Model predictive Control of a Double Stage AC-DC Converter for Grid-Interface of Vanadium Flow Batteries

This paper presents a multi-objective model predictive control algorithm to implement the control of a bidirectional double stage ac-dc power conversion system, used to interface a Vanadium redox flow battery with a power grid. The converter topology is based on two stages conversion, consisting in a dual active bridge, controlled using the phase shift modulation, in cascade with a grid-connected inverter/rectifier. In order to control the charge and discharge of the battery, the dc-link and the grid current, a multi-objective predictive control is introduced embedding the control of both conversion stages in a single control action. The computed steady state and dynamic performance are compared to what can be achieved with standard controllers. The proposed algorithm is tested in simulation using a model developed with the PLECS block library on Matlab/Simulink.