Slid Mode Control of Switch Reluctance Motor Based on Torque Inverse Model

The torque ripple of switched reluctance motor is a crucial problem in its application. The non-linear reflection of position signal of switched reluctance motor is built up by wavelet neural networks and the rotor position angle is estimated. In addition, in this paper the method is discussed that the torque inverse model of switched reluctance motor is trained by the adaptive-neural networks to optimize the required phase current of expected torque and the low ripple torque control is realized by slide mode current controller. The simulation result proves the validity of this method to effectively control the torque changing of switched reluctance motor.