PERFORMANCE EVALUATION OF A NEURO CONTROLLER FOR BRUSHLESS DC MOTOR DRIVE

Abstract

In this dissertation work, an online control strategy is proposed for non-linear brushless dc motor operating in high performance drive environment. The proposed control strategy posses the capability of - simultaneous online identification and control. The control action is implemented in such a way that system trajectories follow a predetermined reference track. The performance of PI controller is also investigated for different Kp and Ki parameters. The performance of Neuro controller for different number of neurons in hidden layer is investigated. Finally a comparison of the performance of Neuro controller with the performance of PI controller is given. Simulation results illustrated that a Neuro controller used in conjunction with adaptive control scheme can result in a flexible control device. A detailed literature survey is reported in the proposed work area in chapter-1. A brief description of artificial neural network (ANN) is given in chapter-2. The modelling and general description of brushless de motor drive is given in chapter-3. A general description of the Neuro controller and PI controller are given in chapter-3. Chapter-5 presents the flow chart for the computer simulation for both the controllers. The simulation results are presented in chapter-6. Chapter-7 gives the overall conclusion of the developed drive system.