Abstract: Numerous problems exist in operating a microgrid with a hybrid energy storage system（HESS）. These include excessive gridconnected power fluctuations and low dispatch efficiency. These issues are induced by ignoring the fluctuations in photovoltaic/wind power output and the dynamic changes in the state of charge of the hybrid energy storage. To address these issues,this study proposed a two-layer microgrid control method taking into account the initial state of charge of the HESS. Specifically,the photovoltaic/wind power output in the microgrid was optimized in the upper layer with the predictive control model,and the microgrid trace dispatch model was established. The initial state of charge on the HESS dispatch day was determined based on the dispatch model in the upper layer,and the capacities of the storage battery and the super-capacitor in the HESS were distributed with the dynamic planning algorithm.Moreover,the change in the state of charge was fed back to the predictive control model in the upper layer for optimizing the gridconnected power interaction. Next,the simulation was performed on the system focusing on minimizing both grid-connected power fluctuations and the state-of-charge deviation of the HESS. The study shows that the proposed control strategy displays desirable gridconnected suppression performance and achieves favorable trace dispatch. The proposed strategy consideres the charge and discharge characteristics of different energy storage elements to achieve coordinated power dispatch in the HESS.