Artificial Neural Network for the Optimal Control of an AC Chopper Fed Induction Motor

Abstract

A novel concept of application of artificial neural networks for the optimal control of an ac chopper fed induction motor is presented. The speed of the induction motor is controlled by controlling the switching angles of the ac chopper. The idea proposed in this paper is to use artificial neural network (ANN) as a non-linear signal converter, which accepts the reference value of the modulation index as input and outputs the values of the primary switching angles. The use of neural network as a controller makes it possible to vary the switching angles optimally such that the selected dominant lower order harmonics are eliminated and high quality output is obtained. Hence, this contributes to the improvement in overall system performance. The ANN based drive system is simulated using SIMULINK toolbox in MATLAB package and performance observed. The superiority of the neural network based controller over the conventional analog and microprocessor based controller is also discussed.