Adaptive RBF neural network controller design for SRM drives

Abstract

In order to solve the problems of the unknown parameters variations, the external load disturbances and the torque ripple of the Switched reluctance motor drives, a combined control strategy of speed and torque is developed. Firstly, a nonlinear speed-loop controller is designed based on error compensated by adaptive radial basis function (RBF) neural network. An adaptive RBF neural network is employed to compensate the controlling errors induced by external load disturbances and parameters variations. The adaptive learning law of RBF neural network weights was developed based on Lyapunov stability theory, so that the stability of the control system can be guaranteed. Secondly, the direct instantaneous torque control method is used in the inner loop to adjust the torque directly to minimize the torque ripple. Finally, comparative studies are carried out among the proposed control scheme, fuzzy control and PI control on a 60KW-6/4 pole SRM, and the results show that the proposed control scheme has a good performance.