IEEE-32331

Dynamic Modeling, Simulation & Analysis of

Induction Motor Drives
P. M. Menghal Dr. A. Jaya Laxmi
Faculty of Degree Engineering, Professor, Dept. of EEE& Coordinator Centre for Energy
Military College of Electronics & Mechanical Engineering, Studies,Jawaharlal NehruTechnological University, College
Secunderabad, 500 015, Telangana. of Engineering, Kukatpally, Hyderabad-500085,
Research Scholar, EEE Dept., JNTU, Anantapur-515002, Telangana, India
Andhra Pradesh, India [email protected]
prashant_menghal@ yahoo.co.in

Abstract—The induction motor plays very important role in vqs , vds q-axis and d-axis components of stator voltage.
industrial sectors, primarily due to its robustness and lowcost. vqr , vqr q -axis and d-axis components of rotor voltage.
When the mechanical load is applied to induction motor which p Number of poles.
requires speed control, some of the drive and controlstrategies ș The angular position of the rotor.
are based on the estimated axis speed of the motor. The speed
measurement directly reduces its stability and increases the cost
Ȧa Reference frame rotating speed.
of drive implementation. This paper proposes an alternative J Moment of inertia (kg/m2).
methodology for estimating the speed of a three phase induction Te Electrical torque.
motor driven by a voltage source inverter, using space vector Tl Load torque.
modulation under the scalar control strategy and based on e (k) Control error.
artificial intelligent controller such as fuzzy, neural and adaptive r (k) Reference signal.
neuro fuzzy system. To validate the performance of the proposed y (k) Output signal.
method under motor load torque and speedreference set point ǻe(k) Changed error.
variations, simulation results are presented.The simulation ǻu(Ri) The crisp ǻu value corresponding to the maximum
results validates under no load, constant load as well as variable
load of the proposed methodand confirms ability to use
membership degree.
induction motor speed control inpractice.The dynamic modeling, II. INTRODUCTION
simulation and analysis of induction motor using Conventional,
PI, Fuzzy Controller, ANN and Adaptive neuro fuzzy controller
in open loop and closed loop have been presented.
T hree phase squirrel cage induction motors play very
important role in industry because of their lowcost, high
efficiency, high robustness, reliability and low maintenance.
Keywords-Volatge Frequncy Control (V/F), Space Vector PWM Nevertheless, they do not have constant speed characteristics
(SPWM), Propotional Integral (PI) Controller, Fuzzy Logic when mechanical loads are changed from no load to full load
Controller(FLC),Artificial Neural Network(ANN), Adaptive Neuro- and do not have capability of variable speed. These short
Fuzzy controller(ANFIS) comings are resolved by using adjustable speed drives. There
are many ways to control the induction motor such as scalar
I. NOMENCLATURE control, vector control and direct control etc[1-4].The scalar
Rs The stator resistance. Volts /Hertz (V/F) control is simple control structure. It uses
Rr The rotor resistance. the stator flux and torque error to generate the stator voltage
Lm The magnetizing inductance of the motor. and frequency[5-8].To verify the concept of the control
Lls The stator leakage inductance. techniques, this paper presents dynamic modeling, simulation
Llr The rotor leakage inductance. and analysis of induction motor drive. The three phase
Ȧr The slip frequency is the frequency of the induction motor mathematical model was derived and
actual rotor current. simulated using MATLAB/SIMULINK. The dynamic analysis
of V/F control induction motor is done using artificial
Llr The rotor leakage inductance referred to stator side.
intelligent controller under no, constant and variable load
Rr The rotor resistance referred to stator side.
conditions. In this paper, AI controllers are developed for
ȥqs ,ȥds q-axis and d-axis components of stator flux. dynamic analysis of induction motor. The training data is
ȥqr , ȥdr q-axis and d-axis components of rotor flux. generated via computer simulation of mathematical models of
iqs , ids q-axis and d-axis components of stator current. the Induction Motor and its drive, which is performedby a
iqr , iqr q-axis and d-axis components of rotor current. three-phase voltage inverter with Space Vector Pulse
WidthModulation (SVPWM), applied to the scalar control[9-
Veltech Multitech Dr.Rangarajan Dr.Sakunthala Engineering College,
Avadi, Chennai (Sponsors)

International Conference on Science, Engineering and Management Research (ICSEMR 2014)

978-1-4799-7613-3/14/$31.00 ©2014 IEEE
 IEEE-32331

16].Fig.1 shows the block diagram of the proposed model and ሺ߰ௗ௦ െ ߰௠ௗ ሻ
the drive system[5]. ݅ௗ௦ ൌ ሺ͸ሻ
ܺ௟௦
൫߰௤௥ െ ߰௠௤ ൯
݅௤௥ ൌ ሺ͹ሻ
ܺ௟௦

