Aiming at the problems of low efficiency and irregular management process due to the uncertainty and stochasticity of renewable energy sources when managing the energy of a hybrid distribution grid with distributed photovoltaic, an optimization method based on two-stage optimization is proposed for energy storage of a hybrid distribution grid with distributed photovoltaic. In the first stage, the day-ahead optimization is implemented based on the predicted output power of wind turbines and photovoltaic, predicted electricity and hydrogen demand, time-price variation information, and physical and economic constraints of system operation, aiming to optimize the resource allocation and minimize the total operating cost of the system. In the second stage, the intraday adjustment model based on backward horizon optimization is introduced to respond to and adjust the energy storage strategy, which effectively compensates for the power fluctuations caused by prediction errors. The experimental results show that the proposed method reduces the operating cost by 13.0% and the deviation by 16.2% compared with the grid-connected energy storage method. The proposed method can improve the efficiency of energy storage in hybrid distribution grids with distributed photovoltaic and has certain practical application value.