Torque Ripple Suppression of Switched Reluctance Motor Based on Torque Sharing Strategy

The torque ripple is the main disadvantage of switched reluctance motor (SRM). In order to reduce the torque ripple of SRM, and improve the performance of the system, the torque sharing strategy was combined with RBF neural network for the purpose of torque ripple suppression by controlling the winding current of each phase. In consideration of the possible error of network, the real-time current compensation was taken to compensate the loss of torque which could suppress the torque ripple of the system in further. The results show that he torque ripple of SRM was suppressed effectively.