Three-Vector-Based Optimizing Selection Model Predictive Direct Dower Control for DFIM

This article introduces a three-vector based optimizing selection model predictive direct power control (TVOS-MPDPC) method is proposed for a doubly-fed induction motor (DFIM) in the all-electric ship application with fast determining duty cycle. In the proposed method, the voltage and flux equations are redesigned to calculate the inverse of the stator-side complex power's phase. In this way, three optimal voltage vectors could be selected to eliminate power errors completely. As a result, only one prediction is necessary and seteting against the established finite-control-set model predictive control approach, the prediction time is reduced. The vector relationship diagram constructed by the selected voltage vector can be leveraged for calculate the action time and duty cycle of each voltage. The simulation results indicate the TVOS-MPDPC can greatly shorten time of forecast and significantly improve the steady-state and dynamic performance.