) 9.563
= (7)
Vin ( s ) 18.43s + 722.9 s 2 + 1997 s + 9.862
3

de(t )
u (t ) PID = K p e(t ) + K i ò e(t )dt + K d (8)
dt
Ki
Gc ( s ) = K p + + Kd s (9)
s
æ K ö
ç K p + i + K d s ÷ G p (s)
C (s) s
= è ø
(10)
Fig. 3. DC motor plant testing rig R( s) æ Ki ö
1+ ç K p + + K d s ÷ G p (s)
è s ø

Kp D(s)

R(s) E(s) U(s) + C(s)

+
Ki/s Gp(s)
+ + + +
- Plant
Kds
Gc(s)
PID Controller

Fig. 4. Plots of actual speed response and plant model Fig. 5. PID control loop

Fig. 6. Tracking and regulating purposes

462 │ J Electr Eng Technol.2018; 13(1): 460-467

 Deacha Puangdownreong

2.3 PID design optimization problem

The PID controller design optimization for DC motor

speed control by the CS can be represented by the block
diagram in Fig. 7. The sum-squared error between R(s) and
C(s) is set as the objective function fobj expressed in (11).
For command tracking purpose, rise time (tr), maximum
percent overshoot (Mp), settling time (ts) and steady-state
error (ess) are set as the inequality constraints stated in (12).
For load regulating purpose, regulating time (treg) and
maximum percent overshoot of load regulation (Mp_reg) are
set as the inequality constraints also stated in (12). The PID
design problem is to search for optimal parameters Kp, Ki Fig. 7. CS-based PID control design optimization
and Kd in (11). The fobj will be fed back to the CS tuning
block to be minimized in order to find the optimal PID
parameters satisfying to the inequality constraints as stated
in (12). Referring to the inequality constraints expressed in
(12), both tracking and regulating constraints are proposed
together to cover two main purposes of control system
design simultaneously. If the tracking part is considered,
while the regulating is ignored, the command tracking
response of the controlled system will be treated, but the
disturbance rejection response will be unpredictable. On
the other hand, if the regulating part is conducted, while
the tracking is not considered, the disturbance rejection
responses of the controlled system will be treated, but the
command tracking response will not be guaranteed.
Therefore, setting the inequality constraints with tracking
and regulating as shown in (12) can guarantee both
command tracking and disturbance rejection responses of
the controlled system. The values of tr, Mp, ts, ess, treg and
Mp_reg appeared in (12) are set from the preliminary study
and simulation of this system. This is the problem-
dependent of each system of interest.
N
Min f obj ( K p , K i , K d ) = å [r (i ) - c(i )]2 (11)
i =1

Subject to: ü
tr £ 0.2 sec.ü ï
ï
M p £ 10% ïï ï
ý : Tracking ï
ts £ 0.5% ï
ï
ess £ 0.1% ïþ ïï
treg £ 0.5 sec. üï ý (12)
ý: Regulating ï
M p _ reg £ 20% ïþ ï
ï Fig. 8. Flow diagram of CS algorithms
0 £ K p £ 10, ü ï
ï
0 £ K i £ 0.1, ý: Search spaces ï
ï ï metaheuristic optimization techniques. In CS algorithm,
0 £ K d £ 4.0 þ þï each cuckoo lays one egg at a time, and dumps it in a
randomly chosen nest. The best nests with high quality of
eggs (solutions) will carry over to the next generation. The
3. Cuckoo Search Algorithms number of available host nests is fixed, and a host can
discover an alien egg with a probability pa ∈ [0, 1]. The CS
In 2009, Yang and Deb proposed the cuckoo search (CS) algorithms can be represented by the flow diagram as
[37] as one of the most powerful population-based visualized in Fig. 8.

http://www.jeet.or.kr │ 463
 Optimal PID Controller Design for DC Motor Speed Control System with Tracking and Regulating Constrained Optimization via Cuckoo~

