A torque model study of switched reluctance motor using BP neural network

Abstract

In the paper, a neural network torque model of a switched reluctance motor (SRM) is established, based on the merits of the backpropagation (BP) neural network in the area of modeling and control of nonlinear systems. The simulation results show that the torque model based on BP-neural network is more robust and adaptive, and can reflect the working properties of SRM more accuracy than the local linearization torque model.