Predictive Cascaded Speed and Current Control for PMSM Drives With Multi-Timescale Optimization

This paper proposes a predictive speed and current control with multi-timescale optimization in a cascade architecture for a permanent-magnet synchronous motor. Considering the difference of timescale characteristics for speed loop and current loop, different sampling times are assigned to the respective subsystem. In the prediction step of the conventional two-timescale system, the coupling between slow and fast sampling models is ignored and the output of the slow-sampling model at asynchronous sampling period is missing, which both weaken the prediction performance of the system. In this paper, the predictions of both slow and fast models for all the prediction instants are analyzed in detail. Besides, a linear estimation method based on virtual instants is proposed to improve the performance of the slow-sampling model for fast prediction instants. The data stream of the proposed method is designed based on the cascaded structure. The strategies are implemented on a field-programmable gate arrays taking advantages of parallel and pipeline processing techniques. Experimental results show that the proposed strategies have a better dynamic performance compared to the conventional method.