Double-Vector-Based Sequential Model Predictive Torque Control of Induction Motor

Abstract

Sequential model predictive torque control (SMPTC) has attracted wide attention because the weighting factor is eliminated. However, since only one voltage vector is applied in each control period, the torque ripple is large. To solve this problem, this paper proposes a double-vector-based sequential model predictive torque control (DSMPTC). In DSMPTC, when there are new voltage vectors among the combined voltage vectors, two voltage vectors are applied in one control period, otherwise, only one voltage vector is applied in one control period. The durations of voltage vectors are determined by torque deadbeat theory. The number of new voltage vectors is determined with a geometric method. The influence of the combined voltage vector sequence on the torque ripple and switching frequency is discussed. The simulation is carried out to verify the correctness of the proposed methods and conclusions. The simulation results show that the dynamic performance of DSMPTC is generally the same as SMPTC, while the steady-state performance of DSMPTC is greatly improved compared with SMPTC.