International Journal of Engineering and Advanced Technology (IJEAT)

ISSN: 2249 – 8958, Volume-9 Issue-1, October 2019

Fuzzy Anti-Windup PID Controlled Induction

Motor

K. Premkumar, T. Thamizhselvan, M. Vishnu Priya, S. B. Ron Carter, L. P. Sivakumar

 response of the system is not sufficient. Generally,

Abstract: The Induction motor has been the centre of focus in conventional PID controller is effectively utilized in the ac
many industrial applications. This is because the Induction motor motor electrical drives. The design of conventional PID
has many advantages like high precision and high torque controller is more complex when this controller is used in the
performance. The Induction motor speed control is very essential
for various applications. Conventionally induction motor speed high rating electrical drives and hence increases the cost.
control was done using Proportional Integral controller. This Soft computing techniques are used in the closed loop
project proposes the speed control of IM using Fuzzy- Anti control of electrical system and it is known as advanced
Windup PID controller. Using this controller the saturation effect intelligent control. The intelligent control is playing a vital
on the speed response of the Induction motor is reduced. A vector role today industries control. Nowadays, microprocessor and
speed control Induction motor by Fuzzy-Anti Windup PID
controller is simulated using MATLAB/ SIMULINK. The control
microcontroller has high computation ability and high speed
system parameters of speed of the induction motor with PI of operation and intelligent control techniques can be
controller, Anti Windup PID controller and the proposed implemented using this microprocessor and microcontroller.
controller are measured and compared. From the test results the The high speed, low power and low cost advanced processors
Fuzzy-Anti Windup PID controller outperforms the other such as FPGA, DSP and ASIC IC’s along with power
considered controller performance.
semiconductor switches such as MOSFET, IGBT and GTO
made the advanced control and widely utilized in electrical
Keyword : Anti Windup, Field Oriented Control, Fuzzy Logic
Control, Induction motor, PID controller. motor controls. Intelligent advanced control has better
performance than conventional adaptive controls. In recent
I. INTRODUCTION four decades, Fuzzy logic control, Neural Network control,
neuro fuzzy control, fuzzy neuro control, evolutionary control
An induction or asynchronous motor is example for an AC
such as genetic algorithm, PSO, BAT algorithm, etc are
electric motor. The current through the rotor is responsible for
applied in the electrical motor control [11]-[21].
torque and it is obtained by law of electromagnetic induction
Fuzzy logic and neural networks controls are relatively
from the interaction of magnetic field of the stator winding
dissimilar, and have distinctive ability function in information
with rotor circuit. Separate excitation, mechanical
processing through stipulating numerical relationships among
commutation or self excitation is not required for induction
frequent variables in a complex system, performing mappings
motor like in universal, DC and large synchronous motors.
with measure of ambiguity, control of nonlinear system to a
Rotor of the induction motor can be categorized by wound
measure not possible with classical linear systems. Fuzzy
rotor type induction motor and squirrel-cage type induction
logic is a method to represent human-behavioral judgment
motor. Normally, squirrel cage type induction motor has more
into a control system. A fuzzy controller can be designed to
advantages than wound rotor type induction motor such as
imitate human behavioral thinking, i.e., the course of action
reliable, rugged, and economical and it is most utilized in the
people use to deduce solutions from what they know. Fuzzy
industrials drives. To design and develop the classical
control has been principally useful to the control of system
controller, the mathematical model of the system under
through fuzzy linguistic variables. The structure of Fuzzy
control should be known. The typical method of estimation of
inference system is depicted in Fig. 1.
mathematical model of a system under control is more
Normally the induction motor speed is controlled using the
complex when parameter of the system is varied or any
PI controller, which produces saturation effect. The objective
disturbance occurred by the disturbance factor, the output
of this proposed work is to create a controller that is able to
overcome the tuning problems and to reduce the saturation
Revised Manuscript Received on October 20, 2019. effect of PID and PI controller. The main part of this
* Correspondence Author work presents the novel techniques to enhance the
K.Premkumar*, EEE department, , Rajalakshmi Engineering College, , performance of the Anti Windup PID controller to enhance
Chennai, Tamilnadu, India. Email: [email protected]
T.Thamizhselvan, EEE department, , Rajalakshmi Engineering College, the speed response of the induction motor. The computer
Chennai, India. Email: [email protected] simulations carried out to demonstrate the proposed schemes
M.Vishnu Priya, ECE department, Saveetha School Engineering, have the low saturation effect, and proper tuning of the Anti
Chennai, Tamilnadu, India. Email: [email protected]
S.B.Ron Carter, EEE department, , Rajalakshmi Engineering College, ,
Chennai, Tamilnadu, India. Email: [email protected]
L.P.Sivakumar, EEE department, , Rajalakshmi Engineering College, ,
Chennai, Tamilnadu, India. Email: [email protected]

