In order to solve the problem of integrated power and heat storage allocation, a power and heat storage configuration method of regional integrated energy system with wind power integration is proposed. Firstly, the dynamic neural network model of wind power prediction is constructed to effectively predict the wind farm output in the regional integrated energy system. Then, the charging and discharging energy behavior of power and heat storage in regional integrated energy system is analyzed, and the relationship among power storage, heat storage and regional wind farm power is established. Considering the coupling relationship between power and heat energy, a joint optimal configuration model of power and heat energy storage is built. Finally, with the actual data of a certain place is selected and three scenarios are set up, the prediction results are substituted into the optimization model to calculate the power and heat storage configuration. The comparison shows that the wind power prediction under this method has better effects, which is conducive to the optimal configuration of energy storage resources and reduce costs.