The energy storage has the functions of reducing peak and filling valley, improving voltage quality, etc. It plays a great role in the power distribution network. This paper studies and formulates the operation strategy of battery energy storage system, then uses it to guide the distribution network storage allocation problem. By maximizing the comprehensive economic benefits as the upper planning objective, the storage capacity is controlled variable, considering the factors such as the investment cost, operation maintenance cost, saving energy loss, saving cost of energy low price charging high price discharge and so on. According to certain index a battery charge-discharge strategy is proposed by heuristic method. This strategy is based on the equivalent load curve that distributed generation output and daily load overlay and considers the sub time electricity price, the strategy formulation as the lower part of the planning, thus two- layer planning model is established. A genetic algorithm with embedded heuristic method is used to solve the model. An improved IEEE 33-bus system is analyzed to verify the effectiveness of the proposed model, and the proposed charging-discharging strategy is compared with the existing typical charging-discharging strategy, which shows the superiority of the strategy in this paper.