‫بسم

الرحمن الرحيم‬
‫اللجنة العلمية الدائمة لترقية أساتذه جامعة األزھر‬
‫التخصص‪ :‬الھندسة الكھربية‬

‫استمارة تمأل بواسطة المتقدم‬

‫لبيان دوره فى البحث معتمدا من باقى الباحثين‬
‫ولو اشترك مع أخرين فى غير تخصصه‬

‫بحث رقم‪ :‬الثالث‬

‫عنون البحث‪:‬‬

‫تحسين التحكم المباشر فى العزم للمحركات الحثية باستخدام المتحكم العصبى العشوائى‬

‫‪Improvement Of Direct Torque Control Of Induction Motor Drives‬‬

‫‪Using Neuro-Fuzzy Controller‬‬

‫الدور‬
‫توقيع‬ ‫التخصص‬
‫كتابة البحث‬ ‫اجراء البحث تحليل النتائج‬ ‫فكرة وبروتكول‬ ‫اسماء المشتركين في البحث‬
‫المشتركين‬ ‫الدقيق‬
‫واعداده للنشر‬ ‫ومناقشتھا‬ ‫وتنفيذه‬ ‫البحث‬
‫اآلت‬
‫‪%70‬‬ ‫‪%70‬‬ ‫‪%70‬‬ ‫‪%70‬‬ ‫د ‪.‬عمرو رفقى عبدالوھاب‬
‫كھربية‬
‫اآلت‬
‫‪% 20‬‬ ‫‪% 20‬‬ ‫‪% 20‬‬ ‫‪% 20‬‬ ‫د‪ .‬سالمة أبو زيد‬
‫كھربية‬
‫اآلت‬
‫‪% 10‬‬ ‫‪% 10‬‬ ‫‪% 10‬‬ ‫‪% 10‬‬ ‫م‪ .‬أحمد حسن عادل‬
‫كھربية‬

‫اسم المتقدم‪ :‬عمرو رفقى على عبدالوھاب‬

‫التوقيع‪:‬‬
 ‫الملخص العربى للبحث‬

‫عنون البحث‪ :‬تحسين التحكم المباشر فى العزم للمحركات الحثية باستخدام المتحكم‬
‫العصبى العشوائى‬
‫يعتبر التحكم المباشر فى العزم طريقة متطورة وبسيطة للتحكم فى التحريك الكھربى للمحركات الحثية‬
‫وكذا فإن لھا بعض الكثير من المميزات بالمقارنة ب طرق التحكم األخرى التى تعتمد على تغير‬
‫التردد‪ ،‬ولكن بھا عيب أساسى اال وھو إرتفاع تموج عزم الدوران‪ .‬يقدم البحث تصميم وتحليل للتحكم‬
‫المباشر فى العزم للتحريك الكھربى الخاص بالمحركات الحثية باستخدام جداول التوصيل مع متحكمات‬
‫تباطؤ للعزم والمجال المغناطيسى و متنبأت بقيم العزم والمجال‪ .‬كما يناقش البحث تطوير التحكم‬
‫التقليدى المباشر بالعزم باستخدام تطوير جداول التوصيل وتطبيق لتقنيات الذكاء اإلصطناعى لتقليل‬
‫التموج فى العزم والمجال المغناطيسى وكذا تقليل تموج التيار الكھربى للحصول على أداء أفضل‬
‫للتحريك الكھربى للمحركات الحثية مع التحكم المباشر فى العزم‪ .‬كما يحتوى البحث على مقارنة بين‬
‫الطرق التقليدية للتحكم المباشر فى العزم والطريقة المقترحة باستخدام المطق العشوائى والعصبى‪.‬‬
‫بمقارنة نتائج البحث يتبن أن التموج فى العزم والمجال المغناطيسى وكذا التيار الكھربى للمحرك‬
‫الحثى أقل بشكل ملحوظ بإستخدام المقترح بإستخدام تقنيات الذكاء اإلصطناعى عنھا فى الطريقة‬
‫التقليدية للتحكم المباشر بالعزم‪ .‬كما أظھرت النتائج إستجابة ديناميكية جيدة جدا مع طريقة التحكم‬
‫المقترحة‪ .‬وكذا استخدام متحكم الشبكات العصيبة بدال من المتحكم التكاملى المباشر يقلل بشكل ملحوظ‬
‫التجاوز الكبير وكذا التباطئ الشديد فى استجابة السرعة ويمنح سرعة المحركة استجابة سريعة منتظمة‬
‫مما يؤكد على أن أستخدام تقنيات التحكم المقترحة تحسن من األداء الديناميكى للتحريك الكھربى‬
‫للمحركات الحثية باستخدام التحكم المباشر فى العزم‪.‬‬

