Economic Scheduling Research Of Microgrid Under Multi-Agent-System Based Day-ahead Electricity Market Environment

Microgrid consists of a variety of distributed generators, with renewable energy as its main resources. Its attractive advantage is low cost, low voltage and low pollution and so on. Microgrid is available to integrate the advantage of a variety of distributed generation and coordinating the contradiction between the distributed system and the main grid. In this case, different resources with diversified characteristics have been utilized efficiently, so Microgrid will be the key component of the smart grid in the future. However, with fast increasing of installed capacity of these distributed generations, surplus electricity in Microgrid becomes an urgent problem to be solved. This article researches and analyzes the Microgrid economic scheduling problems from the perspective of day-ahead electricity market, and maximizes the profit of Microgrid or distributed generator units.On the one hand, this article studies economic scheduling problem of connected Microgrid under the condition of participating in day-ahead electricity market. In this article, a Microgrid economic scheduling system based on Multi-agent-system(MAS) structure is established, which can manage the operation and coordination of distributed generator units in Microgrid, and the transaction between Microgrid and main grid. In the scheduling of microgrid in day-ahead electricity market, the randomness of distributed generation unit, load and electricity price is in consideration. Point estimate method is used to solve the random problem, and also benders decomposition technique can help solve the economic scheduling problem and make the Microgrid operation optimization in this article.On the other hand, in view of the island Microgrid, this article builds the bilevel MAS based Microgrid electricity transaction market. Under this foundation, the economic scheduling of distributed generator units in Microgrid electricity market is researched, and the profit allocation of distributed generator units is in consideration. In addition, the bilevel programming problem is modelled for solving Microgrid day-ahead electricity market bidding problem in this article. And the bilevel programming model is converted into the mixed integer linear programming problem through binary expansion approach. Finally, the bidding problem is shaped into a game based on game theory, and the unique Nash equilibrium is obtained by solving equilibrium problem with equilibrium constraints(EPEC) and auxiliary optimal problem.