In this paper, an adaptive torque sharing function (TSF) control strategy of switched reluctance motor(SRM) based on neural network is proposed, which has the advantage of using BP neural network to automatically adjust the overlap angle online according to different speeds. This paper introduces the driving system and mathematical model of SRM, expounds the conventional TSF control strategy and its disadvantages caused by fixed overlap angle, and explains the reason why this disadvantage causes large torque ripple. According to this shortcoming, an adaptive TSF control strategy is proposed. This adaptive control strategy uses back propagation (BP) neural network to establish the nonlinear relationship between the speed and the overlap angle, and automatically adjusts the overlap angle according to the speed. Compared with the conventional control method, this control strategy significantly reduces the torque ripple.
 

SRM; TSF; BP neural network; torque ripple;