Design of the MPC Weight Factors for a Grid-connected Renewable Energy Converter

Model Predictive Control (MPC) combines the characteristics of several advanced control strategies and has gained space in power converter applications, such as grid integration of renewable energy sources. The tuning of the MPC controller with modulation has many degrees of freedom, which makes its systematic tuning an open scientific challenge. This paper applies an MPC with modulation to regulate the current in a grid-connected converter with an inductive or L-filter following a model in the synchronous frame. The proposed controller tuning is based on the closed-loop poles of the MPC control system. The degrees of freedom were reduced to only four parameters: cost function weight factors and predictive and control time horizons. First, the control parameters are constrained to guarantee stability and then to estimate the theoretical dynamic performance of the control system. Simulations enable the MPC to be adjusted for better dynamic performance of the grid-connected converter. Experimental testing using a 1 kW workstation supports the proposed controller tuning. The designed controller shows improved power control and energy quality compared to the Finite Control Set predictive controller.