Minimization of torque ripple in switched reluctance motor based on MPC and TSF

Abstract

In this paper, a novel control method based on model predictive control (MPC) and torque sharing function (TSF) is addressed to minimize the torque ripple of switched reluctance motor (SRM). Firstly, an accurate SRM model is established based on flux‐linkage characteristic curves obtained from the locked rotor test, which can predict the future operation state of the SRM drive system. Secondly, an improved TSF curve is proposed, and genetic algorithm is used to optimize TSF parameters to reduce torque ripple in commutation region. Besides, the MPC is integrated into the TSF control framework to replace the traditional hysteresis controller, and establishes a new TSF‐based model predicted torque control (MPTC) method, which avoids the frequency conversion problem caused by torque or current hysteresis controller. Further, a sector division scheme is addressed to decrease the number of candidate voltage states, whereby the computation of the controller in each sampling period is effectively reduced. Finally, in order to verify the correctness of the MPTC method, it is compared with the traditional direct instantaneous torque control (DITC). Both simulations and experiments on a three‐phase 12/8 pole SRM have confirmed that MPTC scheme can not only decreases torque ripple and stator copper losses, but also has higher efficiency than DITC. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.