A Single-Objective FCS-MPC Method for Three-Level APF

Abstract

Finite control set model predictive control (FCS-MPC) scheme is applied to active power filter (APF), which has some advantages such as fast dynamic response, needlessness of modulation model. However, some problems exist such as large calculation and difficult selection for weighting factors. In this paper, a single-objective FCS-MPC method is proposed for neutral-point-clamped (NPC) three-level APF. The mixed logic dynamic (MLD) model of APF is built as predictive model. Then based on the traditional FCS-MPC scheme, the proposed method optimized the multiple-objectives by one cost function item without using weighting factors. This method decreases the amount of calculation caused by setting weighting factors and calculating cost function items as well as the number of switch states required to calculate. Finally, simulation and experiment results verify the feasibility and appropriate performance of the proposed method.