Search In this Thesis
   Search In this Thesis  
العنوان
Proportional integral control for electromechanical system using neural network /
المؤلف
Sallam, Nada Mohamed Abd Elmonem Mahmoud.
هيئة الاعداد
باحث / ندى محمد عبد المنعم محمود سالم
مشرف / صبري فؤاد سراية سراية
مشرف / محمد شريف مصطفى القصاصي
مشرف / محمد معوض عبده عبد السالم
الموضوع
Artificial Neural Networks. Neural Networks. Electromechanical system - Network.
تاريخ النشر
2019.
عدد الصفحات
online resource (84 pages) :
اللغة
الإنجليزية
الدرجة
ماجستير
التخصص
هندسة النظم والتحكم
تاريخ الإجازة
1/1/2019
مكان الإجازة
جامعة المنصورة - كلية الهندسة - هندسة الحاسبات ونظم التحكم
الفهرس
Only 14 pages are availabe for public view

from 109

from 109

Abstract

Different types of electric motors are used for the different applications like manufacturing, home appliances and many more in past decades. In all such electric motors e.g. universal motors, DC motors, induction motors (IMs), synchronous motors etc., IMs are most widely used because of their property of self-start i.e. they do not require any external force for running. Induction motors have several basic characteristics like robustness, low maintenance, high efficiency, reliability, high starting torque which makes IM a desirable machine for applications. Induction motor is dynamic nonlinear system with uncertainty in the machine parameters. The aim of this thesis is to control speed of induction motor based on Proportional Integral – Artificial Neural Network controller and achieve high dynamic performance to overcome the drawbacks in conventional controller, using neural network (NN) controller, based on Back Propagation (BP) for training a given dataset. The effectiveness of the controller was achieved by applying external noise such as changing the reference speed signal as well as reducing the load torque on the motor. The controller was designed by MATLAB Simulink. MATLAB/SIMULINK software is used to develop a three phase induction motor model. Also the performances of the two controllers PI and ANN have been verified. It is found that the usage of PI-ANN controller achieved better performance and fast response for the motor speed.