ሺ߰ௗ௥ െ ߰௠ௗ ሻ
݅ௗ௥ ൌ ሺͺሻ
ܺ௟௦
Where ȥmq and ȥmd are theflux linkages over Lm in the q
and d axes. The flux equations are written as follows:
Fig. 1 Block diagram of the Proposed Induction Motor Drive. ߰௤௦ ߰௤௥
߰௠௤ ൌ  ܺ௠௟ ൬ ൅ ൰ ሺͻሻ
ܺ௟௦ ܺ௟௥
III. DYNAMIC MODELING & SIMULATION OF INDUCTION
MOTOR DRIVE ߰ௗ௦ ߰ௗ௥
߰௠ௗ ൌ  ܺ௠௟ ൬ ൅ ൰ ሺͳͲሻ
The dynamic model of induction motor can be developed ܺ௟௦ ܺ௟௥
by writing differential equations for voltage and torque. For
dynamic analysis, poly phase windings are transformed into ͳ
ܺ௠௟ ൌ ଵ ଵ ଵ ሺͳͳሻ
two phase windings (q-d).In other words, the parameters ofan ൅ ൅
௑೘ ௑೗ೞ ௑೗ೝ
induction machine are transferred toanother d-q model which
remains stationary[1-10].
In the above equations,the speed Ȧr is related to the torque
by the following mechanical dynamic equation as:
݀߱௠ ‫߱݀ ʹܬ‬௥
ܶ௘ ൌ  ܶ௟௢௔ௗ ൅ ‫ܬ‬ ൌ  ܶ௟௢௔ௗ ൅ ሺͳʹሻ
݀‫ݐ‬ ‫ݐ݀ ݌‬
then Ȧr is achievablefrom above equation, where:
p: number of poles.
J: moment of inertia (kg/m2).
The model constructed according to the equations has been
simulated by using MATLAB/SIMULINK as shown in Fig.3
in conventional mode of operation of induction motor. A 3
phase source is applied to conventional model of an induction
Fig. 2 d q Model of Induction Motor
motor and the equations are given by:
In Fig.2stator inductance is the sum of the stator leakage
inductance and magnetizing inductance (Lls = Ls + Lm), and the
rotor inductance is the sum of the rotor leakage inductance and ܸ௔ ൌ  ξʹܸ௥௠௦ ‫݊݅ݏ‬ሺ߱‫ݐ‬ሻ ሺͳ͵ሻ
magnetizing inductance (Llr = Lr + Lm). From the equivalent ʹߨ
circuit of the induction motorin d-q frame, the model equations ܸ௕ ൌ  ξʹܸ௥௠௦ ‫ ݊݅ݏ‬൬߱‫ ݐ‬െ ൰ሺͳͶሻ
͵
are derived. The flux linkages are given by: ଶ஠
ܸ௖ ൌ  ξʹܸ௥௠௦ •‹ ቀɘ– ൅ ቁሺͳͷሻ
ଷ
By using Parks Transformation, voltages are transformed
ͳ ݀߰௤௦ ߱௘ to two phase in the d-q axes, and are applied to induction
ൌ ‫ݒ‬௤௦ െ ߰ െ ܴ௦ ݅௤௦ ሺͳሻ
߱௕ ݀‫ݐ‬ ߱௕ ௗ௦ motor. In orderto obtain the stator and rotor currents
ͳ ݀߰ௗ௦ ߱௘ ofinduction motor in two phase, Inverse park transformation is
ൌ ‫ݒ‬ௗ௦ െ ߰ െ ܴ௦ ݅ௗ௦ ሺʹሻ applied in the last stage [5-8].
߱௕ ݀‫ݐ‬ ߱௕ ௤௦

ͳ ݀߰௤௥ ሺ߱௘ െ ߱௥ ሻ
ൌ ‫ݒ‬௤௥ െ ߰ௗ௥ െ ܴ௦ ݅௤௥ ሺ͵ሻ
߱௕ ݀‫ݐ‬ ߱௕

ͳ ݀߰ௗ௥ ሺ߱௘ െ ߱௥ ሻ
ൌ ‫ݒ‬ௗ௥ ൅ ߰௤௥ െ ܴ௦ ݅ௗ௥ ሺͶሻ
߱௕ ݀‫ݐ‬ ߱௕

By substituting the values of flux linkages in the above

equations, the following current equations are obtained as:
൫߰௤௦ െ ߰௠௤ ൯
݅௤௦ ൌ ሺͷሻ Fig.3 Induction motor in d-q model
ܺ௟௦
 IEEE-32331

