Available online at www.sciencedirect.

com

ScienceDirect
Procedia Technology 11 (2013) 141 – 146

International Conference on Electrical Engineering and Informatics 2013 (ICEEI 2013)

Speed Tracking of Indirect Field Oriented Control Induction Motor

using Neural Network
Azuwien Aida Boharia, Wahyu Mulyo Utomoa, Zainal Alam Harona, , Nooradzianie
Muhd. Zina, Sy Yi Sim a, and Roslina Mat Ariff a
a Faculty of Electrical and Electronic Engineering, Universiti Tun Hussein Onn Malaysia, 86400 Parit Raja, Batu Pahat, Johor, Malaysia

Abstract

Neural network based indirect field oriented control of induction motor drive is presented in this paper. The purpose of this motor is
to investigate the performance of neural network controller which is applied on three phase induction motor drive system. The
introduction of neural network using indirect field oriented control principles for decoupling control of induction motor are
discusses. A decoupling controller and flux estimator were implemented upon this neural network using MATLAB/SIMULINK
software. The training procedure of the neural network, description of the control system is given in this paper. The comparative
results show that the neural network based indirect field oriented control schemes of induction motor drive is more effective and,
hence, found to be a suitable replacement of the conventional proportional integral derivative controller.

Keywords:Induction motor, field oriented control, neural network controller.

1. Introduction

An induction motor (IM) is designed to operate from a three phase source of alternating voltage and it is a type of
synchronous AC motor. The advantages of the induction motor includes high reliability, relatively simple, has rugged
structure, low cost, has a wide speed range, high efficiencies and robustness. These advantages are proven to be a
crucial development of the electrical drives, taking the induction machine as the execution element, for all related
aspects like starting, braking, speed reversal and speed change. The dynamic operation of the induction machine drive
system is a part of the overall performance of the system. The advanced control methods have made the induction
motor for high performance applications possible due to the rapid improvements in power electronics and
microprocessor.
The induction motor control in the field of electric motor drives has attracted a lot of researchers over time[1-14].
To meet the high standards of performance set by dc drives, variable speed ac drives employing induction motors and
the field-oriented control method have been developed in recent years. Decoupling terms between the flux and the
torque is highly advocates to provide fast dynamic response as well as good steady-state performance. High dynamic
performance in induction motor can be achieved by means of field-oriented control where it provides a suitable
mathematical description of three phase induction motor.
The translation of coordinate from the fixed references stator frame to the frame of rotating synchronous is
implied by the vector control [1]. Independent control of the torque and flux of the complex dynamic of the motor
were permitted but this decoupling technique, introduced in early 1970s [2].Thus, a good performance with parameter
insensitive property cannot be achieved without applying more sophisticated control techniques such as adaptive
control [3]. Artificial neural networks attracted a great attention in the field of power electronics due to their inherent
parallelism which allows for high speed processing and permits implementation of real time control application. The
ability to perform in noise environments and the tolerance to faults and to miss data are possessed by the NN’s[4].

a
E-mail address: he110247@siswa.uthm.edu.my

2212-0173 © 2013 The Authors. Published by Elsevier Ltd. Open access under CC BY-NC-ND license.
Selection and peer-review under responsibility of the Faculty of Information Science & Technology, Universiti Kebangsaan Malaysia.
doi:10.1016/j.protcy.2013.12.173
142 Azuwien Aida Bohari et al. / Procedia Technology 11 (2013) 141 – 146

To identify and control nonlinear dynamic systems and nonlinear parameter estimation, the use of Neural
Networks (NN’s) has been proposed [5][6]. An estimation of the stator flux and trained to map the nonlinear behavior
of a rotor flux is performed on the proposed NN’s [7]. High performance control has been the objectives of [8], which
presents robust neural controller and against variation of parameter. Without knowing the exact mathematical model
of the system, the desired mapping between the input and outputs signals of the system can be learned by the ANN’s.
The NN’s are excellent estimators in non-linear system because it does not use the mathematical model of the system
[9-10].
This paper will present a neural network (NN) based indirect field oriented control (IFOC) of induction motor
(IM) drive. The performance of NN controller which is applied in three phase IM drive system will be investigated in
this paper. The theoretical principle, numerical simulation procedures and the results of these methods are discussed
and compared with conventional PI controller.