New solutions x(t+1) for cuckoo i can be generated by 4. Results and Disscussions
using a Lévy flight as stated in (13). Symbol Lévy(l) in
(13) represents a Lévy flight providing random walk with The CS is applied to design an optimal PID controller
random step drawn from a Lévy distribution having an for DC motor speed control system. Referring to Fig. 7, the
infinite variance with an infinite mean as expressed in (14), CS algorithms were coded by MATLAB running on Intel(R)
where t stands for the time. In another way, the step length Core(TM) i5-2430M, 2.4 GHz, 8 GB-RAM computer.
s of cuckoo flight can be calculated by (15), where u and v The search parameters of the CS are set according to the
are drawn from normal distribution as stated in (16), where recommendations [37-40], i.e. n =20, a =1.0, b =1.5 and
N is denoted as the normal distribution. Standard pa = 0.3. Max_Gen = 100 is set as the TC. The design
deviations of u and v are also expressed in (17), where G is approach runs 100 trials with different initial solutions to
the standard Gamma function. obtain the best solutions (Kp, Ki and Kd ). After the search
process stopped, the convergent rates of the proposed
xi(t +1) = xi(t ) + a Å Lévy(l) (13) objective function in (11) are depicted in Fig. 9. The optimal
-l parameters of the PID controller are successfully obtained as
Lévy » u = t , 1 < l £ 3 (14)
stated in (18).
u
s = 1/ b (15)
|v| 0.03
Gc ( s ) PID _ CS = 3.25 + + 2.66 s (18)
u » N (0. s u ) v » N (0. s v2 )
2
(16) s
1/ b
æ G(1 + b ) sin(pb / 2) ö ü
su = ç , 0 < b £ 2ï For comparison, the second method of Z-N is applied in
( b -1)/ 2 ÷
è G[(1 + b ) / 2]b 2 ø ý (17) this work to design the PID controller for DC motor speed
sv = 1 ï
þ control system. Details of the Z-N design tuning rules are
omitted in this article. Readers can find its more details
From their preliminary parametric studies, Yang and Deb from [12]. The PID controller designed by the second
provided the recommendations for users to set the search method of Z-N is then stated in (19).
parameters of the CS for most optimization problems [37,
38, 39, 40]: the numbers of nests n = 15 to 25, a = 1.0, b = 16, 284.08
Gc ( s ) PID _ ZN = 4,914.53 + + 371.05s (19)
1.5 and the probability fraction pa = 0.25 to 0.5. s
Once the CS is applied to design the PID controller for
DC motor speed control system, its algorithms can be
represented step-by-step to show how the CS find the
optimal parameters Kp, Ki and Kd of the PID controller as
follows.
Step-0 Perform the objective function fobj as (11) with
the inequality constraint functions as (12), the search
spaces of Kp, Ki and Kd as (12), Max_Gen and Gen=1. The
numbers of nests (numbers of cuckoos) = n. In this step,
the initial solutions x* (initial values of Kp, Ki and Kd) are
randomly generated within the given search spaces.
Step-1 If Gen<=Max_Gen, n cuckoos find the new
nests by Lévy flight in (15) associated with formulations
(16) and (17) to lay their eggs in the random nests. Fig. 9. Convergent rates of objective function
Otherwise, report the optimal solution x* (Kp, Ki and Kd).
Step-2 If m (m<=n) cuckoos’ eggs are found by hosts,
m cuckoos have to find the new nests by Lévy flight in (15)
associated with formulations (16) and (17) again to lay
their eggs in the random nests.
Step-3 New solutions (Kp, Ki and Kd) will be calculated
via (13). Then, they are evaluated by the objective function
fobj in (11) and the inequality constraint functions in (12).
Step-4 If the fobj with new solutions is less than the fobj
with old solutions, the solutions x* (Kp, Ki and Kd) will be
updated. Otherwise, the old solutions x* are maintained.
Step-5 Update Gen= Gen+1, and go back to Step-1 to
proceed the next generation.
Fig. 10. Responses without and with PID designed by Z-N

