An Improved Parallel Predictive Torque Control for Permanent Magnet Synchronous Motor

Abstract

A predictive torque control strategy with parallel structure (P-PTC) was recently proposed to overcome the tedious tuning of weighting factor in the conventional predictive torque control (C-PTC). However, P-PTC still has some problems, such as the introduction of other parameters to be tuned and the lack of theoretical support, et al. To solve the above problems, an improved P-PTC without any extra parameters tuning (IP-PTC) is proposed for permanent magnet synchronous motor (PMSM). Firstly, two parallel cost functions are respectively constructed to select three voltage vectors that have the best control effect on electromagnetic torque and stator flux to form two new finite control sets (FCSs). Secondly, three different cases are divided according to the intersection number of these two FSCs. Finally, the corresponding optimal voltage vector selection scheme is proposed for these three cases. C-PTC, P-PTC and the proposed IP-PTC are studied through theory and experiments. The results show that when compared with C-PTC and P-PTC, the proposed IP-PTC has neither weighting factor nor any extra parameters, and it is well supported by theory. Furthermore, the proposed IP-PTC has the best static and dynamic performance among the three.