PI controllers are vary much sensitive to parameter Fig. 6. Speed output terminal of induction motor is applied as
variations inherent in real plant operations. The gain equation an input to fuzzy controller, and in the initial start of induction
for PI Controller motor the error is maximum, so according to fuzzy rules FC
produces a crisp value. Then this value will change the
T = Kp e + Ki e dt frequency of sine wave in the speed controller. The sine wave
The output of the PI controller is updated by varying the is then compared with triangular waveform to generate the
PI controller gains (Kpand Ki) based on the control law in the firing signals of IGBTs in the PWM inverters. The frequency
presence of parameter variation and drive nonlinearity. of these firing signals also gradually changes, thus increasing
the frequency of applied voltage to Induction Motor [12,14].
As discussed earlier, the crisp value obtained from Fuzzy Logic
Controller is used to change the frequency of gating signals of
PWM inverter. Thus, the output AC signals obtained will be
variable frequency sine waves. The sine wave is generated with
amplitude, phase and frequency which are supplied through a
GUI. Then the clock signal which is sampling time of
simulation is divided by crisp value which is obtained from
FLC. So by placing three sine waves with different phases, one
can compare them with triangular waveform and generate
Fig. 4Simulated Induction Motor Model in Conventional Mode necessary gating signals of PWM inverter. So at the first
The use of PI controllers for speed control of induction sampling point the speed is zero and error is maximum. Then
machine drives is characterized by an overshoot during whatever the speed rises, the error will decrease, and the crisp
tracking mode and a poor load disturbance rejection. This is value obtained from FLC will increase. So, the frequency of
mainly caused by the fact that the complexity of the system sine wave will decrease which will cause IGBTs switched ON
does not allow the gains of the PI controller to exceed a certain and OFF faster. It will increase the AC supply frequency, and
low value. If the gains of the controller exceed a certain value, the motor will speed up. The inputs to these blocks are the
the variations in the command torquecontroller gains are very gating signals which are produced in speed controller block.
high. Thus the motor reaches the reference speed rapidly and The firing signals are applied to IGBT gates that will turn ON
without overshoot.Step commands are tracked with almost zero and OFF the IGBTs.
steady state error and no overshoot. Load disturbances are
rapidly rejected and variations of some of the motor parameters
are fairly well dealt, it become too high and they will
destabilize the system. To overcome this problem we propose
the use of a limiter ahead of the PI controller [11]. This limiter
causes the speed error to be maintained within the
saturationlimits. Fig.5 shows the structure of PI controller.

Fig. 6 Fuzzy Controller.

TABLE I. MODIFIED FUZZY RULE DECISION

ǻe
NB NS ZZ PS PB
PB ZZ NS NS NB NB

PS PS ZZ NS NS NB
Fig. 5 PI controller e
ZZ PS PS ZZ NS NS
IV. ARTIFICIAL INTELLIGENT CONTROLLER
Despite ofthe concentration devoted to speed control NS PB PS PS ZZ NS
induction motor, many control methods cannot be directly
NB PB PB PS PS ZZ
applied to practical systems. Artificial Intelligent control
techniques are generally classified as expert system control,
(b) Artificial Neural Network (ANN):One of the most
fuzzy-logic control, neural-network control,Adaptive Neuro
important features of Artificial Neural Networks (ANN) is their
Fuzzy Inference System(ANFIS) and genetic algorithm.
ability to learn and improve their operation using a neural
Various artificial intelligent controllers are as follows:-
network training data[7-8].Fig. 7 depicts the network structure
(a) Fuzzy Logic Controller: The speed of induction motor of the NNC, which indicates that the neural network has three
is adjusted by the fuzzy controller. In Table-I, the fuzzy rules layered network structure. The first is formed with five neuron
decision implemented into the controller are given. The inputs ¨(ȦANN(K+1)), ¨(ȦANN(K)), ȦANN, ȦS(K-1), ¨(ȦS(K-
conventional simulated induction motor model as shown in 2)). The second layer consists of five neurons. The last one
Fig. 4 is modified by adding Fuzzy controller and is shown in contains one neuron to give the command variation ¨(ȦS(K)).
 IEEE-32331

