Abstract: As the number of microgrids continues to increase， the frequent energy interaction between the microgrid and the distribution network will have an adverse impact on the operation of the distribution network. In addition， the gradual increase in penetration rate of wind power leads to a large amount of wind power entering the distribution network， which also imposes a serious burden on the operation of the distribution network. In order to reduce the unnecessary energy interaction between the microgrid and the distribution network， promote the local consumption of wind power， this paper constructs an energy transaction mode for multi-microgrid and distribution network that considers wind power uncertainty. The optimization goal of this mode is to maximize the revenue of the system operation entity and minimize the total cost of power consumption of the multi-microgrid. This mode integrates the load and wind power output in the area through operation entity， and divides the internal peak-to-valley time of the system to improve the utilization of wind power. In order to enhance the ability to cope with the uncertainty of wind power， the transaction mode uses the load transfer rate to describe the user response of TOU prices and establish a model for the TOU price of regional multi-microgrid systems considering wind power uncertainty. Finally， the multi-objective optimization problem in the model is solved by the NSGA-II algorithm combined with the Monte Carlo method. The example proves that the proposed transaction mode is beneficial to reduce the cost of electricity used in microgrids， improve the overall benefits of the multi-microgrid， and realize the friendly interaction between the multi-microgrid system and the distribution network.

Key words: microgrid, regional multi-microgrid system, wind power accommodation, time-of-use （TOU） prices, elitist nondominated sorting genetic algorithm（NSGA-Ⅱ）