‫‪1‬‬
Volume 2, Issue 10, October, 2015
ISSN : 3159-0040
www.jmest.org
 Journal of Multidisciplinary Engineering Science and Technology
ISSN: 3159-0040 (Online)
Vol. 2 Issue 10, October - 2015
Paper Title Pages
Soil-cement Stabiliza on For Road Pavement Using Soils Obtained From Agu-awka In Anambra State
2668-2670
Authors: Okonkwo V. O.; Nwokike V. M.

Effect Of Curing Age On The Compressive Strength Of Concrete Made From Local Granite Chippings
2671-2675
Authors: Okonkwo V. O.; Nwokike V. M.
Suitability Of Lean Construc on Approach In Nigerian Project Delivery
2676-2679
Authors: Samaila Adamu; Razali Abdulhamid
Development Of Emcs In Power Plants
2680-2686
Authors: Hassan Saad Mohamed; Sayed Ahmed Nagy; Salah El-deen Kamal El-sayed
Analy cal Solu ons To Elas c Func onally Graded Cylindrical And Spherical Pressure Vessels
2687-2693
Authors: Ahmet N. Eraslan; Tolga Akis
Clustering Using Principal Component Analysis As An Input Tool
2694-2699
Authors: Evelyn Nkiruka; Joseph Uchenna; Okeke
The Role Of Modeling, Simula on And Analysis Stage In Mechatronics Systems Design Educa on
2700-2715
Authors: Farhan A. Salem
The Role Of Control System/algorithm Subsystems In Mechatronics Systems Design
2716-2732
Authors: Farhan A. Salem
Numerical Analysis Of Interfacial Behavior With Turbulence Effect In Water Model Of A Continuous 
Cas ng Mold 2733-2739
Authors: Rajat Kumar Das; Chetana Tripathy; Sukanta Kumar Dash
Stream Sediment Geochemistry As A Tool For Assessing Mineral Potentials In Arinta And Olumirin 
Waterfalls In Eki  And Osun States, Southwestern Nigeria 2740-2752
Authors: Olusiji Samuel Ayodele; Segun Ajayi Akinyemi
Determina on Of Bioac ve Molecules And An oxidant Ac vity In S nging Ne le (ur ca Dioica)
Authors: Akram Hijazi; Bouchra Sayed-ahmad; Mariam Akil; Hassan Rammal; Hanan Assaf; Houssein  2753-2758
Annan

Numerical Solu on Of 2d Spl Heat Conduc on Model For Analyzing Thermal Wave Oscilla ons

2759-2764
Authors: T. N. Mishra; K. N. Rai
Design And Construc on Of A Dust Extractor Machine
2765-2768
Authors: Engr Ogbeide S. O.
A Novel Sequen al Frac onal Order Kalman Filter Considering Colored Noise
2769-2775
Authors: Mojtaba Asad; Behrooz Safarinejadian; Mokhtar Sha Sadeghi
Op miza on Of Natural Gas Dehydra on Using Triethylene Glycol (teg)
2776-2779
Authors: Evbuomwan, B. O; Felicia, O. R
Pa ern Discovery Of Adhd Disorder Using Graph Theory On Task-free Fmri Data.
2780-2789
Authors: Reza Khsorowabadi; Farnaz Mohammadi; Mohammad Rostami; Hamidreza Pouretemad

