Model Predictive Control for Space Vector Modulation of a Three-Level ANPC Inverter for Efficient Loss Distribution and Neutral Point Balancing

Abstract

This paper presents an efficient loss distribution and Neutral Point (NP) balancing method for a grid-connected Three-Phase Three-Level Active Neutral Point Clamped (3L-ANPC) inverter. A Finite Control Set Model Predictive Control (FCS-MPC) technique is used to select the optimum redundant active voltage vectors and redundant zero voltage vectors during the Space Vector Modulation (SVM). A direct discrete loss model is designed for the IGBT switches of the 3L-ANPC inverter, to calculate the junction temperature T <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">j</sub> of the IGBT switches for each sampling point, using information about conduction and switching losses, and thermal properties of the housing and cooling. To evaluate the performance of FCS-MPC, a T <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">j</sub> comparison is held out with a reference from literature. This reference utilizes a decision table based on an energy loss model. Furthermore, the performance of the FCS-MPC technique for loss distribution is analyzed during steady state. The FCS-MPC technique for NP balancing is evaluated during steady and transient states.