Published By:
Retrieval Number: A1113109119/2019©BEIESP Blue Eyes Intelligence Engineering
DOI: 10.35940/ijeat.A1113.109119 184 & Sciences Publication
 Fuzzy Anti-Windup PID Controlled Induction Motor

Windup PID controller. More specifically, the objectives governors control and manufacturing process control. In
of the paper are following: To design the Fuzzy logic tuner, recent manufacturing industries is used PID controller in the
Design of Anti Windup PID controller, Performance analysis control loops.
of proposed controller, and Comparative analysis between PI, In proportional control action, steady state error of the control
Anti Windup PID controller. parameter is reduced by increasing proportional gain,
conversely it is produces more oscillation in the control
parameter. In integral control, the output error is completely
eliminated by decreasing integral gain but it affects the
performance at transient state. In derivative control action,
system response damped out completely by increasing
derivative parameter but damping effect decreases with high
value of gain. Some of demerits of the PID Controller are
Fig. 1.Fuzzy Control System given as, difficult to tune the parameter of the PID controller.
It reduces robustness; consequently, it cannot provide optimal
This paper deals about the tuning of the anti windup control for the system. When load varies it becomes unstable,
PID controller using fuzzy tuner for speed control of and gives more overshoot, it produces saturation effect, and
induction motor. Induction motor speed by field oriented Due to the integral controller the response of PID controller
control is elaborated in section 2. The conventional controller produces windup phenomenon.
and proposed controller scheme is explained in section 3. Saturation occurs in actuator due to control action provided
MATLAB simulink model of the conventional and proposed by the controller cross limits of the capacities of the control
system is demonstrated in section 4. Investigation on actuator or control system. The apple of integral controller in
simulation results and effectiveness of the proposed the PID controller integrated with saturation of actuator will
controllers are given in section 5. In section 6, concluding produces detrimental effects. If the higher error signal in the
remarks and scope for future work is presented. Integrator will saturates the actuator, the response loop will be
not active due to saturated output from the actuator even if the
II. INDUCTION MOTOR CONTROL BY FIELD control system response changes. The Integrator control
ORIENTED CONTROL
action become unstable due to integration of the error signal
In field oriented control method, flux of the motor is to very high value and also the output takes a long time to go
controlled thus by decompose the current of the AC motor steady-state. While the error is lastly reduced, the integrator
into component of flux and component of torque. These two may be so high i.e., the integrator is taken more time for come
components are processed independently and finally back to the normal value. This result is known as Integrator
combined to generate the motor phase currents in actual form. windup [6]-[7].
This method provide the solutions for the problem of boost To contract with integrator windup phenomenon, controller
adjustment, effective control of torque and enhance the parameters are intended to neglect the actuator limits and an
transient and stead state performance of motor. Various anti-windup control action system is included in the integral
techniques have been developed for the control of maximum action. There are a number of methods existing to decrease
torque. In [3], stator flux inversely proportionally to the the possessions of integrator windup. The anti windup
angular speed to provide the maximum torque than classical regulator is used to conquer the windup occurrence. Three
techniques. Conversely, in this technique, maximum torque commonly used Anti windup schemes are as follows:
cannot be produces for low speed region. Maximum torque of Saturation feedback anti windup PID regulator can able to
the induction motor is controlled by voltage control method in eliminate the saturation effect. It is depicted in the Fig.2,
the region of field weakening region with taking into account involves an additional feedback loop, decreasing the input to
of current and voltage constraints. But, this method ignores the Integrator by equal level of saturation error percentage.
analysis of the stator resistor [1]-[2]. Instantaneous This is also known as a back calculation method.
maximum-torque generation technique has been proposed Back-calculation only requires tuning one parameter, the time
for control of induction motor. In this method, the rotor flux constant Tt, but it only applies to PID control. Disadvantages
is generated using input current and then torque is of Anti windup controller are follows, Tuning methods are
commutated. At time of rotor flux attain the final steady value, required to initialize the gain value of anti windup PID
the torque component is developed using input current controller, Due to constant gain, and it will affect the control
component current [4]-[5]. system performance of the system. In order to overcome the
drawbacks of Anti windup PID controller, fuzzy based online
III. SPEED CONTROL OF INDUCTION MOTOR tuner used in this work. Now a day, Fuzzy logic controller can
The various control techniques for speed control of Induction be used as an online tuner, because Fuzzy logic incorporates
motor are following, the knowledge of human experts. The proposed Fuzzy Anti
windup PID controller has been explained in the proceeding
A. PID Controller section.
The PID controller is accepted as a universal controller for the
feedback system. PID controller is developed in the year
1940s and became a standard controller in the automatic

