To solve the problem of low reliability caused by typical wind power forecasting provided in the form of ‘power points’

an optimized control strategy for energy storage charging and discharging decisions based on ‘wind power interval prediction’ is proposed. Firstly

a wind power interval prediction model based on the dual attention mechanism and the Huber quantile loss function （DA-BiLSTM-QH） is established. Secondly

by integrating interval prediction information

microgrid load

energy storage status and real-time electricity price

a multi-objective optimization model of wind-storage microgrid system is established

and a solution strategy based on deep reinforcement learning DQN algorithm is designed. The effectiveness of the proposed strategy is verified through simulation of typical operating conditions.