Comparison of Finite Control Set Model Predictive Control Methods for Five-Level NNPC Inverter

Abstract

Based on the topology of five level nested neutral point clamped (NNPC) grid connected inverter, this paper compares two type of finite control set model predictive control (FCS-MPC) methods to reduce the amount of calculation time compare with the traditional FCS-MPC. The first method is based on space vector modulation approach which uses 60- degree voltage state vector for the prediction to reduce the calculation time of each control period in FCS-MPC algorithm. While the second method is to simplify the calculation time by using the FCS-MPC based on adjacent state vector method. Through comparative analysis, it can be seen that the 60 degrees sector FCS-MPC have similar performances regarding the current THD; the upper and lower DC voltage deviation of DC- link capacitors. The main difference is on the computation time which results to 35 control periods compare to 125 for the traditional FCS-MPC. The adjacent state vector method results to 7-29 control period.