Published By:
Retrieval Number: A1113109119/2019©BEIESP Blue Eyes Intelligence Engineering
DOI: 10.35940/ijeat.A1113.109119 185 & Sciences Publication
 International Journal of Engineering and Advanced Technology (IJEAT)
ISSN: 2249 – 8958, Volume-9 Issue-1, October 2019

Fig. 2.Back Calculation Anti Windup PID controller

B. Fuzzy-Anti Windup PID Controller Fig. 4.Membership Function of the Input Error
The Fig.3 shows the block diagram of the Fuzzy Anti
windup PID controller.

Fig. 5.Membership Function of Rate of Change of Error

Fig. 3.Fuzzy Anti windup PID controller

Three coefficients kp, ki,and kd are optimized by using
three fuzzy logic controller. As a result, the three individual
fuzzy based proportional, fuzzy based integral and fuzzy
based derivative controllers are combined to create the final
fuzzy PID controller. The fuzzy PID controller receives two Fig. 6.Membership Function of proportional gain
inputs that are error e (t) and rate of change of error de (t). The
error and rate of change error are distributed between -150 to
150. The Fig.4 depicts the range of membership function of
the input error. The error membership function is distributed
from -150 to +150. The error membership function is divided
into five bell shaped membership functions. The Fig.5 depicts
the membership function of rate of change of error. The rate
of change of error is also distributed as the same way of the
error membership function. The Fig.6 shows the membership
function of the proportional gain. The proportional gain
membership function varies from 0 to 20. The membership Fig. 7.Membership Function of integral gain
function is distributed into five membership functions, i.e,
small, medium, big, very big and very very big. The Fig.7
depicts the membership function of integral gain. The integral
gain is distributed from 0 to 50. The Fig.8 depicts the
membership function of derivative gain. The derivative gain
is distributed from 0 to 2.

Fig. 8.Membership Function of derivative gain

Published By:
Retrieval Number: A1113109119/2019©BEIESP Blue Eyes Intelligence Engineering
DOI: 10.35940/ijeat.A1113.109119 186 & Sciences Publication
 Fuzzy Anti-Windup PID Controlled Induction Motor

The Fig.9 shows the structure of Fuzzy online tuner. in Fig.12, Fig.13 and Fig.14 respectively.

Fig. 12. Simulink model of PI Controller

Fig. 9.Structure of Fuzzy online tuner

IV. SIMULATION MODEL

The Fig.10 depicts the simulink model of open loop control of
Induction motor.

