NEURAL CONTROL OF ELECTRICAL SYSTEMS

Abstract

Adaptive control for high perfo1•mance drive systems has recently become an important subject of research. In applications such as robotics, aircraft, the rotor of the electric motor should follow a pre­ selected track at all time. Parameter uncertainties, unknown load variation, or sudden external disturbances should not affect the tracking accuracy. A neural control technique is proposed for identification and controi of electric motors. The main feature of the proposed controller is a neural network, which performs two functions. The first is to identify the nonlinear system dynamics at all times. Hence, detailed and elaborated models for the electric motors are not needed. Furthermore, unknown nonlinear dynamics which are difficult to model such as load disturbances, system noise and parameter variation can be recognized by the trained neural network.