An Iterative Learning Based Direct Torque Control Strategy of DC-Biased Vernier Reluctance Machines for Torque Ripple Reduction
Zimin Li, Wubin Kong, Yongjun Cheng, Ronghai Qu
Abstract
This article proposes an iterative learning-based direct torque control (IL-DTC) strategy for dc-biased Vernier reluctance machines (dc-VRMs) to effectively improve the torque control accuracy. Firstly, the model for dc-VRMs in stator flux synchronous rotating reference frame is proposed for DTC, considering the harmonic currents and saturation of self-inductance. Based on the model, torque ripple is implied to be suppressed by injecting third zero-sequence current (ZSC). Then, a simplified iterative learning controller (ILC) with varying gain coefficient is designed for dc-VRMs to generate the desired ZSC online. The convergence and parameter design principle of the ILC are analyzed elaborately to ensure the control accuracy. Finally, comparative experiments with conventional proportional-integral-based DTC scheme and torque ripple suppression scheme are carried out to verify the superior torque control accuracy of the proposed IL-DTC scheme.