Fig. 13. Simulink model of Anti Windup PID Controller

Fig. 10. Simulink model of open loop control of

Induction motor Fig. 14. Simulink model of Fuzzy Tuned Anti Windup
The DC voltage source is used to provide dc supply to the PID Controller
universal bridge. The Universal Bridge is a Series RC snubber
circuits are connected in parallel with each switch device. The V. SIMULATION MODEL
universal bridge is a converter circuit which can act as both
inverter as well as a rectifier. Here the universal bridge is used The Fig.15 shows the speed and torque response of the
as an inverter. The PWM generator is used to provide a pulse open loop control of Induction motor. The Fig.16 shows the
to the switches in the universal bridge. The three- phase V-I speed and electromagnetic torque response of closed loop
measurement block is used measure the voltage and current of control using PI controller in the Induction motor. The Fig.17
the universal bridge. The asynchronous machine is the three shows the speed and electromagnetic torque response of
phase Induction motor which is used as a load. Induction motor control using Anti windup PID controller.
In the closed loop control the vector control of induction
motor is used to the speed controller. The speed of the
induction motor is taken as the actual speed and the reference
speed is compared and speed error is calculated. The torque is
converted into current and thus produces the pulses for the
universal bridge. In the vector control there is theta
calculation, Id calculation, flux calculation, ABC to dq
calculation, dq to ABC calculation and Iqs calculation Fig. 15. Speed and torque response of the open loop
[8]-[10]. control of Induction motor

Fig. 11. Simulink model of closed control of Induction Fig. 16. Speed and torque response of closed loop
motor control using PI
Simulink model of PI controller, anti windup PID controller
controller and Fuzzy tuned anti windup controller are shown

Published By:
Retrieval Number: A1113109119/2019©BEIESP Blue Eyes Intelligence Engineering
DOI: 10.35940/ijeat.A1113.109119 187 & Sciences Publication
 International Journal of Engineering and Advanced Technology (IJEAT)
ISSN: 2249 – 8958, Volume-9 Issue-1, October 2019

Fig. 19. Response of sudden load variation from no load

to full load with PI controller
Fig. 17. Speed and torque response of Induction motor
control using Anti windup PID controller

.
Fig. 20. Response of sudden load variation from full
load to no load with PI controller

Fig. 18. Speed and torque response of Induction motor

using Fuzzy anti windup PID controller
The Fig.18 shows the rotor speed and electromagnetic
torque of Induction motor when controlled by Fuzzy anti
windup PID controller. The figures show the pulse to the Fig. 21. Response of sudden load variation from no load
universal bridge, vector conversions, three phase voltage and to full load with Anti Windup PID controller
current of the Induction motor output.
Table-I: Comparison of time domain specifications.
Time Domain PI Anti Windup Fuzzy Anti
Specification Controller PID Windup PID

Rise Time 0.0148 0.0145 0.0144

Settling Time 0.0266 0.0243 0.0241

Overshoot 2.1793 1.3025 0

Fig. 22. Response of sudden load variation from full
load to no load with Anti windup PID controller
Undershoot 0.1611 0 0

Peak value 154.1125 150.8707 150

Peak Time 0.0262 0.0254 -
From this comparison Table-I, it is clear that the Fuzzy
Anti windup controller provides the best control system
parameters like rise time, settling time. All other proposed
controllers provide higher overshoot and undershoot. Where Fig. 23. Response of sudden load variation from no load
as in Fuzzy Anti windup PID controller there is no overshoot to full load with Fuzzy tuned Anti Windup PID
and undershoots. Therefore the Fuzzy Anti windup PID controller
controller has the best steady state and transient response.
The sudden load is applied at step time 0.2 sec, and the load
torque applied will be 0 to 20 for no load to full load. Where
the full load torque will be applied at 0.2 sec. The load torque
applied will be 20 to 0 for full load to no load. Where the no
loads torque will be applied at 0.2 sec and corresponding
speed and torque response is depicted in Fig.19, Fig.20, Fig. 24. Response of sudden load variation from full
Fig.21, Fig.22, Fig.23 and Fig.24. The no load to full load load to no load with Fuzzy tuned Anti windup PID
torque, full load to no load torque will applied to the PI controller
controller, Anti windup PID controller and Fuzzy Anti
windup PID controller. This response shows that there is a VI. CONCLUSION
undershoot; it takes some time to come back to the steady
state performance. The three phase Induction Motor speed control has been
performed in MATLAB Simulink environment. The
Induction motor speed has
been controlled with PI
controller, Anti windup