The aim of the proposed NNC is to compute the command ‘A’. The dynamic behaviours of the PI controller, FLC
variation based on the future output variation controller, Neural Network controller and ANFIS are shown in
¨(ȦANN(K+1)).Hence, with this structure, a predictive control Fig.9 to Fig. 27at no load, constant load, variable load and
generating mode conditions in open loop and closed loop
with integrator has been realised. At time k, the neural mode.
network computes the command variation based on the output
at time (k+1), while the later isn’t defined at this time. In this (i) At No load conditions
case, it is assumed that ȦANN(K+1) Ł ȦANN(K)[14-15]. The rise time drastically decreaseswhen Adaptive neuro
fuzzy controlleris added to simulation model and both the
results are taken in same period of time. In ANFIS based
simulation, it is apparent from the simulation results shown in
Fig. 9 and Fig.18 torque-speed characteristics converge to
zero in less duration of time when compared with conventional
and PI Controller, FLC, and ANN. ANFIS controller has less
overshoot and settles faster in comparison with PI, FLC and
ANN. It is also noted that there is no steady-state error in the
speed response during theoperation when ANFIS controller is
activated.In conventional and PI controller, oscillations occur,
whereas in AI controller, no oscillations occur in the torque
Fig. 7 Neural network controller
response before it finally settles down as shown in Fig. 10 and
(c)Adaptive Neuro Fuzzy Controller (ANFIS):Adaptive
Fig 19. Good torque response is obtained with AI controller as
Neuro-Fuzzy Inference System (ANFIS)replaces the draw-
compared to conventional, PI and FLC at all time instants and
backs of Fuzzy Logic Control and Artificial Neural
it is observed that speed response is better than conventional,
Network.Adaptive neuro fuzzy combines the learning power of
PI controllers, FLC and ANN controller. Under dynamic
neural network with knowledge representation of fuzzy
condition, with the ANFIS controller, speed reaches its steady
logic.Neuro fuzzy techniques have emerged from the fusion of
state value faster with negligible ripple as compared to
Artificial Neural Networks (ANN) and Fuzzy Inference
Conventional PI, FLC and ANN as shown in Fig. 11 and Fig.
Systems (FIS) and have become popular for solving the real
20.
world problems. A neuro fuzzy system is based on a fuzzy
system which is trained by a learning algorithm derived from (ii) At constant load conditions:
neural network theory. In this paper, the inputs will be e(k) and A drive with conventional and PI controller has a peak
ǻe(k)[12].Fig.8 shows the overall structure of Adaptive Neuro- overshoot, but in case of fuzzy controller, neural network
Fuzzy model. controller and ANFIS, it is eliminated as shown in Fig. 12 and
Fig. 21. The PI controller is tuned at rated conditions in order
to make a fair comparison. Fig. 12, Fig. 13, Fig. 14and Fig.21,
Fig.22, Fig. 23 shows the simulated starting performance of the
drive with conventional, PI, FLC, neural and ANFIS based
drive systems, in open loop and closed loop respectively.
Although the PI controller is tuned by trial and error to give an
optimum response at this rated condition, the ANFIS controller
yields better performance in termsof faster response time and
lower starting current. It is worth mentioning here that the
performance obtained by the proposed AI controller is faster
than the PI controller, i.e. it achieves the steady state condition
Fig.8 Adaptive Neuro-Fuzzy model. faster than the PI controller.
V. SIMULATION RESULTS & DISCUSSION (iii )At variable load conditions:
Drive with conventional and PI controller speed response
A complete simulation model for scalar v/f controlled
has small peak at 0.5sec, but in case of fuzzy controller, neural
Induction motor drive incorporating PI, Fuzzy Logic
network and ANFIS controller speed response, it is quick and
Controller, Neural network controller and ANFIS is developed
smooth response which is shown in Fig.17 and Fig.26. Fig.15,
in open loop and closed loop mode. Neural networkcontroller
Fig. 16, Fig. 17 and Fig. 24, Fig. 25 Fig. 26 shows the
isdesigned by adjusting the weights and ANFIS is developedon
waveforms of Speed, Torque, Torque-Speed characteristics
a fuzzy system which is trained by a learning algorithm derived
with five controllers. Fig.17 and Fig.26 shows the speed
from neural network theory in order toget simulated results.
response for step change in the load torque using the
The performance of the artificial intelligent based induction
conventional, PI, fuzzy, neural and ANFIS controller,
motor drive is investigated at different operating conditions. In
respectively. The motor starts from standstill at load torque =
order to prove the superiority of ANFIS, a comparison is made
10 Nms and at t =0.5s, a sudden full load of 30 Nms is applied
with the response of conventional PI, FLC and Neural Network
to the system, then it is controlled by fuzzy, neural and ANFIS
based induction motor drive. The parameters of the induction
controller. Since the time taken by the conventional and PI
motor considered in this study are summarized in Appendix
controlled system to achieve steady state is much higher than
 IEEE-32331

fuzzy, neural and ANFIS controlled system, the step change in 200
Fig. 11 Speed Responses:Different Controllers at No Load.
Torque- Speed Characteristics of Conventional Controller at Const Load

load torque is applied at t = 1.5 sec. The motor speed follows 150
_______ Conventional

its reference with zero steady-state error and a fast response

nN-m
100

uei
using a fuzzy controller, neural and ANFIS. On the other hand,

q
50

Tor
0

the conventional and PI controller shows steady-state error -50

0 200 400 600 800 1000 1200 1400 1600 1800

with a high starting current. It is to be noted that the speed

Speed in RPM

(a) Conventional Controller.

response is affected by the load conditions. This is the 200
Torque- Speed Characteristics
200
Torque- Speed Characteristics

Torque in N-m

Torque in N-m
150 _______ PI Control 150 _______ Fuzzy Logic Control

drawback of a PI controller with varying operating conditions. 100

50

0
100

50

It is to be noted that the ANN controller and ANFIS gives

-50 -50
0 200 400 600 800 1000 1200 1400 1600 1800 0 200 400 600 800 1000 1200 1400 1600 1800
Speed in RPM Speed in RPM
Torque- Speed Characteristics Torque- Speed Characteristics
200 200

Torque in N-m

Torque in N-m
better responses in terms of overshoot, steady-state error and 150

100

50
_______ ANN Control
150

100

50
_______ ANFIS Control

fast response when compared with PI and fuzzy. These figures -50
0

0 200 400 600

Speed
800
in
1000
RPM
1200 1400 1600 1800
0

-50
0 200 400 600 800
Speed
1000
in RPM
1200 1400 1600 1800

also show that the neuro and ANFIS controller based drive (b) PI &Artificial Intelligent Controller.
system can handle the sudden increase in command speed Fig.12 Torque –Speed Characteristics of Different Controllers at Const. Load.
Torque Response of Conventional Controller at Const Load
quickly without overshoot, under- shoot, and steady-state error, 200

Torque in N-m
_______ Conventional
150

whereas the PI and fuzzy controller-based drive system has 100

steady-state error and the response is not as fast as compared to 50

neural network and ANFIS. Thus, the proposed ANFIS based -50
0 0.5 1 1.5
Time in Sec ...
2 2.5 3

drive has been found superior to the conventional PI-controller, (a) Conventional Controller.
FLC, and ANFISbased system. 200
Torque Responses at Const Load
_______ PI Control

