Enhanced FS-PTC: Dynamic Weighting Factors for Optimal Flux and Torque Control

Abstract

Recently, Finite State-Predictive Torque Control (FS-PTC) of induction motors has gained significant interest in high-performance motor drive applications. The effectiveness of FS-PTC relies on the successful minimization of a cost function achieved by selecting an appropriate voltage vector. Typically, the cost function for FS-PTC is composed of errors between the predicted and reference values of torque and flux; hence a weighting factor, \mathbit{\lambda}, is normally employed to establish different priorities between torque and flux. However, determining the optimal is a complex undertaking, since an incorrect or suboptimal choice can needlessly compromise torque or flux responses. This paper introduces an online tuning approach for the weighting factor, based on the dynamic change of flux error. Instead of using a fixed value, the weighting factor is dynamically adjusted using a simple P or PI controller. The proposed method's performance is evaluated in this paper, considering various configurations of the controller's settings. Simulation results demonstrate that the proposed technique enhances the overall torque and flux ripples across a broad range of operating speeds, surpassing the performance of the fixed value weighting factor technique.