Model Predictive Torque Control Using Modified TSF and Parameter Adjustment of SRMs for Torque Ripple Reduction

Abstract

This article introduces a model predictive torque control (MPTC) scheme using a modified hybrid torque sharing function (TSF) and the adaptive adjustment regarding parameters for switched reluctance motors (SRMs) to mitigate the torque ripple effect and enhance dynamic performance. Compared with classic TSF control, torque model prediction is adopted in MPTC instead of hysteresis control to reduce torque ripple. First, an enhanced hybrid TSF is designed for torque ripple suppression, where the rising reference phase torque can be adjusted in real-time according to the actual phase torque in the demagnetizing interval. Second, adaptive adjustment of angles based on the approximate model and the look-up table is utilized to further minimize torque ripple under multiple operating conditions. Then, the linear active disturbance rejection control with the adaptive bandwidth parameter is applied to the speed controller for better dynamic performance under multiple operating conditions. Eventually, experiments on a 12/8 SRM are conducted to ascertain the efficacy of the proposed MTPC by comparison.