Auto-Tuned Two-Step Horizon FCS-MPC for a Grid-Connected CSC Inverter-based PV System

Abstract

In this paper, an optimized auto-tuning of Finite Control Set-Model Predictive Control (FCS-MPC) is proposed for a 9-level Crossover Switches Cell (CSC9) inverter. The system under study, a grid-connected single-phase CSC9 inverter-based PV system, is designed to extract maximum power from the PV panels and feed it to the grid with unity power factor and low current total harmonic distortion (THD). These objectives are achieved while regulating the CSC capacitor’s voltage at its reference value to maintain the 9 voltage levels at the inverter output terminals. In the cost function design, the minimization of the switching transitions along with the errors on the capacitor voltage and grid current (consequently current THD) were selected as control objectives, where the higher priority was given to the latter. The objective of reducing the switching events is assumed to have the least priority while the weighting factor on the capacitor voltage is dynamic and its value is determined by a pre-defined polynomial. Moreover, the optimization of the cost function is performed over two-step prediction horizon. The performance of the proposed control design (current THD enhancement and switching transitions reduction) is compared in simulation with the fixed weighting factor and one-step horizon without switching transitions constraint FCS-MPC case scenario.