Search In this Thesis
   Search In this Thesis  
العنوان
Fault Detection in Induction Motors Using Artificial Intelligence /
المؤلف
Tolba, Mohamed Ali Hassan.
هيئة الاعداد
باحث / Mohamed Ali Hassan Tolba
مشرف / Ahmed Abd El-Tawaab Hassan
مشرف / Yehia Sayed Mohamed
مشرف / Abd El-Rahman Abd El-Rahman El-kafas.
الموضوع
Electrical engineering.
تاريخ النشر
2013.
عدد الصفحات
p 109. :
اللغة
الإنجليزية
الدرجة
ماجستير
التخصص
الهندسة الكهربائية والالكترونية
تاريخ الإجازة
1/1/2013
مكان الإجازة
جامعة المنيا - كلية الهندسه - Electrical Engineering
الفهرس
Only 14 pages are availabe for public view

from 129

from 129

Abstract

Induction Motors applications are presented in different processes in the industry. However, the induction motors, as other machines, can fault during operation, which can damage the motor. The fault conditions that motor exposed include ground faults, unbalanced supply voltage, over voltage, under voltage, reversal supply voltage, phase loss and overload. Induction motors modeling has continuously attracted the attention of researchers not only because such machines are made and used in different applications but also due to their varied modes of operation under both steady and dynamic states. In this work, the steady state motor models under balanced and unbalanced of stator supply voltages are derived using the positive and negative sequence equivalent circuits taking skin effect into account. The steady state performances of three-phase squirrel-cage induction motor taking the skin effect into account under balanced and unbalanced supply voltages are presented. The mathematical dynamic models of the induction motors are derived using d and q variables in a stationary reference frame and also the dynamic performance of the induction motors under different fault conditions. Digital protection of three-phase induction motors with a simple and reliable system using artificial neural networks (ANN’s) for monitoring and detecting external faults are proposed in this thesis. The ANNs technique are trained and tested based on the measured RMS values of stator voltages and currents and motor speed signals. In this work, the results show that the feed-forward ANN is more promising scheme in the case where fault data from induction motor is available. Also, these results will help in the development of better induction motor maintenance strategies and serve as the basis of on-line induction motor condition monitoring and diagnosis. The proposed system using “Matlab/Simulink” Software is used to simulate the induction motor under different types of external faults.