Variable Action Period Predictive Flux Control Strategy for Permanent Magnet Synchronous Machines

Abstract

This article introduces a method to achieve the variable action period (VAP) of voltage vector in the finite control set model predictive control (FCS-MPC). In the conventional FCS-MPC, the control behavior is limited since the action period of a voltage vector is normally fixed to the sampling period. This article aims to enhance the control performances by means of regulating the action period of selected voltage vectors, together with the selection of an optimal control action. A VAP-based predictive flux control strategy is thus proposed. The optimal voltage vector and its optimal action period are both obtained by means of a two-step cost function minimization. In this manner, the future states of the control objectives are taken into account in the construction of cost function. The action period for the selected voltage vectors is no longer fixed to the sampling period, which provides an additional control freedom to obtain improved reference tracking behaviors. Experimental validations and comparison studies demonstrate that enhanced steady-state performances as well as excellent dynamic behaviors can both be obtained with the proposed strategy adopted.