Abstract: A new approach based on an improved dynamic commutation strategy and improved wavelet neural network is proposed to solve the problem of high torque ripple and the poor antijamming performance of control systems due to the structure of switched reluctance motor. A dynamic commutation strategy is designed on the basis of the direct instantaneous torque control system to reduce the turnoff phase current and increase the excitation degree of the turn-on phase current, which reduces the torque ripple of the system. In addition, the speed controller is improved on the basis of the wavelet neural network, and the output torque of the system is introduced to allow the system to respond to disturbances effectively and enhance its robustness. Simulation results show that the torque ripple of the proposed scheme can be reduced by 22.12% to 64.14% compared with the model based on traditional direct instantaneous torque control and the torque control based on a wavelet neural network, and the static error after load mutation is only 0.02r/min to 0.01r/m. Therefore, the proposed method can effectively reduce torque ripple and improve anti-interference capability.