Deadbeat Indirect Torque Control of Switched Reluctance Motors with Current Vector Decomposition

Abstract

This article presents an improved deadbeat indirect torque control (ITC) method for switched reluctance motors (SRMs) with the primary goal of reducing torque ripple. The proposed control approach comprises two parts: a torque-to-current conversion scheme that the proposed method achieves excellent current and a deadbeat controller (DBC). In the conversion scheme, a second-order SRM Fourier-series model is constructed by integrating the current vector decomposition method. Subsequently, an iterative learning controller (ILC) is designed based on this model to achieve precise conversion from the electromagnetic torque to the q-axis current, which eliminates the need for additional modeling processes. Within the proposed DBC controller, a novel recursive least squares (RLS) estimator is introduced to effectively tackle the issue of model variations. This integration enables the adaptive calibration of the predictive model, ultimately guaranteeing optimal performance in the current control. Furthermore, the consistency of the model employed in both the DBC and conversion scheme empowers the RLS to further refine the accuracy of torque-to-current conversion, thereby improving torque ripple suppression performance. Comparative experiments are conducted on a 12/8 SRM to evaluate the proposed control method’s performance. The experimental results show that the proposed method achieves excellent current tracking and torque ripple suppression performance in SRM drives.