—This article presents a novel torque ripple minimization approach based on the genetic algorithm (GA) for the switched reluctance motor (SRM) drives. The fitness function of the GA is designed on the basis of two main optimization criteria. The first optimization criteria; three objectives are defined to optimize commutation angles minimizing torque ripple and copper loss. The second optimization criteria; an objective is defined to eliminate the negative torque which decreases the average torque in SRM drives. The achievement of all objectives depends on the adjusting appropriate the commutation angles of the torque sharing functions (TSF) during the commutation period. All of conventional TSFs are dependent on three different commutation angles, namely turn-on, turn-off, and overlapping. Due to the nonlinear phase inductances of SRM, the delay of current rising and falling time are not unity in overlapping commutation region. To overcome the separation of the incoming and outgoing phase currents during the commutation region, rise angle, and fall angle are used instead of overlapping angle. GA is used to optimize the commutation angles of conventional (sinusoidal) TSF and improved (sinusoidal) TSF. At the same time, the elimination negative torque effects on torque ripple, average torque, and copper loss is investigated.
Read full abstract