Published By:
Retrieval Number: A1113109119/2019©BEIESP Blue Eyes Intelligence Engineering
DOI: 10.35940/ijeat.A1113.109119 188 & Sciences Publication
 Fuzzy Anti-Windup PID Controlled Induction Motor

controller, Fuzzy Anti windup controller. The simulation on Power, Energy and Control (ICPEC), Sri Rangalatchum Dindigul,
2013, pp. 290-295.
results shows that the Fuzzy Anti windup PID controller has 18. K. Premkumar, and B.V. Manikandan, “Speed control of Brushless DC
clear better performance for providing settling time, motor using bat algorithm optimized Adaptive Neuro-Fuzzy Inference
overshoot, undershoot, peak and peak time in System,” in Applied Soft Computing, vol.32, 2015, pp.403-419.
comparison with PI controller, Anti windup PID controller. 19. K. Premkumar, and B.V. Manikandan, “Fuzzy PID supervised online
ANFIS based speed controller for brushless dc motor,”
Neurocomputing, vol.157, 2015, pp.76-90.
REFERENCES 20. K. Premkumar, and B.V. Manikandan, “Bat algorithm optimized fuzzy
PD based speed controller for brushless direct current motor,” in
1. P.M. Menghal and A. Jaya Laxmi, “Fuzzy Based Real Time Control Engineering Science and Technology, an International Journal, vol. 19,
of Induction Motor Drive,” in Procedia Computer Science, vol. 85, no.2, 2016, pp.818-840.
2016, pp. 228-235. 21. K. Premkumar, and B.V. Manikandan, “Adaptive Neuro-Fuzzy
2. Faa-Jeng Lin and Rong-Jong Wai, "Hybrid control using recurrent Inference System based speed controller for brushless DC motor,” in
fuzzy neural network for linear induction motor servo drive," in IEEE Neurocomputing, vol.138, 2014, pp. 260-270.
Transactions on Fuzzy Systems, vol. 9, no. 1, Feb 2001, pp. 102-115.
3. H. Sathishkumar and S.S. Parthasarathy, “A novel neuro-fuzzy
controller for vector controlled induction motor drive,” Energy AUTHORS PROFILE
Procedia, vol.138, 2017, pp. 698-703.
4. M. Hafeez, M. N. Uddin, N. A. Rahim and H. W. Ping, "Self-Tuned
Dr.K. Premkumar received the B.E. degree from Anna
NFC and Adaptive Torque Hysteresis-Based DTC Scheme for IM
University, Chennai, Tamilnadu, India, in 2005, and
Drive," in IEEE Transactions on Industry Applications, vol. 50, no.
M.E. degree from Anna University, Chennai,
2, March-April 2014, pp. 1410-1420.
Tamilnadu, India, in 2007, all in faculty of electrical
5. Abdelkarim Ammar, Abdelhamid Benakcha, and Amor Bourek,
and electronics engineering. At present, he is doing
“Closed loop torque SVM-DTC based on robust super twisting speed
Ph.D. in Anna University, Chennai, Tamilnadu, India.
controller for induction motor drive with efficiency optimization,” in
Also he is working as an Associate Professor in the Department of Electrical
International Journal of Hydrogen Energy, vol.42, no.28, 2017, pp.
and Electronics Engineering of Rajalakshmi Engineering College, Chennai
17940-17952.
Tamilnadu, India. His current research interests include design of speed and
6. H. Shin and J. Park, "Anti-Windup PID Controller With Integral State
current controller is based on PID controller, fuzzy logic controller, ANFIS
Predictor for Variable-Speed Motor Drives," in IEEE Transactions on
controller and CANFIS controller for the special electrical machines.
Industrial Electronics, vol. 59, no. 3, March 2012, pp. 1509-1516.
7. Bruno Leandro Galvão Costa, Clayton Luiz Graciola, Bruno
Augusto Angélico, Alessandro Goedtel, and Marcelo Favoretto
Dr.Thamizhselvan T, Associate Professor in the
Castoldi, “Metaheuristics optimization applied to PI controllers
Department of Electrical and Electronics Engineering
tuning of a DTC-SVM drive for three-phase induction motors,” in
has 12 years of teaching and research experience
Applied Soft Computing, vol.62, 2018, pp. 776-788.
including 6 years of research experience in Power
8. S. Selvi and S. Gopinath, "Vector control of induction motor drive
Electronics and Renewable Energy Systems. He
using optimized GA technique," in IEEE 9th International