Evolu onary Trend In The Development Of Computer Programming Languages And Concepts
2790-2794
Authors: Calister Nnenna; Onu F. U.; Asogwa Tochukwu Chijindu; Edeh Eucharia Ujunwa
Comparison Of The Effects Of Alkali And Silica On The Tensile Strength Properties Of Epoxy Based 
Composites 2795-2797
Authors: Menderes Koyuncu
Evalua ng Asr Systems Under Noisy Condi on Using Aurora Database
2798-2800
Authors: Saimir Tola; Ligor Nikolla
Design Analysis Of Kitchen-waste-heat Energy For The Production Of Telephones’ Chargers As An 
Alterna ve Grid Technology 2801-2805
Authors: Mgbachi C. A; Oluka S. I,
Innova ve Design Of An An -bacterial Shopping Cart A achment
2806-2810
Authors: Feng-chyi Duh; Jen-chao Tai
Markov Decision Process Based Multi Agent Learning Style Intelligent Tutoring System For Modelling 
Students 2811-2819
Authors: Agwu C. O.; Achi I. I.; Prof. Inyiama H. C.; Prof. Bakpo F. S.
Noise Suppression In Automa c Speech Recogni on Using Hidden Markov Model
2820-2821
Authors: Saimir Tola; Ligor Nikolla
A Profi T-maximizing Approach For Transmission Expansion Planning Using A Incen ve Mechanism
2822-2825
Authors: Alireza Sedaghati; Mehrab Najafi
The Effect Gorbunov In Prac ce
2826-2829
Authors: Sutorikhin V. A.; Shchedrivyi K. V.
Brain Tumor Segmentation Using Biography Based Neural Clustering (bbnc) And Genetic Processing 
For Medical Images 2830-2833
Authors: Chandanpreet Kaur; Bikrampal Kaur
Appraisal Of The Impact Of Corruption On Sustainable Development In Nigerian Construction 
Industry 2834-2842
Authors: Kasimu M. A; Kolawole A. F
Analy cal And Numerical Solu ons To A Rota ng Fgm Disk
2843-2850
Authors: Eray Arslan; Ahmet N. Eraslan
A Comparative Analysis Of The Dispersion Curves In The Semi-classical Theory Of Laser And 
Electromagne cally Induced Transparency 2851-2854
Authors: Ranjana Bora Bordoloi; Rajib Bordoloi; Gauranga Dhar Baruah
Stability Mi ga on Of Dg Based Electric Power Grids Using An Extrinsic Iner a
2855-2859
Authors: Mahmoud El-dabah; Mohamed Ibrahim; Mohamed El-gazzar; El-said Othman

Feasibility Of Redwood Wood For Paper Produc on And The Effects Of Bea ng On Paper’s Proper es

2860-2862
Authors: Ayhan Gençer; Ufuk Özgül; Saadettin Murat Onat; Fadime Yurdakurban; Hikmet Yazici

THIS MANUSCRIPT HAS BEEN WITHDRAWN AT THE AUTHOR`S REQUEST

Authors: Author's name withdrawn at their request 2863-2871

A First Approach On Posi oning Difis System For Containing Oil And Oil-gas Leaks
2872-2876
Authors: Dimitrios-periklis A. Giannoulis; Dionissios P. Margaris
Ethical Guidelines For Good Prac ce In Teaching And Research
2877-2880
Authors: Ajaegbu, Chigozirim; Ajaegbu, Oguchi Onyeizu
A Comparative Study Of Linear Predictive Analysis Methods With Application To Speaker 
Iden fica on- Over A Scrip ng Programing 2881-2885
Authors: Ervenila Musta; Vangjush Komini
Laboratory Inves ga on On Levee Breach And Risk On Floodplain Of Alluvial Fan River
2886-2894
Authors: Md. Serazul Islam; Shahnaj Pervin; Tetsuro Tsujimoto; Takashi Tashiro
Membrane Solu ons For Circular Toroidal Shells Under Internal Hydrosta c Pressure
2895-2900
Authors: Enoma N; Unueroh U. G; Egware H. O; Itoje H. J
The Bending Moment Changes Of Tunnel Lining Due To Kars fica on
2901-2904
Authors: Ashkan Rahmani; Vahid Hosseinitoudeshki
A Development Of The Exhibition Or Performance Tree Shape Robot Having A Growth Reproduction 
Func on 2905-2909
Authors: Hong Seok Lim
Consolidating Data For Forming An Information Image Of Geospatial Objects From Heterogeneous 
Social Media 2910-2912
Authors: Iryna Khmil
Improvement Of Direct Torque Control Of Induc on Motor Drives Using Neuro-fuzzy Controller
2913-2918
Authors: Ahmed Hassan Adel; Dr. Salama Abo-zaid; Dr. Amr Refky
Designing An Integrated Web Based Lecturer/students Interaction System For Institutions And 
Colleges 2919-2927
Authors: Igbajar Abraham; Ekwueme Victoria A.
THIS MANUSCRIPT HAS BEEN WITHDRAWN AT THE AUTHOR`S REQUEST
2928-2931
Authors: Author's name withdrawn at their request
Wi-fi Coverage Area Analysis And Pathloss Parameter Modeling Using The Linear Regression Method
2932-2935
Authors: Emerole K. C.; Ezeh G. N.; Nwogu U. O
 Journal of Multidisciplinary Engineering Science and Technology (JMEST)
ISSN: 3159-0040
Vol. 2 Issue 10, October - 2015

