Torque Ripple Minimization in Switched Reluctance Motor Based on BP Neural Network

Abstract

The instantaneous torque control for torque ripple minimization of switched reluctance motor (SRM) by BP neural network is presented. As SRM has a highly nonlinear characteristics, neural network is well suited for its control. After static torque characteristics of SRM having been measured, the torque model and the inverse torque model are developed based on BP neural network of Levenberg-Marquardt algorithm. The torque ripple minimization can be achieved by optimum profiling of the phase current based on instantaneous torque control. An efficient commutation strategy for minimizing torque ripple as well as avoiding power converter voltage saturation over a wide speed range of operation is proposed. Simulation results verify the feasibility of this torque ripple minimization technique.