464 │ J Electr Eng Technol.2018; 13(1): 460-467

 Deacha Puangdownreong

Fig. 11. Responses without and with PID designed by CS

Fig. 12. DC motor speed control system testing rig
4.1 Simulation results

The PID controllers for DC motor speed control system

obtained by the second method of Z-N in (19) and the CS
in (18) are conducted for simulation via MATLAB as
visualized in Fig. 10 and Fig. 11, respectively. From Fig.
10, it was found that the PID controller designed by the Z-
N provides very fast and very high overshoot response, i.e.
tr = 0.17 msec., ts = 2.88 msec., Mp = 64.48% and ess =
0.00%. Once considering the controller gains in (19), it
was found that the values of Kp, Ki and Kd are very great
and probably cannot be realized by any hardware. This
leads the PID controller designed by the Z-N impractical
for implementation. Fig. 13. System responses (experimental results)
Referring to Fig. 11, it was found that the PID controller
designed by the CS gives the reasonable response
satisfying the inequality constraints in (12): for tracking, CS are depicted in Fig. 13. The PID controller designed
tr=0.15 msec., ts=0.42 sec., Mp=5.58% and ess = 0.00% and by the Z-N cannot be implemented due to the high values
for regulating, Mp_reg=16.47% and treg=2.36 sec. These of controller gains. From Fig. 13, the system response at
results show the main advantage of the proposed design 1,000 rpm provides tr=1.25 sec., ts = 2.60 sec., Mp = 6.50%,
approach and obtained responses superior to the Mp_reg=18.50%, treg=2.25 sec. and ess=0.00%. Once
conventional design method such as the Z-N design tuning comparing to the simulation results, the system output from
rules. When considering the controller gains in (18), it was experimental results gives slower response. This is because
found that the values of Kp, Ki and Kd are possible to be of the DC generator (load) added into the system. It makes
realized by any hardware. This leads the PID controller the entire time constant of the system greater. The time-
designed by the CS practical for realization and delay occurred in the output response as appeared in Fig.
implementation. 13 is due to the TCA785 trigger circuits and power
electronic devices installed in full-bridge rectifier circuits.
4.2 Experimental results However, when considering the overall system response, it
was found that the experimental results are corresponding
To confirm the simulation results, the DC motor speed to the simulation ones. Moreover, the developed system
control system with PID controller needs to be implemented. can efficiently regulate the load disturbance of 90% of
The DC motor control system is developed as a testing rig motor rated.
by using an analog technology as shown in Fig. 12 to
perform the experimental results. An optimal PID controller
designed by the CS is realized by op-amp LM335 and RC 5. Conclusions
network to ensure real-time operation. Signal conditioning
circuits including low-pass filter and zero-span circuit are Design of an optimal PID controller for the DC motor
also employed. The electrical load is applied for the step speed control system with tracking and regulating
load disturbance input. With this scheme, the DC generator constrained optimization by using the cuckoo search (CS),
(load) is also conducted to generate the output voltage for one of the most powerful population-based metaheuristics,
electrical load. has been proposed in this article. Based on the modern
From testing, the experimental results of the DC motor optimization, the sum-squared error between the referent
speed control system with PID controller designed by the input and the controlled output has been set as the

http://www.jeet.or.kr │ 465
 Optimal PID Controller Design for DC Motor Speed Control System with Tracking and Regulating Constrained Optimization via Cuckoo~