2. Dynamic Modeling of Induction Motor

The induction motor model has been derived in a number of different reference frames. This makes it easier to fix
the reference frame to a particular motor quantity and adjust the model accordingly. Most of induction motors are the
rotary type with basically a stationary stator and a rotating rotor. The dynamic model of the induction motor is derived
by transforming the three phase quantities into two phase direct and quadrature axes quantities. The mathematical
model in compact form can be given in the stationary reference frame as follows [11].

ªVds º ª Rs  Ls p 0 Lm p 0 º ªids º
« » « 0 « »
«Vqs » « Rs  Ls p 0 Lp »» «iqs » (1)
u
«Vdr » « Lm p Zr Lm Rr  Lr p Zr Lr » «idr »
« » « » « »
«¬Vqr »¼ ¬ Zr Lm Lm p Zr Lr Rr  Lr p ¼ «¬iqr »¼

The induction model in the stator-fixed d-q reference frame is described as:
ௗ
ܸ௦ ൌ ܴ௦ ݅௦ ൅  ௗ௧ ሺߖ௦ ሻ 

ௗ
ܸ௥ ൌ Ͳ ൌ ܴ௥ ݅௥ ൅ ௗ௧ ሺߖ௥ ሻ െ ݆߱௥ ߖ௥ 

ߖ௦ ൌ ‫ܮ‬௦ ݅௦ ൅  ‫ܮ‬௠ 

ߖ௥ ൌ ‫ܮ‬௠ ݅௦ ൅  ‫ܮ‬௥ ݅௥ 

Whereas the mechanical equation is given as follow:

The induction motor stator flux and torque are calculated in the flux and torque calculator as follows:

ߖௗ௦ ൌ ‫׬‬ሺܸௗ௦ െ ܴ௦ ݅ௗ௦ ሻ݀‫ݐ‬ 

ߖ௤௦ ൌ ‫׬‬൫ܸ௤௦ െ ܴ௦ ݅௤௦ ൯݀‫ݐ‬ 

ȁߖ௦ ȁ ൌ  ටߖௗ௦ ଶ ൅ ߖ௤௦ ଶ 

అ೜ೞ
ߠఅೞ ൌ  –ƒିଵ ቀ ቁ 
అ೏ೞ

The electromagnetic torque of the machine can be presented as follow:

ଷ௣
ܶ௘ ൌ ଶ ଶ ൫ߖௗ௦ ݅௤௦ െ ߖ௤௦ ݅ௗ௦ ൯ 


 Azuwien Aida Bohari et al. / Procedia Technology 11 (2013) 141 – 146 143


3. IFOC Control System Description

IFOC is a control procedure for operating the motor that results in fast dynamic response and energy efficient
operation at all speeds. It commutates the motor by calculating voltage and current vectors based on motor current
feedback. It maintains high efficiency over a wide operating range and allows for precise dynamic control of speed
and torque. The IFOC controls the stator currents represented by a space vector. It transforms three-phase stator
currents (A, B, C) into a two-phase time variant system (α,β). A two-coordinate time invariant system (dq) is obtained
from that time variant system. In this structure, the motor flux generating part is d (direct) and a toque generating part
is q (quadrature).

4. Neural Network

Inspired by the successful function of the human brains, the neural network (NN) was developed for solving many
large scale and complex problems. Based on ability to process some information and also to analyze the input and
output simultaneously, it makes NN suitable for dynamic and nonlinear system. The development of the structure and
learning algorithm of the NN is explained as follows [12].

4.1. Proposed NN speed controller

This paper proposed a NN control method of IFOC based on hysteresis to reduce the overshoot and torque ripple.
The NN control is added to the speed controller to produce the torque reference. The block diagram of the proposed
NNIFOC of IM drive is shown in Fig 1.

