Research on control strategy of dual active full bridge converter based on deep deterministic strategy gradient

Abstract

In order to cope with the need for fast and adaptive voltage regulation in the face of large-scale disturbance in future smart grid EV grid-connected converters, a control strategy based on deep reinforcement learning is proposed in this paper. For electric vehicle grid-connected dual active bridge (DAB) DC–DC converter, a deep deterministic strategy gradient (DDPG) reinforcement learning intelligent controller based on an actor-critic architecture is designed. Through real-time online learning and training by collecting a large amount of system data, the intelligent controller can automatically adjust the power parameters of the DAB converter to ensure that the DC converter has strict stability in the face of various system disturbances. Simulation results show that the control strategy is effective. This research provides an innovative and feasible solution, which has important practical application value to the voltage regulation of electric vehicle grid-connected converters.