Improvement Of Direct Torque Control Of

Induction Motor Drives Using Neuro-Fuzzy
Controller
Eng. Ahmed Hassan Adel, Dr. Salama Abo-Zaid, Dr. Amr Refky
Dept. of Electrical Engineering
Al-Azhar University
Cairo, Egypt
ahmedhassanadel@yahoo.com

Abstract— Direct torque control (DTC) is an
advanced and simple control method for induction
motor drive has many advantages over other
variable frequency control methods, but it has a
common disadvantage of high torque ripple. This
paper introduces design and analysis for DTC of
induction motor drives using the switching table,
hysteresis controllers and flux and torque
estimator. This paper also proposes improvement
of the conventional DTC using the improvement of
the switching table and the application of the
artificial intelligent techniques to minimize the
torque, flux and current ripples and to get better
performance of the motor. At the end of this paper
there is a comparison between the conventional
DTC and the proposed one.
Keywords— Induction Motor Drive; Direct
Torque Control (DTC); switching table; flux and
torque estimator; artificial neural networks (ANN) ;
Fuzzy logic control (FLC).
I. INTRODUCTION
In many variable speed drive applications(e.g.
traction drives for electric vehicles), torque control is
required or desired, but precise, closed-loop control of
speed is not necessary. Torque control provides
greatly improved transient response, avoidance of
nuisance over-current trips, and elimination of load
dependent controller parameters[1]. DTC was
introduced by Takahashi (1984) in Japan and then in
Germany by Depenbrock (1985)[2].
DTC method is characterised by its simple
implementation and a fast dynamic response.
Furthermore, the inverter is directly controlled by
the algorithm, i.e. a modulation technique for the
inverter is not needed. The main advantages of
DTC are absence of coordinate transformation and
current regulator, absence of separate voltage
modulation block[3].
Common disadvantages of conventional DTC
are high torque ripple, It also needs flux and torque
estimators and therefore, accurate machine
parameters are required. For that reason the
application of artificial neural network and fuzzy
logic control attracts the attention of many scientists
from all over the world. The reason for this trend
is the many advantages which the architectures of
ANN have over traditional algorithmic methods
[3],[4].
II. STRATEGY OF DIRECT TORQUE CONTROL
DTC technique can be easily implemented using
two hysteresis controllers (one for flux and the other
for torque), torque and flux estimators and a switching
table to select the proper voltage vector[5]. The basic
functional blocks used to implement the DTC scheme
are represented in Fig.1.
Fig. 1. Block diagram of DTC of I.M drives
The torque control of the inverter fed machine is
carried out by hysteresis control of magnitude of stator
flux and torque that selects one of the six active and
two zero inverter voltage vector ( ) The use of a
switching table for voltage vector selection provides
fast torque response, low inverter switching
frequency and low harmonic losses without the
complex field orientation by restricting the flux and
torque errors within respective flux and torque
hysteresis bands with the optimum selection being
made[5].
In a voltage fed three phases, the switching
commands of each inverter leg are complementary So
for each leg a logic state (i=a,b,c) can be defined ;
is 1 if the upper switch is commanded to be closed and
0 if the lower one in commanded to be close (first)[7].
The following equation give the voltage vectors:

www.jmest.org
JMESTN42351145 2913
 Journal of Multidisciplinary Engineering Science and Technology (JMEST)
ISSN: 3159-0040
Vol. 2 Issue 10, October - 2015
TABLE II. SWITCHING TABLE FOR MODIFIED CLASSICAL DTC