Fig. 1. Complete block diagram of proposed NNIFOC

4.2. Structure of NNIFOC

To design the neural network control some information about the plant is required. Basically, the numbers of
input and output neuron at each layer are equal to the number of input and output signals of the system respectively.
Further the number of hidden layers and the total neurons is depended on the complexity of the system and the
required training accuracy. To implement search efficiency optimal control of an induction motor drive, a multilayer
perceptrons neural network control is developed. Based on the type of the task to be performed, the structure of the
proposed NNIFOC is as shown in Fig 2.

Fig.2. Block diagram of neural network speed control for induction motor drive
144 Azuwien Aida Bohari et al. / Procedia Technology 11 (2013) 141 – 146

The controller consists of input layer, hidden layer and output layer. Based on number of the neuron in the layers,
the NNFOC is defined as a 1-3-1 network structure. The first neuron of the output layer is used as a torque reference
signal (a21=mf). The connections weight parameter between jth and ith neuron at mth layer is given by wmij, while bias
parameter of this layer at ith neuron is given by bmi. Transfer function of the network at ith neuron in mth layer is
defined by:

S m 1
nim ¦w
j 1
m
ij am
j
1
 bim  

The output function of neuron at mth layer is given by:

 aim f m
(nim ) 

Where f is activation function of the neuron. In this design the activation function of the output layer is unity and for
the hidden layer is a tangent hyperbolic function given by:

2
 f m
(nim ) m  1 
1  e  2 ni

Updating of the connection weight and bias parameters are given by:

wF (k ) 
 wijm (k  1) wijm (k )  D
wwijm

 wF ( k ) 
bim (k  1) bim ( k )  D
wbim

where k is sampling time, α is learning rate, and F performance index function of the network.

5. Simulation Results

A simulation was constructed for the entire system in Fig 1 with the implementation of NN based IFOC scheme
for the three-phase induction motor using Borland C++, and then embedded as S-function in Simulink-Matlab
software. In this paper, the pattern of the reference speed that will be use to verify the effectiveness of the proposed
controller is shown in Fig 3 the parameters for the motor are 240V, 50Hz, Poles=4, Rs=0.3Ω, Rr=0.25Ω, Ls=0.0415H,
Lr=0.0412H, Lm=0.0403H, J=0.1kg-m2.

140
REFERENCE SPEED

120

100
Speed (rad/s)

80

60

40

20

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (s)

Fig. 3. Pattern of the reference speed

 Azuwien Aida Bohari et al. / Procedia Technology 11 (2013) 141 – 146 145

The output response of the system based on the speed reference patern is shown in Fig 4 and Fig 5.
160 50
PID CONTROLLER PID CONTROLLER
NN CONTROLLER NN CONTROLLER
140
45

120

40
100
Speed (rad/s)

Speed (rad/s)
80 35

60
30

40

25
20

0 20
0 0.5 1 1.5 2 2.5 3 0.5 0.505 0.51 0.515 0.52 0.525 0.53 0.535 0.54 0.545 0.55
Time (s) Time (s)

Fig.4. Acceleration ramp response of NN and PID speed controller

100
160
PID CONTROLLER
PID CONTROLLER
99 NN CONTROLLER
NN CONTROLLER
140
98

120
97

100 96
Speed (rad/s)
Speed (rad/s)

80 95

94
60
93
40
92

20
91

0 90
2 2.5 3 3.5 4 4.5 5 3.7 3.705 3.71 3.715 3.72 3.725 3.73 3.735 3.74 3.745 3.75
Time (s) Time (s)

Fig.5. Deceleration ramp response of NN and PID speed controller

Referring to the Fig 4, the acceleration ramp of the system by using both controllers in between time 0 to 2
seconds, shows that NN speed controller follow the shape of speed reference pattern and have a response improved
from the conventional PID controller.
While, in Fig 5 between time 2 to 3 seconds, the speed continue at 130 rad/s. Then, NN speed controller follow the
shape of deceleration ramp at the time of 3 to 5 seconds. The results for the both figure shows the great reduce in
ripple of the transient respose and improved the performance at the same time rather than PID controller.