objective function, while the rise time, the maximum [10] J. J. Rubio, P. Cruz, L. A. Paramo, J. A. Meda, D.
overshoot, the settling time and the steady-state error have Mujica and R. S. Ortigoza, “PID anti-vibation control
been set as tracking constraints and the regulating time and of a robotic arm,” IEEE Latin America Transactions,
the maximum overshoot of load regulation have been set as vol. 14, no. 7, pp. 3144-3150, 2016.
regulating constraints. As simulation results compared with [11] P. T. Garran and G. Garcia, “Design of a PID
the Ziegler-Nichols (Z-N) tuning rules, it was found that controller for a coupled tanks system employing
the Z-N provided very high gains of PID controller which ADRC,” IEEE Latin America Transactions, vol. 15,
cannot be realized and implemented, while the CS revealed no. 2, pp. 189-196, 2017.
an optimal PID controller for DC motor speed control [12] J. G. Ziegler and N. B. Nichols, “Optimum settings
system satisfying both tracking and regulating constraints. for automatic controllers,” Trans. ASME, vol. 64, pp.
In this work, the simulation results have been confirmed by 759-768, 1942.
the experimental ones from a testing rig of DC motor speed [13] G. H. Cohen and G. A. Coon, “Theoretical con-
control system developed by analog technology. For the sideration of retarded control,” Trans. ASME, vol. 75,
overall system response, the experimental results are pp. 827-834, 1953.
corresponding to the simulation ones. In future research, [14] V. Zakian, Control Systems Design: A New Frame-
the proposed design approach will be extended to design work, Springer-Verlag, 2005.
the proportional-integral-derivative-accelerated (PIDA) [15] V. Zakian and U. Al-Naib, “Design of dynamical and
controllers for real-world control systems. Moreover, the control systems by the method of inequalities,” in
CS algorithms need some modifications in order to speed Proc. of the IEE International Conference, vol. 120,
up the search process and guarantee the exactly optimal pp. 1421-1427, 1973.
solution found in general case. [16] S. Mehrdad and C. Greg, “An adaptive PID controller
based on genetic algorithm processor,” in Proc. of the
International Conference on Genetic Algorithm in
Acknowledgements Engineering System, pp. 88-93, 1995.
[17] Y. Mitsukura, T. Yamamoto and M. Kaneda, “A
The author would like to thank the anonymous reviewers
design of self-tuning PID controller using a genetic
for their helpful and constructive comments that greatly
algorithm,” in Proc. of the International Conference
contributed to improving the final version of the paper.
on American Control Conference, pp. 1361-1365, 1999.
Also, they would like to thank the Editors for their generous
[18] J. F. Whidborne, “A genetic algorithm approach to
comments and support during the review process.
design finite-precision PID controller structures,” in
Proc. of the International Conference on American
References Control Conference, pp.4338-4342, 1999.
[19] R. Bindu, K. Mini and K. Namboothiripad, “Tuning
[1] C. L. Phillips and R. D. Harbor, Feedback Control of PID controller for DC servo motor using genetic
Systems, Prentice-Hall, 1991. algorithm,” in Proc. of the International Journal of
[2] B. C. Kuo, Automatic control systems, John Wiley & Emerging Technology and Advanced Engineering
Sons, 2003. (IJETAE), vol. 2, no. 3, pp. 310-314, 2012.
[3] R. C. Dorf and R. H. Bishop, Modern Control [20] S. Srikanth and G. R. Chandra, “Modeling and PID
Systems, Prentice-Hall, 2005. control of the brushless DC motor with the help of
[4] K. Ogata, Modern Control Engineering, Prentice- genetic algorithm,” in Proc. of the IEEE International
Hall, 2010. Conference on Advances in Engineering, Science and
[5] J. J. Rubio, “Hybrid controller with observer for the Management (ICAESM-2012), pp. 639-644, 2012.
estimation and rejection of disturbances,” ISA Trans- [21] N. P. Adhikari, M. Choubey and R. Singh, “DC motor
actions, vol. 65, pp.445-455, 2016. control using Ziegler Nichols and genetic algorithm
[6] A-I. Carlos, S-R. Hebertt and S. S-C. Miguel, “A technique,” Int. J. Electr. Electron. Comput. Eng., 2012.
linear active disturbance rejection control for a ball [22] Y. Arturo, J. C. Ren, J. R. Troncoso, L. M. Velazquez
and rigid triangle system,” Mathematical Problems in and R. A. Osornio-Rios, “PID-controller tuning opti-
Engineering, vol. 2016, pp. 1-11, 2016. mization with genetic algorithms in servo systems,”
[7] J. J. Rubio, “Sliding mode control of robotic arms International Journal of Advanced Robotic Systems,
with deadzone,” IET Control, Theory and Applica- vol.10, pp. 324, 2013.
tions, vol. 11, no. 8, pp. 1214-1221, 2017. [23] A. T. El-Deen, A. A. H. Mahmoud and A. R. El-Sawi,
[8] N. Minorsky, “Directional stability of automatically “Optimal PID tuning for DC motor speed controller
steered bodies,” American Society of Naval Engin- based on genetic algorithm,” International Review of
eering, vol. 34, pp. 284, 1922. Automatic Control (IREACO), vol. 8, no. 1, pp. 80-85,
[9] A. Dwyer, Handbook of PI and PID Controller 2015.
Tuning Rules, Imperial College Press, 2003. [24] A. J. Mohammed, “A particle swarm optimization