Torque in Nm
150
_______ Fuzzy Logic Control
Open Loop: Simulation Results
Torque- Speed Characteristics of Conventional Controller with No Load
100

50
_______ ANN Control
_______ ANFIS Control
200

_______ Conventional
0
150

-50
nNm
-

100 0 0.5 1 1.5 2 2.5 3

ei

Time in Sec ...

u
q

50
o
Tr

0
(b) PI &Artificial Intelligent Controller.
Fig. 13 Torque Responses:DifferentController at Const Load.
-50
0 200 400 600 800 1000 1200 1400 1600 1800
Speed in RPM

(a) Conventional Controller. 2000

Speed Response of Conventional Controller at Const Load

Torque- Speed Characteristics Torque- Speed Characteristics Speed in RPM 1500

200 200
Torque in N-m

150 _______ PI Control 150 _______ Fuzzy Logic Control 1000 _______ Conventional
100 100
500
50 50

0 0 0
0 0.5 1 1.5 2 2.5 3
-50 -50 Time in Sec ...
0 200 400 600 800 1000 1200 1400 1600 1800 0 200 400 600 800 1000 1200 1400 1600 1800

200
Speed in RPM
Torque- Speed Characteristics
200
Speed in RPM
Torque- Speed Characteristics (a) Conventional Controller.
Torque in N-m

_______ ANN Control

150 150 Speed Responses at Const Load
100 100 2000
Speed in RPM

50 50
1500
0 0

-50 -50 1000 _______ PI Control

0 200 400 600 800 1000 1200 1400 1600 1800 0 200 400 600 800 1000 1200 1400 1600 1800 _______ Fuzzy Logic Control
Speed in RPM 500
_______ ANN Control
_______ ANFIS Control
(b) PI &Artificial Intelligent Controllers. -500
0

0 0.5 1 1.5 2 2.5 3

Fig.9 Torque –Speed Characteristics of different Controller at No Load Time in Sec ...
Torque Response of Conventional Controller with No Load
200 (b) PI &Artificial Intelligent Controller.
Torque in N-m

_______ Conventional
150

100
Fig.14 Speed Responses of Different Controller at Const. Load
Torque- Speed Characteristics of Conventional Controller at Variable Load
300

_______ Conventional
50 250

200
0
nNm
-

150
ei

-50
u

0 0.5 1 1.5 2 2.5 3

q

100
o
Tr

Time in Sec ... 50

(a) Conventional Controller.

Time in Sec ...
-50
0 200 400 600 800
Speed in RPM
1000 1200 1400 1600 1800

Torque Responses at No Load (a) Conventional Controller.

200 Torque- Speed Characteristics Torque- Speed Characteristics
300 300
_______ PI Control
Torque in N-m

Torque in N-m
Torque in Nm

_______ PI Control _______ Fuzzy Logic Control

200 200
_______ Fuzzy Logic Control
100 100 100

_______ ANN Control 0 0

_______ ANFIS Control 0 200 400 600

Speed in RPM
800 1000 1200 1400 1600 1800 0 200 400 600 800
Speed in RPM
1000 1200 1400 1600 1800

0 300
Torque- Speed Characteristics
300
Torque- Speed Characteristics
Torque in N-m

Torque in N-m

_______ ANN Control _______ ANFIS Control

200 200

100 100
-100
0 0.5 1 1.5 2 2.5 3 0 0

0 200 400 600 800 1000 1200 1400 1600 1800 0 200 400 600 800 1000 1200 1400 1600 1800
Time in Sec ... Speed in RPM Speed in RPM

(b) PI &Artificial Intelligent Controllers. (b) PI &Artificial Intelligent Controller.

Fig. 10 Torque Responses:Different Controllers at No Load. Fig.15 Torque Speed Characteristics of Different Controllers at Variable
Speed Response of Conventional Controller with No Load
Load.
2000 Torque Response of Conventional Controller at Variable Load
200
Speed in RPM

Torque in N-m

1500
_______ Conventional
_______ Conventional
100
1000

0
500

0 -100
0 0.5 1 1.5 2 2.5 3 0 0.5 1 1.5 2 2.5 3
Time in Sec ... Time in Sec ...
Torque Response of Conventional Controller with No Load
(a) Conventional Controller. (a) Conventional Controller.
Torque Responses at Variable Load
300
Speed Responses at No Load _______ PI Control
Torque in N m

2000 200 _______ Fuzzy Logic Control

Speed in RPM

_______ ANN Control

_______ PI Control 100 _______ ANFIS Control
1000
_______ Fuzzy Logic Control 0

0
_______ ANN Control -100
0 0.5 1 1.5 2 2.5 3
_______ ANFIS Control Time in Sec ...
-1000
0 0.5 1
Time in Sec ...
1.5 2 2.5 3 (b) PI &Artificial Intelligent Controller.
Fig. 16Torque Responses:Different Controllers at Variable Load.
(b) PI &Artificial Intelligent Controller.
 IEEE-32331

2000
Speed Response of Conventional Controller at Variable Load
Fig. 25Torque Responses:AI Controller at Variable Load.
Speed in RPM Speed Responses at Variable Load
1000 2000

Speed in RPM
1500
_______ Conventional _______ PI Control
0 1000
_______ Fuzzy Logic Control
500 _______ ANN Control
-1000 _______ ANFIS Control
0 0.5 1 1.5 2 2.5 3 0
Time in Sec ...
-500
0 0.5 1 1.5 2 2.5 3
(a) Conventional Controller. Time in Sec ...

