Performance analysis of single phase induction motor using neural network predictive controller and pi controllers

Abstract

In recent years the industrial processes are completely automated due to the detonation in the areas including integrated circuit technologies the invention of power semiconductor switches advent of excellent computers and the development of sophisticated algorithms to provide better co ordination among them To the fullest utility of rapid technological development the grubby gasoline machineries and direct current machineries are quickly replaced with induction motors For the cost effective operation tight control of machine speed is essential but it is difficult to control the machineries due to its inherent nonlinear nature and interaction complexity To accomplish strict control an unremitting supervision of the machine speed and implying the requisite manipulations at its input are indispensable To provide an accurate manipulation the control law that are used to the motor must be accurately defined here the prime intention is a design of superior controller to attain better process characteristics such as enhanced time domain specifications disturbance signal rejection expert set point tracking effective load and parameter change handling capabilities Artificial neural network technique is a promising tool to control such nonlinear systems as they can capture inherent nonlinearity of the process with the use of an explicit model newline