A Study of Trading Strategies for Distribution Systems Containing Multiple Microgrids Based on Three-layer Games

With the large-scale access of multi-microgrids to the distribution network, the shape of the distribution network will develop in the direction of hierarchical structure and diversified subjects. In this context, this paper proposes a multi-microgrid distribution system trading strategy that takes into account the distribution network trend, and realizes the balance of interests of the operating entities at all levels on the basis of meeting the constraints on the safe operation of the distribution network. Firstly, a three-level game model of "distribution grid-multi-microgrid-users" is constructed. According to the decision-making status of different levels, the distribution network operator (DSO), as the first-level leader of the system, sets the purchase and sale price to maximize the benefits, and at the same time realizes the optimal system trend through active management. The microgrid operator (MGO) as a follower responds to the DSO tariff and as a second-level leader sets the energy sales price for the lower-level producers and consumers (PCUs) to guide the users to use energy. Each PCU as a follower responds to the DSO price to adjust its electricity consumption strategy, and plays a cooperative game with other PCUs to reduce the amount of externally purchased electricity. Then, the KKT condition is used to reconstruct the three-layer model into an internal and external dual model, and the outer model is divided into the optimal trend and electricity price setting problems, which are solved by SOCP and Kriging metamodel equilibrium algorithms, respectively. Finally, the revenue generated from the cooperation between producers and consumers is allocated using the asymmetric Nash bargaining method and solved by the alternating direction multiplier method (ADMM).The example results show that the proposed method can improve the economic benefits of each subject, realize the coordinated and optimized operation of each party, and effectively improve the power quality of distribution network.