Abstract

Finite Control-Set Model Predictive Control (FCS-MPC) has become widely used in power electronics due to its conceptual simplicity and ability to include multiple control objectives. However, FCS-MPC presents some challenges, specifically variable switching frequency and high current ripple due to the application of a single vector per state. On the other hand, the Optimal Switching Sequence Model Predictive Control (OSS-MPC) technique addresses this problem, but generates high computational costs. This article proposes a Virtual Vector OSS-MPC (VV-OSS-MPC) for reduced computational cost. VV-OSS-MPC can determine the optimal switching between all possible options without evaluating all sectors, thereby reducing the computational burden. The proposed strategy presents almost identical performance to OSS-MPC, measured by Total Harmonic Distortion (THD) and dynamic response time. The results are experimentally validated on a two-level inverter with RL load, obtaining an execution time reduction of 50% compared with OSS-MPC.