Sector
√ [ ] (1) 1 2 3 4 5 6
Hψ HTe

Eight switching states can be taken according to 1 V2 V3 V4 V5 V6 V1

equation (1): two zero voltage vectors and six non-zero 1 0 V0 V7 V0 V7 V0 V7
voltage vectors show by Fig. 2[7]. -1 V0 V7 V0 V7 V0 V7
1 V3 V4 V5 V6 V1 V2
-1 0 V0 V7 V0 V7 V0 V7
-1 V0 V7 V0 V7 V0 V7

III. DIRECT TORQUE CONTROL USING FUZZY LOGIC

Fuzzy logic controller can be used to replace the
switching table of the conventional DTC, and this result
in improving the performance of the motor and
reducing the torque ripple. Fig. 3 shows block diagram
for DTC of I.M using fuzzy logic controller.

Fig. 2. Partition of the dq plane into 6 angular sectors

In Fig. 2 the stator flux is represented by the direct

axis and the torque is represented by the quadrature
axis ,It can be seen that selecting the proper voltage
vector results in maintaining the torque and stator flux
within the limits . The switching table proposed by
Takahashi is given by Table1:
TABLE I. SWITCHING TABLE FOR CLASSICAL DTC

Sector
1 2 3 4 5 6
Hψ HTe
1 V2 V3 V4 V5 V6 V1
1 0 V0 V7 V0 V7 V0 V7 Fig. 3. Block diagram of DTC using fuzzy logic controller
-1 V6 V1 V2 V3 V4 V5 A. Design of DTC using fuzzy logic
1 V3 V4 V5 V6 V1 V2 A designer of Fuzzy logic controller must choose
-1 0 V0 V7 V0 V7 V0 V7 memberships for input and output variables and put
-1 V5 V6 V1 V2 V3 V4 the rules for the control process. Figure 4 shows
functional block diagram for fuzzy logic controller; in
this diagram the decision making is made according to
The voltage vector table receives the flux level fuzzified inputs and rule base which the designer put
, the torque level and the sector number and according to data base, then the output is deffuzified.
generates appropriate control for the inverter from a
look-up table as in Table 1[6].
Table 1 can be simply modified by applying zero
voltage vectors (V0 , V7) for the torque decrease states
(-1) ,and this modification will result in decreasing the
torque ripple largely , a considerable reason for this
decrease is that applying the zero voltage vectors
result in reducing the inertia of the motor at this instant
and this result in reducing the torque with a percent
which is more suitable than the percent given by
applying the vectors in table 1 for the torque decrease
states[8]. Table 2 illustrates this modification.

