A two-layer optimization model of a leader-follower game is constructed for peer-to-peer electricity trading between distributed energy power stations and industrial parks.The profit maximization of the distributed energy station is taken as the goal in the upper layer，with decision variables including tariff time period division，tariff pricing，and the charging/discharging power of on-site energy storage equipment.The difference between electricity supply and demand is used as the state variable，and a certain range of difference is used to determine the valley and peak price periods.A tariff response coefficient is designed，which depends on the demand response effect of the power load elasticity matrix at each period，thereby determining the tariff for each time period.The highest production schedule matching degree and the cost-effectiveness of electricity procurement in industrial parks are taken as the goal in the lower layer，with the load adjustment for demand response as the decision variable.The production scheme matching degree is defined as the degree of consistency between the power variation at each moment of the industrial park after participating in the demand response and the original production scheme.The multi-objective problem at the lower level is solved by the augmented ε-constraint method to obtain the Pareto optimal solution set.The example results show that methods of tariff time period division and tariff pricing in peer-to-peer electricity trading can improve the profit of distributed energy power stations，reduce the cost of electricity for users，effectively mitigate wind and solar curtailment，and motivate the trading entities to participate in peer-to-peer electricity trading.