A fuzzy iterative approach for determination of current waveform for switched reluctance motors using a torque sharing function at positive and negative torque production regions

This paper deals with a fundamental control issue in switched reluctance motor (SRM) drives - the torque ripples. Normally, torque ripple minimization is achieved by using a look-up table, i.e., the look-up table uses stored magnetic characteristics to provide the reference current for some specified torque. A number of techniques for the generation of reference current profiles that minimize the torque ripples have been suggested. But, because of highly nonlinear characteristics of the SRM, all these schemes are not fully successful. Our work deals with a novel algorithm for generation of current waveforms by modulation of the reference current pattern of the two-phase-on scheme with torque sharing functions at both positive and negative torque production regions by means of a multiplying factor for minimizing the torque ripple. This multiplying factor is computed with the aid of fuzzy logic, which is well suited to compensate for the nonlinearities of the system. Performance of the proposed strategy is verified by simulation.