Fig. 4. Functional block diagram for the fuzzy logic

controller

www.jmest.org
JMESTN42351145 2914
 Journal of Multidisciplinary Engineering Science and Technology (JMEST)
B. Input and Output variables
Flux error, torque error and flux position are the
input variables of fuzzy logic controller and the stator
voltage vector ( V0 - V7 ) is the output variable. Figures
5-8 show memberships for the inputs and output
variables.
Fig. 5. Membership for the flux error
Fig. 6. Membership for the torque error
Fig. 7. Membership for the flux position
Fig. 8. Membership for the stator voltage vector
C. Rule base
As shown in figure 4 decision making is made
according to the inputs and the rule base. A designer
of FLC puts the rules according to data base for the
control method used for the system to be controlled.
The rules of DTC using fuzzy logic is put according to
figure 2 and table 2 mentioned above.
www.jmest.org
JMESTN42351145
ISSN: 3159-0040
Vol. 2 Issue 10, October - 2015
IV. ARTIFICIAL NEURAL NETWORKS
Artificial neural networks (ANN) are, as their name
indicates, computational networks which attempt to
simulate, in a gross manner, the networks of nerve cell
(neurons) of the biological (human or animal) central
nervous system[9]. Neural networks are composed of
simple elements called neurons operating in parallel.
As in nature, the connections between elements
largely determine the network function[10].
Fig. 9. Multiple Layer Neural Network
(∑ ) (2)
where;
R : number of elements in input vector.
S : number of neurons.
A neural network can be trained to perform a
particular function by adjusting the values of the
connections (weights) between elements. Typically,
neural networks are adjusted, or trained, so that a
particular input leads to a specific target output[10].
Neural networks have self-adapting compatibilities
which makes them well suited to handle non-linarites,
uncertainness and parameter variations. Neural
networks are capable of generalization in regions of
the input space, where little or no training data are
available[11].
Fig. 10. Basic principle of operation for the ANN
A neural network is designed and trained so that it
can replace the conventional PI speed controller to get
the desired response. Network given is feed forward
with one input, three hidden and one output layer.
The hidden layers has 'log-sigmoid', 'tan-sigmoid' and
'pure line' as transfer functions. The training is done
using Levenberg–Marquardt Back-propagation
Algorithm with 20 neurons in each hidden layer using
Matlab simulink.
2915
 Journal of Multidisciplinary Engineering Science and Technology (JMEST)
ISSN: 3159-0040
Vol. 2 Issue 10, October - 2015

V. SIMULATION OF DIRECT TORQUE CONTROL

A. Model and Parameters
Figure 11 shows simulink model of DTC for induction motor drive, while Tables III. ,IV. and V. show the
parameters of the induction motor, DTC and speed controller respectively.

Fig. 11. Simulink model of DTC

TABLE III. PARAMETERS FOR INDUCTION MOTOR

Rated Rated Rated Pole Stator Stator Rotor Rotor Mutual

voltage power frequency pairs resistance inductance resistance inductance inductance
220 2238 60 2 0.435 2 0.816 2 69.3
V VA Hz ‫ــــ‬ Ω mH Ω mH mH

TABLE IV. PARAMETERS FOR DIRECT TORQUE CONTROL

Sampling Hysteresis torque Hysteresis flux Max. switching Initial flux

time band width band width frequency
20 0.5 0.01 20 0.3
µs N.m wb KHz wb

TABLE V. PARAMETERS FOR SPEED CONTROLLER

Upper limit Lower limit Sampling Cut off Ki Kp Ramp 2 Ramp 1

torque torque time frequency speed speed
17.8 -17.8 7*20 100 10 5 1800 -1800
N.m N.m µs Hz ‫ـــ‬ ‫ـــ‬ r.p.m r.p.m

The following simulation results is drawn in per unit values for a base values of 12.5 N.m for the torque , 1705
r.p.m for the speed ,0.3 wb for flux and 25A for the stator current.

www.jmest.org
JMESTN42351145 2916
 Journal of Multidisciplinary Engineering Science and Technology (JMEST)
ISSN: 3159-0040
Vol. 2 Issue 10, October - 2015

B. Results

Fig. 12. Electromagnetic torque for classical DTC

Fig. 17. Stator flux circle for modified fuzzy DTC

Fig. 16 shows Stator flux circle for classical DTC
which has a large value of ripple, while Fig. 17 shows
Fig. 13. Rotor speed for classical DTC
Stator flux circle for modified fuzzy DTC which has a
minimum value of ripple.

Fig. 14. Torque for modified classical Fuzzy DTC

Fig. 15. Torque in (p.u) for Neuro-Fuzzy DTC Fig. 18. stator current in per unit for classical DTC
Fig. 12 shows electromagnetic torque in per unit for
the classical DTC which has unaccepted ripple value
of torque (0.38 p.u.). Fig. 14 shows electromagnetic
Torque in (p.u) for modified classical Fuzzy DTC the
ripple value of torque is reduced and became 0.175
p.u. which is an accepted. Also, it is clear the fast
dynamic response of the torque and the under and
over shoots presented in the torque results . Fig. 15
shows electromagnetic torque in (p.u) for Neuro-Fuzzy
DTC, here the under and over shoots are reduced.

