A SLIDING MODE POSITION CONTROL OF A SYNCHRONOUS MACHINE USING NEURAL NETWORK DISTURBANCE OBSERVER

In this paper, we propose a sliding mode technique to control the field-oriented synchronous machine. The sliding mode controller is designed for a class of non linear dynamic systems to tackle the problems with model uncertainties, parameter fluctuations and external disturbances. Our aim is to make the position control robust to parameter variations. The use of the nonlinear sliding mode method provides very good performance for motor operation and robustness of the control law despite the external/internal perturbations. An observer is considered to overcome the problem of torque disturbance. A load torque observer is designed based on neural network technique without affecting the overall system response. Simulation results are given to highlight the performance of the proposed control echnique under load disturbances and parameter uncertainties.