2000
Speed Responses at Variable Load Fig.26 Speed Responses of AI Controller at Variable Load.
Generating Mode: Simulation Results
Speed in RPM

1500

1000
_______ PI Control 200
Torque -Spe ed Characte rstics of Conevntiona l Controlle r

_______ Fuzzy Logic Control 150

Motoring Generating

500 _______ ANN Control

*m
100

einN
0 _______ ANFIS Control

o
Trqu
50

-500
0 0.5 1 1.5 2 2.5 3 0

Time in Sec ...

-50
0 200 400 600 800 1000 1200 1400 1600 1800 2000

(b) Artificial Intelligent Controller. 200

Spe ed in RPM

Torque -Spee d Cha ra cte rstics of PI Controlle r

Fig.17 Speed Responses of Different Controllers at Variable Load. 150

Motoring Generating

Closed Loop: Simulation Results

*m
100

rqueinN
Torque- Speed Characteristics Torque- Speed Characteristics

o
200 200 50

T
Torque in N-m

Torque in N-m

150 _______ PI Control 150 _______ Fuzzy Logic Control

100 100 0

50 50

0 0 -50
0 200 400 600 800 1000 1200 1400 1600 1800 2000
-50 -50 Spe e d in RPM
0 200 400 600 800 1000 1200 1400 1600 1800 0 200 400 600 800 1000 1200 1400 1600 1800
Speed in RPM Speed in RPM Torque-Spee d Chara cte rstics of Fuzzy Controller
Torque- Speed Characteristics Torque- Speed Characteristics 200
200 200
Torque in N-m

Torque in N-m

_______ ANN Control _______ ANFIS Control Motoring

150 150 Generaing
100 100 150

50 50

0 0

TorqueinN*m
100
-50 -50
0 200 400 600 800 1000 1200 1400 1600 1800 0 200 400 600 800 1000 1200 1400 1600 1800
Speed in RPM Speed in RPM
50

Fig.18Torque –Speed Characteristics: AI controller at No Load.

0
Torque Responses at No Load
150
_______ PI Control
Torque in N m

-50
100 _______ Fuzzy Logic Control 0 200 400 600 800 1000 1200 1400 1600 1800 2000
Spe ed in RPM
_______ ANN Control
50 _______ ANFIS Control Torque-Speed Characterstics of ANN Controller
200

0 Motoring Generaing

-50 150
0 0.5 1 1.5 2 2.5 3
Time in Sec ...

TorqueinN*m
100

Fig. 19Torque Responses:AI Controller at No Load. 50

Speed Responses at No Load
2000
Speed in RPM

0
1000 _______ PI Control
_______ Fuzzy Logic Control
0 _______ ANN Control -50
0 200 400 600 800 1000 1200 1400 1600 1800 2000
_______ ANFIS Control Speed in RPM

-1000 Torque-Speed Chara cterstics of ANFIS Controlle r

0 0.5 1 1.5 2 2.5 3 250

Time in Sec ... Motoring Generating

200

Fig. 20 Speed Responses:AI Controller at No Load. 150

*m

Torque- Speed Characteristics Torque- Speed Characteristics

einN

200 200
Torque in N-m

Torque in N-m

100
rqu

150 _______ PI Control 150 _______ Fuzzy Logic Control

To

100 100 50

50 50

0 0 0

-50 -50
0 200 400 600 800 1000 1200 1400 1600 1800 0 200 400 600 800 1000 1200 1400 1600 1800
-50
Speed in RPM Speed in RPM 0 200 400 600 800 1000 1200 1400 1600 1800 2000
Speed in RPM
Torque- Speed Characteristics Torque- Speed Characteristics
200 200
Torque in N-m

Torque in N-m

_______ ANN Control _______ ANFIS Control

150 150

100 100

50

0
50

0 Fig.27. Torque –Speed Characteristics: Conventional &AI controller in

-50 -50
0 200 400 600 800
Speed in RPM
1000 1200 1400 1600 1800 0 200 400 600 800 1000
Speed in RPM
1200 1400 1600 1800
Motoring and Generation Mode.
Fig.21 Torque –Speed Characteristics of AI Controller at Const. Load.
150
Torque Responses at Const Load
(iii) Generating Mode
_______ PI Control
Torque in N m

_______ Fuzzy Logic Control

100

50
_______
_______
ANN Control
ANFIS Control Recently, because of the increasing importance
0 ofdistributed energy resources such as wind turbines andmicro
-50
0 0.5 1 1.5 2 2.5 3 turbines, there is renewed attention to inductiongenerators. The
Time in Sec ...
model described in this paper can also beused to model an
Fig. 22Torque Responses:AI Controller at Const Load.
Speed Responses at Const Load
induction generator. The only change,compared with motoring
2000
operation, is that an external mover rotates the motor at a speed
Speed in RPM