466 │ J Electr Eng Technol.2018; 13(1): 460-467

 Deacha Puangdownreong

(PSO) based optimum of tuning PID controller for a method based on multi-objective ant colony
separately excited DC motor (SEDM),” Eng. & Tech. optimization and differential evolution to design PID
Journal, vol. 29, no. 16, pp. 3322- 3323, 2011. DC motor speed controller,” International Review on
[25] H. E. A. Ibrahim and M. A. Elnady, “A comparative Modelling and Simulations (IREMOS), vol. 5, no. 2,
study of PID, fuzzy, fuzzy-PID, PSO-PID, PSO-fuzzy, pp. 905-912, 2012.
and PSO-fuzzy-PID controllers for speed control of [37] X. S. Yang and S. Deb, “Cuckoo search via Lévy
DC motor drive,” International Review of Automatic flights,” in Proc. of the World Congress on Nature &
Control (IREACO), vol. 6, no. 4, pp. 393-403, 2013. Biologically Inspired Computing (NaBIC 2009), pp.
[26] A. Jalilvand, A. Kimiyaghalam and A. H. Kord, 210-214, 2009.
“Optimal tuning of PID controller parameters on a [38] X. S. Yang, “Cuckoo search and firefly algorithm:
DC motor based on advanced particle swarm optimi- overview and analysis,” Studied in Computational
zation algorithm,” International Journal on Technical Intelligence: Cuckoo search and Firefly Algorithm
and Physical Problems of Engineering (IJTPE), vol. Theory and Applications, Springer, pp. 1-26, 2014.
3, no. 4, pp. 10-17, 2011. [39] X. S. Yang and S. Deb, “Engineering optimisation
[27] A. Jalilvand, R. Aghmasheh and E. Khalkhali, by cuckoo search,” Mathematical Modelling and
“Optimal design of PID power system stabilizer in Numerical Optimisation, vol. 1, no. 4, pp. 330-343,
multi-machine power system using PSO and genetic 2010.
algorithm,” International Review of Electrical [40] X.S. Yang and S. Deb, “Multiobjective cuckoo search
Engineering (IREE), vol. 6, no. 2, pp. 907-912, 2011. for design optimization,” Computers and Operations
[28] M. I. Solihin, L. F. Tack and M. L. Kean, “Tuning of Research, vol. 40, no. 6, pp. 1616-1624, 2013.
PID controller using particle swarm optimization [41] M. Kishnani, S. Pareek and R. Gupta, “Optimal
(PSO),” in Proc. of the International Conference on tuning of PID controller by cuckoo search via Lévy
Advanced Science, Engineering and Information flights,” in Proc. of the International Conference on
Technology (ICASEIT2011), 2011. Advances in Engineering and Technology Research
[29] S. Bouallegue, J. Haggege, M. Ayadin and M. Benrejeb, (ICAETR), 2014.
“PID-type fuzzy logic controller tuning based on [42] R. Sethi, S. Panda and B. P. Sahoo, “Cuckoo search
particle swarm optimization,” Elsevier journal on algorithm based optimal tuning of PID structured
Engineering Applications of Artificial Intelligence, TCSC controller,” Computational Intelligence in
vol. 25, no. 3, pp. 484-493, 2012. Data Mining, vol. 1, pp. 251-263, 2014.
[30] H. T. Dorrah, A. M. El-Garhy and M. E. El-Shimy, [43] D. Puangdownreong, C. Kiree, D. Kumpanya and S.
“Design of PSO-based optimal fuzzy PID controllers Tunyasrirut, “Application of cuckoo search to design
for the two-coupled distillation column process,” in optimal PID/PI controllers of BLDC motor speed
Proc. of the International Middle East Power Systems control system,” in Proc. of the Global Engineering
Conference (MEPCON2010), 2010. & Applied Science Conference, pp. 99-106, 2015.
[31] M. Nasri, N. P. Hossein and M. Maghfoori, “A PSO- [44] P. C. Krause, Analysis of Electric Machinery, McGraw-
based optimum design of PID controller for a linear Hill Book Company, 1983.
brushless DC motor,” World Academy of Science, [45] P. C. Krause, O. Wasynczuk, S. D. Sudhoff, Analysis
Engineering and Technology, vol. 20, pp. 211-215, of Electric Machinery and Drive Systems, IEEE
2007. Press, Piscataway, NJ, 2002.
[32] C. Kiree, D. Kumpanya, S. Tunyasrirut and D.
Puangdownreong, “PSO-Based Optimal PI(D) Con-
troller Design for Brushless DC Motor,” Journal of Deacha Puangdownreong received
Electrical Engineering & Technology, vol. 11, no. 3, his B.Eng., M.Eng. and Ph.D. degrees,
pp. 715-723, 2016. all in Electrical Engineering, from
[33] M. Ünal, A. V. Topuz and H. Erdal, “Optimization of Southeast Asia University (SAU),
PID controllers using ant colony and genetic algori- Bangkok, Thailand, King Mongkut's
thms,” Springer, 2013. Institute of Technology Ladkrabang
[34] H. E. A. Ibrahim and A. A. H. Mahmoud, “DC motor (KMITL), Bangkok, Thailand and
control using PID controller based on improved ant Suranaree University of Technology
colony algorithm,” International Review of Automatic (SUT), Nakhon Ratchasima, Thailand, in 1993, 1996 and
Control (IREACO), vol. 7, no. 1, pp. 1-6, 2014. 2005, respectively. He is currently an associated professor
[35] M. J. Blondin and P. Sicard, “PID controllers and of electrical engineering of Southeast Asia University,
anti-windup systems tuning using ant colony Bangkok, Thailand. His research interests include control
optimization,” Journal of Power Electronics and synthesis and design, metaheuristics applications to various
Applications (EPE), pp. 1-10, 2013. control engineering problems.
[36] I. Chiha, H. Liouane and N. Liouane, “A hybrid

http://www.jeet.or.kr │ 467