A Novel Torque Boundary-Based Model Predictive Torque Control for PMSM Without Weighting Factor

In this article, a novel torque boundary-based model predictive torque control (MPTC) is proposed to improve the torque and flux performance in permanent magnet synchronous motor (PMSM) drives by straightforwardly reducing the corresponding ripples. For this purpose, an optimal sequential list of two voltage vectors is considered as candidate input and the time durations of the two vectors to be applied are determined based on a predefined torque ripple tolerance band. Considering the torque outcomes both at the switching instant and at the end of the control period, the torque can be limited within the upper and lower boundaries of the preset tolerance band during the whole control period. Meanwhile, by monitoring the number of defined valid vector sequences, the torque boundaries can be optimized online. As a result, the torque ripple can be restrained to a fairly small range, and the torque boundary design work is avoided. In addition, according to the predicted torque outcomes of the candidate inputs, a group of vector sequences can be excluded from the control set before optimization, reducing the computation cost. Moreover, due to the preset torque ripple tolerance, the weighting factor in the cost function is eliminated, thereby avoiding the corresponding tuning work. Experimental results are presented to reveal the effectiveness of the proposed strategy.