received his B.E. from Meenakshi college of
Conference on Intelligent Systems and Control (ISCO), Coimbatore,
Engineering in 2005 and M.E in Power Electronics & Drives from SSN
2015, pp. 1-7.
Engineering college in 2007 and Ph.D from Anna University, Chennai in
9. S. Chacko, C. N. Bhende, S. Jain and R. K. Nema, "A novel on line
2018.He has published 12 research publications in international journals, 10
rotor resistance estimation technique using EA tuned fuzzy
papers in the proceedings of refereed international and national conferences
controller for vector controlled induction motor drive," in IEEE
with 2 best paper awards.
International Conference on Electrical, Computer and
Communication Technologies (ICECCT), Coimbatore, 2015, pp.
M.Vishnu Priya B is an Assistant Professor in the
1-11.
faculty of Electronics and Communication Engineering
10. M. N. Uddin, Z. R. Huang and A. B. M. S. Hossain, "Development
in Saveetha School of Engineering, Chennai Tamilnadu,
and Implementation of a Simplified Self-Tuned Neuro–Fuzzy-Based
India. she got her Masters and Bachelors degree from
IM Drive," in IEEE Transactions on Industry Applications, vol. 50,
Anna University, India. Her research area includes solar
no. 1, Jan.-Feb. 2014, pp. 51-59.
energy, power electronics, soft computing, and electrical machines. She is
11. K. Premkumar and B.V. Manikandan , “Stability and Performance
currently pursuing her Ph.D in the area of multilevel inverter at Saveetha
Analysis of ANFIS Tuned PID Based Speed Controller for Brushless
University.
DC Motor,” in Current Signal Transduction Therapy, vol.13, no.1,
2018, pp. 19-30.
Ron Carter S B is an Assistant Professor in the faculty
12. Premkumar, Kamaraj, Manikandan, Bairavan Veerayan, Kumar, and
of Electrical & Electronics Engineering in Rajalakshmi
Chellappan Agees, “Antlion Algorithm Optimized Fuzzy PID
Engineering College, India. He got his Masters and
Supervised On-line Recurrent Fuzzy Neural Network Based Controller
Bachelors degree from Anna University, India. His
for Brushless DC Motor,” Electric Power Components and Systems,
research area includes renewable energy, power
vol.45, no.20, 2018, pp.2304-2317.
converters, Control Engineering and ANN. Also, he has
13. K. Premkumar and B.V. Manikandan , “GA-PSO optimized online
done consultancy projects in the relevant area.
ANFIS based speed controller for Brushless DC motor,” in Journal of
Intelligent & Fuzzy Systems, vol. 28, no. 6, 2015, pp. 2839-2850.
Dr.P.Sivakumar was born in Tiruchirapalli, India on
14. K. Premkumar and B.V. Manikandan , “Online Fuzzy Supervised
July 3, 1975. He did his U.G. and P.G. at Regional
Learning of Radial Basis Function Neural Network Based Speed
Engineering College, Trichy, India and SASTRA
Controller for Brushless DC Motor,” in Lecture Notes in Electrical
University, Tamil Nadu, India. He is presently working as
Engineering, vol.326, 2015, pp.1397-1405.
an Associate Professor at Rajalakshmi Engineering
15. K. Premkumar and B.V. Manikandan, “Novel bacterial foraging-based
College, Chennai, India. His current research interests
ANFIS for speed control of matrix converter-fed industrial BLDC
include dispersed power generators based on PV and wind power
motors operated under low speed and high torque,” in Neural
generations, DG sourced power system optimization and power electronics
Computing and Applications, vol. 29, no.12, June
control.
2018, pp.1411–1434.
16. M. John Prabu, P. Poongodi, and K. Premkumar, “Fuzzy supervised
online coactive neuro-fuzzy inference system-based rotor position
control of brushless DC motor,” in IET Power Electronics, vol.9,
no.11, September 2016, pp.2229 – 2239.
17. K. Premkumar and B. V. Manikandan, "Adaptive fuzzy logic speed
controller for brushless DC motor," in 2013 International Conference

Published By:
Retrieval Number: A1113109119/2019©BEIESP Blue Eyes Intelligence Engineering
DOI: 10.35940/ijeat.A1113.109119 189 & Sciences Publication