Torque Ripple Intelligent Minimization of Switched Reluctance Motor

The torque ripple of switched reluctance motor is a crucial problem in its application fieled.This text uses the adaptive fuzzy-neural networks to the learning of its static torque inverse model off line based on the torque-angle characteristic of a 6/4 structure motor.Then according to the torque distribution function,each phase torque was distributed,and optimal was real-time gained by fuzzy-neural networks on line.At the same time,current hysteretic comparator was utilized to track the current profile,which results in low ripple torque control of switched reluctance motor.The simulation result proves validity of this method.