This paper describes a three-stage model predictive torque control method with sequential structure. The prediction trajectory of the torque and flux-linkage are extrapolated to two control instants. The first-stage selects alternative voltage vectors from all basic voltage vectors according to the torque prediction performance at the next one control instant, and the second-stage further reduces the range of candidate voltage vector by considering the whole prediction performance of the torque at next two control instants. Furthermore, the flux-linkage prediction performance at two control instants is evaluated in the third-stage to select an optimal vector voltage. Since three-stage prediction of the proposed method considers the comprehensive control performance of torque and flux-linkage at two control instants, the probability of adjacent periods with the same vector increases. Therefore, the switching frequency reduction of the inverter and improvement of the steady-state control performance are attained simultaneously in the proposed method. Ultimately, it is proved by comparative experiments.
Three-stage Series Model Predictive Torque Control for PMSM Drives
2022-10-28
672980 byte
Conference paper
Electronic Resource
English
Deadbeat Model Predictive Torque Control with Discrete Space Vector Modulation for PMSM Drives.
BASE | 2017
|Modulated Model Predictive Speed Control for PMSM Drives
IEEE | 2018
|