1500

1000

500
_______
_______
_______
PI Control
Fuzzy Logic Control
ANN Control
higher than the synchronousspeed. This change can be
0 _______ ANFIS Control implemented in Simulink by usinga negative load torque
-500
0 0.5 1 1.5
Time in Sec ...
2 2.5 3
instead of a positive one used formotoring.Fig. 27 shows the
Fig.23 Speed Responses of AI Controller at Const. Load.
d
torque-speed curve of the inductionmachine in both motoring
300
Torque- Speed Characteristics
300
Torque- Speed Characteristics
and generating regions. Themachine accelerates freely to
Torque in N-m

Torque in N-m

_______ PI Control _______ Fuzzy Logic Control

200 200

100 100 almost the synchronous speedwith direct ac start-up, and then a
0

0 200 400 600 800 1000 1200 1400 1600 1800

0

0 200 400 600 800 1000 1200 1400 1600 1800

large negative torque isapplied so that the machine starts
generating. With AI controllers the motor reaches steady state
Speed in RPM Speed in RPM
Torque- Speed Characteristics Torque- Speed Characteristics
300 300
Torque in N-m

Torque in N-m

_______ ANN Control _______ ANFIS Control

200

100
200

100
value faster.
0 0

0 200 400 600 800

Speed in RPM
1000 1200 1400 1600 1800 0 200 400 600 800 1000
Speed in RPM
1200 1400 1600 1800
VI. CONCLUSION
Fig.24Torque –Speed Characteristics of AI Controller at Variable Load.
150
Torque Responses at Variable Load Dynamic modeling, simulation and analysis of the
_______ PI Control
induction motor drive are presented using conventional, PI,
Torque in N m

100 _______ Fuzzy Logic Control

_______ ANN Control
50 _______ ANFIS Control
FLC ,NN and ANFIS. From the speed curves in five models,
0

-50
the fuzzy controller decreases the rise time drastically, in the
0 0.5 1 1.5 2 2.5 3
Time in Sec ... manner which the frequency of sine waves are changing
 IEEE-32331