6. Conclusion

The NN based IFOC of induction motor (IM) drive system has been presented in this paper. The purpose of this
motor is to investigate the performance of NN compared to the PID controller. The comparative results show that NN
based IFOC schemes of IM drive is more effective and, hence, found to be a suitable replacement of the conventional
PI controller. It is concluded that field oriented control by using neural network are useful and able to improve the
performance. The result shows that the performance of the system is improved by reducing the ripple.

Acknowledgement

The author would like to gratitude University Tun Hussein Onn Malaysia for any valuable supports during
conducting this research and in preparing this manuscript.
146 Azuwien Aida Bohari et al. / Procedia Technology 11 (2013) 141 – 146

References
[1] Goyat, S. and Ahuja, R. K.., 2012. “Speed Control of Induction Motor using Vector or Field Oriented Control”, IJAET.
[2] Blaschke, F., 1972. “The principle of field orientation as applied to the new transvector closed-loop control system for rotating-filed
machines,” Siemens. Vol.34, pp. 217-220.
[3] Von Zuben, F.J., Netto, M.L.A., Bim, E., Szajner, J., 1994. "Adaptive vector control of a three-phase induction motor using neural
networks," Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on , vol.6,
no., pp.3750-3755 vol.6.
[4] Ba-Razzouk, A., Cheriti, A., Olivier, G.,1997."A neural networks based field oriented control scheme for induction motors," Industry
Applications Conference, 1997. Thirty-Second IAS Annual Meeting, IAS '97., Conference Record of the 1997 IEEE , vol.2, no., pp.804-811
vol.2.
[5] Wishart, M.T., Harley, R.G., 1995. "Identification and control of induction machines using artificial neural networks," Industry Applications,
IEEE Transactions on , vol.31, no.3, pp.612-619.
[6] Narendra, K.S., Parthasarathy, K., 1990. "Identification and control of dynamical systems using neural networks," Neural Networks, IEEE
Transactions on , vol.1, no.1, pp.4-27.
[7] Ba-Razzouk, A., Cheriti, A., Olivier, G., Sicard, P., 1995 , "Field oriented control of induction motors using neural networks
decouplers," Industrial Electronics, Control, and Instrumentation, 1995., Proceedings of the 1995 IEEE IECON 21st International Conference
on , vol.2, no., pp.1428-1433 vol.2.
[8] Miloudi, A., Draou, A., 2001. "Neural controller design for speed control of an indirect field oriented induction machine drive," Industrial
Electronics Society, 2001. IECON '01. The 27th Annual Conference of the IEEE , vol.2, no., pp.1225-1229 vol.2.
[9] Ohnishi T., Miyazaki H., and Okitsu H., 1988. ” High Efficiency Drive of an Induction Motor by Means of V/F Ratio Control.” IECON
Proceedings 14 annual Conference, vol. 3, pp. 780-785.
[10] Lorenz R.D., Yang S.M., 1992. ”Efficiency-optimized flux trajectories for closed-cycle operation of field-orientation induction machine
drives” IEEE Transactions on Industry Applications, Vol. 28 , pp.574 – 580.
[11] Kumar, Rajesh, Gupta, R.A., Bhangale, S.V., Gothwal, Himanshu; 2007. "Artificial neural network based direct Torque Control of Induction
Motor drives," Information and Communication Technology in Electrical Sciences (ICTES 2007), 2007. ICTES. IET-UK International
Conference on , vol., no., pp.361-367, 20-22.
[12] Yatim A.H.M., and Utomo W.M., 2004. ” On-Line Optimal Control of Variable Speed Compressor Motor Drive Using Neural Control
Model.” PECon Proceedings Conference, pp. 83-87