Fig. 19. stator current for the modified fuzzy DTC

Fig. 18 shows Stator current for classical DTC
which has a large value of ripple, while Fig. 19 shows
Stator current for modified fuzzy DTC which has a
minimum value of ripple. Figures 20-23 shows a
comparison between the conventional and proposed
methods of DTC. Swift speed response for the
proposed method is obvious.

Fig. 16. Stator flux circle for classical DTC

www.jmest.org
JMESTN42351145 2917
 Journal of Multidisciplinary Engineering Science and Technology (JMEST)
Fig. 20. stator flux ripple in per unit for
classical and fuzzy DTC
Fig. 21. Electromagnetic torque in per unit
for classical and fuzzy DTC
Fig. 22. Speed response for Classical DTC
Fig. 23. Speed response for Neuro-Fuzzy DTC
VI. CONCLUSION
This paper proposes improvement of DTC using
FLC as switching state selector based on modified
classical DTC and neural speed controller. It can seen
from the simulation results above for the simulation
model given and the parameters mentioned above that
flux, torque and current ripples is reduced remarkably
for the proposed method. The simulation results show
that the torque has very good dynamic response for
the mentioned DTC methods. Applying ANN to replace
conventional PI speed controller result in reducing the
over and under shoots and lead to swift speed
www.jmest.org
JMESTN42351145
ISSN: 3159-0040
Vol. 2 Issue 10, October - 2015
response as shown in simulation results. Therefore
using the proposed method results in improving the
motor performance.
REFERENCES
[1] 1- Thomas G. Habetler , Michele Pastorelli ,
Leon M. Tolbert " Direct Torque Control of Induction
Machines using Space Vector Modulation" IEEE
Transactions on Applications , vol. 28, No. 5,
September/October 1992 pp: 1045-1053 .
[2] 2- Alnasir, Z.A. , Almarhoon A.H. " Design of
Direct Torque Controller of Induction Motor (DTC)"
Alnasir, Z.A. et al. / International Journal of
Engineering and Technology (IJET) ,2001, pp: 54-70 .
[3] 3- Devender Kumar, Ishan Thakur, Kanika
Gupta "Direct Torque Control for Induction Motor
using Intelligent Artificial Neural Network Technique"
International Journal of Emerging Trends &
Technology in Computer Science (IJETTCS) , Volume
3, Issue 4, July-August 2014, pp:44-50.
[4] 4- M. Moghadasian , R. Kianinezhad , F. Betin
, G. A. Capolino " Torque Ripple Minimization in Direct
Torque Control of Six-Phase Induction Machines
Using Fuzzy Inference Systems " ,2010 IEEE .
[5] 5- Rajendra S. Soni , S. S. Dhamal " Direct
Torque Control of Three Phase Induction Motor using
Fuzzy Logic " Proceedings of International
Conference on Electrical, Electronics and Computer
Engineering, 25 th August 2013, Bhopal, India :34-38 .
[6] 6- Melvin Koshy , Mrs. Lekshmi A. " Direct
Torque Control Schemes for Induction Motor ",2011 .
[7] 7- Riad Toufouti, Salima Meziane, Hocine
Benalla " Direct Torque Control Strategy of Induction
Motors" Acta Electrotechnica et Informatica No. 1, Vol.
7, 2007 pp:1-7 .
[8] 8- Ahmed Hassan Adel, Amr Refky, Salama
Abo-Zaid, Mahmoud Elwany " Torque Ripple
Reduction in Direct Torque Control of Induction Motor
Drives by Improvement of the Switching Table ",
Journal of Multidisciplinary Engineering Science and
Technology, Vol. 1 Issue 5, December - 2014, pp:238-
243.
[9] 9- Daniel Graupe "Principles of Artificial
Neural Networks", Advanced Series on Circuit &
Systems -Vol. 6.
[10] 10- Neural Network Toolbox 6 , Matlab User's
Guide.
[11] 11- N.Vahdatifar, Ss.Mortazavi,
R.Kianinezhad " Neural Network Based Predictive
DTC Algorithm for Induction Motors", World Academy
of Science, Engineering and Technology 47 2010,
pp:771-775.
2918