An Improved Model Predictive Control for Three-level inverter

Model predictive control (MPC) has been gaining significant attention due to its ease to consider multiple nonlinear constraints and outstanding dynamic response. However, finite control set mode predictive control (FCS-MPC) suffers from huge demand of computing resources and cumbersome tuning of weighting factors, which are obstacle for its application. To resolve this contradiction, this paper presents an effective method to carefully select the voltage vectors aiming to simplify the traditional model predictive current control (MPCC) and make weighting factor tuning got simplified. Firstly, a voltage prediction model is used to replace the traditional current prediction model by equivalent changes. Then, a new cost function and fast optimization method based on the geometric characteristics of the space vector of the output voltage of the three-level NPC inverter is proposed. Finally, the proposed control strategy is verified on the hardware-in-the-loop simulation platform that it can greatly improve the computing efficiency of MPCC while maintaining the current tracking performance with good dynamic and static performance.