Simulation Research on no Position Control Strategy of Switched Reluctance Motor Based on On-Line Training of BP Neural Network

For the rotor position estimation model of switched reluctance motor position free control system, This paper presents an online training rotor position identification method based on BP neural network. By generating the M program of neural network, dynamically extracting the training samples, batch training the neural network, and finally obtain the accurate estimation of rotor position and angle. The simulation results show that under different working conditions, the error between the actual rotor position and the estimated rotor position of the system is small, the estimated rotor position angle accuracy is high, and has good dynamic response ability, so as to realize the accurate control of switched reluctance without position.