Variable Vector Model Predictive Control for Permanent Magnet Synchronous Motors Using a Single DC-Link Current Sensor

In this article, a variable vector model predictive control (VV-MPC) using a single dc-link current sensor is proposed for permanent magnet synchronous motor (PMSM) drives. In VV-MPC, the linear modulation region is divided into several small regions corresponding to different predictive expected voltage vector synthesis methods. Then, the durations of vectors are flexibly allocated to reconstruct the current successfully. Based on the modulation characteristic of MPC, to improve the accuracy of reconstructed current, the vector action sequence is flexibly allocated and the current sampling moments are properly set, with the reconstructed error compensating instantaneously. Compared with conventional MPC methods, VV-MPC has better steady and dynamic performance. Besides, VV-MPC has a relatively low computation burden due to full consideration of the computation complexity. Theoretical analysis and experimental results verify the effectiveness of VV-MPC.