Direct Model Predictive Control with Constrained Power Losses for Grid-Tied Converters with LCL Filters

This paper proposes a direct model predictive control (MPC) method that improves the efficiency of grid-tied converters with LCL filters, while minimizing the grid current distortions. To do so, the MPC problem not only accounts for the grid current reference tracking minimization, but it also limits the converter power losses by introducing explicit constraints on them. To this aim, different problem formulations are examined to limit the losses. Specifically, the FCS-MPC problem is designed such that it either limits the losses per semiconductor device or the three-phase converter losses. The presented results verify the effectiveness of the proposed strategy and demonstrate the performance benefits compared to conventional control and modulation methods, such as voltage-oriented control (VOC) with space vector modulation (SVM) or discontinuous pulse width modulation (DPWM).