according to the percentage of error from rated speed. The [9] Sayeed Nurul Ghani, “Digital Computer Simulation of Three-phase
frequency of these firing signals also gradually changes, thus Induction Machine Dynamics-AGeneralized Approach” IEEE Tras.on
Industry Applications , Vol 21. No I, Jan/Feb1988,pp 106-114.
increasing the frequency of applied voltage to Induction Motor. [10]C. N. Weygandt and S. Charp, “Electromechanical Transient
According to the direct relation of induction motor speed and Performanceof Induction Motors,” AIEE Trans. Vol-65,1965,pp 1000-
frequency of supplied voltage, the speed also will increase. 1009.
With results obtained from simulation, it is clear that for the [11]A. L. Orille,G.M.A. Sowilam, .A. Valencia, “A NewSimulation of
same operating condition of induction motor, fuzzy controller Symmetrical Three Phase Induction Motor under Transformation of
has better performance than the conventional PI controller. By Park,”ElsevierComputers & Industrial Engineering Vol. 37
comparing Adaptive neuro fuzzy model with FLC model, it is 1999, pp 359-362.
[12]Narongrit Pimkumwong, Amorn Onkrong,Tirasak Sapaklom,“Modeling
apparent that by adding learning algorithm tosupplied voltage, and Simulation of Direct Torque Control InductionMotor Drives via
the speed will also increase. With results obtained from Constant Volt/Hertz Technique” Elsevier Procedia on Engineering 31,
simulation, it is clear that for the same operating condition of 2012,pp 1211-1216.
induction motor, fuzzy controller has better performance than [13]Tiago Henrique dos Santosa, Alessandro Goedtel Sergio Augusto
the conventional PI controller. By comparing Neural network Oliveira da Silvab,Marcelo Suetake, “Scalar control of an induction
motor using a neural sensorlesstechnique,”Elsevier Electric Power
controller with FLC model, it is apparent that by adding Systems Research,108 ,2014, pp 322– 330.
learning algorithm to the control system will decrease the rising [14] H. C. Stanley, “An Analysis of the Induction Machine,” AIEE Trans,
time more than expectation and it proves ANFIS controller has vol-57,1938,pp 751-757
better dynamic performance as compared to NN, FLC and [15] P. Menghal, A Jaya Laxmi, “Neural Network Based Dynamic
conventional PI controller.The comparative results prove that Performance of Induction Motor” Springer Proceeding Advances in
Intelligent Systems and Computing (AISC) vol.259,Apr 2014 539-552.
the performance of scalar v/f -control drive with ANFIS [16] P M Menghal, A Jaya Laxmi ““Application of Artificial Intelligence
controller is superior to that with conventional P-I, fuzzy and Controller for Dynamic Simulation of Induction Motor Drives” Asian
neural network controller. The transient response of induction Power Electronics Journal, Vol. 7, No. 1, Sep 2013,pp 23-29.
machine using ANFIS has been improved and therefore the [17] P M Menghal, A Jaya Laxmi, N Mukhesh “Dynamic Simulation of
dynamic response of the same has been made faster. For Induction Motor Drive Using Artificial Intelligent Controller” IEEE
variable loads, when there is a sudden change in load, the International Conference on Control, Instrumentation, Energy &
Communication(CIEC14), 28-31 Jan 2014, pp 356-360.
ANFIS controller reaches its steady state value faster and there [18] P M .Menghal, A Jaya Laxmi, "Adaptive Neuro Fuzzy Based Dynamic
are no overshoots as compared to the PI, Fuzzy and NN Simulation of Induction Motor Drives," IEEE International Conference
controller. By dynamic modelling and simulation ANFIS Fuzzy Systems,7-10 July 2013, pp 1-8.
controller gives best performance as compared to the other AI [19] P M .Menghal, A Jaya Laxmi, " Neural Network Based
controllers. DynamicSimulation of Induction Motor Drive," IEEE International
Conference on Power, Energy and Control (ICPEC-13),6-8 Feb 2013 ,
APPENDIX 'A' pp 566-571.
[20] P M .Menghal, A Jaya Laxmi, "Adaptive Neuro Fuzzy Interference
The following parameters of the induction motorare chosen (ANFIS) based simulation of Induction motor drive," International
for the simulation studies: Review on Modeling and Simulation (IRMOS), Vol. 5, No. 5, Oct
2012, pp. 2007-2016.
V = 220 f = 50 HP = 3 Rs = 0.435 [21] P M .Menghal, A Jaya Laxmi, "Artificial intelligence based induction
Rr = 0.816 Xls = 0.754 Xlr = 0.754 Xm = 26.13 motor drive," Michael Faraday IET India Summit, Kolkata, India,
p=4 J = 0.089 rpm = 1710 November 25, 2012, pp. 208-212.
[22] P M .Menghal, A Jaya Laxmi “Artificial Intelligence Based Dynamic
Simulation of Induction Motor Drives” IOSR Journal of Electrical and
REFERENCES Electronics Engineering (IOSR-JEEE) ISSN: 2278-1676 Volume 3,
Issue 5 (Nov. - Dec. 2012), pp 37-45.
[1] K. L . Shi, T . F. Chan, Y. K. Wong and S. L . HO, "Modeling and [23] M. Nasir Uddin and Muhammad Hafeez, "FLC-Based DTC Scheme to
simulation of the three phase induction motor Using SIMULINK," Improve the Dynamic Performance of an IM Drive," IEEE Trans.on
Int.J. Elect. Engg. Educ., Vol. 36, 1999, pp. 163–172. Industry Applications , Vol . 48 , No. 2, Mar/Apr 2012, pp. 823-831.
[2] Tze Fun Chan and Keli Shi, "Applied intelligent control of induction [24] M. Nasir Uddin and Muhammad Hafeez.,"FLC-Based DTC Scheme to
motor drives," IEEE Willey Press, First edition, 2011. Improve the Dynamic Performance of an IM Drive,” IEEE Trans. on
[3] P.C. Krause, "Analysis of Electrical Machinery and Drives Industry Applications , Vol -48 , No 2, Mar/Apr 2012; 823-831.
System,"IEEE Willey Press, 2000. [25] M. Nasir Uddin, Hao Wen, "Development of a Self-Tuned Neuro-Fuzzy
[4] Ned Mohan, "Advanced Electric Drives: Analysis, Control Modeling Controller for Induction Motor Drives," IEEE Trans on Industry
using Simulink,"MNPERE Publication,2001. Application. Vol. 43, No. 4, July/August 2007,pp. 1108-1116.
[5] Tiago Henrique dos Santos, Alessandro Goedtel, Sergio Augusto [26] M Nasir Uddin, Tawfik S. Radwan and Azizur Rahman, "Performance
Oliveirada Silva and Marcelo Suetake, “Scalar control of an induction motor of Fuzzy logic based indirect vector control for induction motor drive,"
usinga neural sensorlesstechnique,” Elsevier Electric Power IEEE Trans on industry application, Vol. 38, No.5 Sept /Oct 2002, pp.
System Research, vol 108,2014,pp 322-320. 1219-1225.
[6] Howard E. Jordan, “Analysis of Induction Machines in Dynamic [28] Besir Dandil,Muammer Gokbulut Fikrat Ata, "A PI Type Fuzzy –
Systems,”IEEE Trans.on Power Apparatus And Systems Vol. Pas-84, Neural Controller for Induction Motor Drives,"Journal of Applied
No. 11 November 1965,pp 1080-1088. Sciences 5(7) 2005, pp. 1286-1291.
[7] H. H. Hwang, “Unbalanced Operations of AC Machines,”IEEE [29] Pradeep Chatterjee, B.M. Karan and P.K. Sinha,"Fuzzy Control of
Transon Power Apparatusand Systems Vol. PAS-84,NO. 11,Nov 1965, Induction Motor with Reduced Rule Base," Serbian Journal of
pp 1054-1066. Electrical Engineering, Vol 4, No. 2,Nov 2007, pp. 147-159.
[8] P. C. Krause, C. H. Thomas, “Simulation of Symmetrical Induction [30] I.H. Altas and A.M. Sharaf,"A Generalized Direct Approach for
Machinery,” IEEE Transon Power Apparatusand Systems Vol. PAS- Designing Fuzzy Logic Controllers in Matlab/Simulink GUI
84,NO. 11,Nov 1965, Environment," International Journal of Information Technology and
pp 1038-1053. Intelligent Computing, Int. J. IT&IC No.4 Vol..1, 2007.