Research On Distributed Coordinated Operation Method Of Islanding Microgrid

Microgrid(MG)is an effective way to integrate and control all kinds of distributed energy.In recent years,MG has gradually become a hot spot in the field of power system research because of its advantages of flexibility,efficiency and reliability.However,the output power of renewable energy in MG has great randomness and volatility,which will seriously affect the power supply reliability and economic benefits of MG.Therefore,it is of great significance to study how to coordinate the power output of various distributed power sources to ensure the internal power balance of the system and realize the optimal economic dispatch of MG.Based on this,this paper mainly focuses on the research of the operation benefit of MG in two stages: day-ahead and real-time.This paper established a day ahead distributed economic optimal scheduling model based on multi-agent technology,which to maximize the benefits of island MG system while keeping voltage stabilization and satisfying the capacity limits.In this model,load unit have the same rights as other units to participate in the establishment of clearing price in the island MG power market.Aiming at the nonconvex characteristics of the model,the paper analyzed the conditions should to be satisfied when the alternating direction multiplier method(ADMM)is applied to non-convex optimization.And the ADMM algorithm was used to solve the distributed optimization of the model.In the solution process,each agent point and the neighboring agent point only need to exchange the expected transaction power and electricity price information.Until the convergence conditions are met,the island MG does not complete the settlement of market energy transactions.Therefore,the privacy of each unit is better protected.The model is simulated in three different weather conditions,and the power and operating benefits of various stakeholders are compared under the three weather conditions.The results verify the correctness and feasibility of the proposed model.Day-ahead scheduling model depends on the forecast information or previous optimization results,this may be cause to inaccurate scheduling results.In order to improve the model,a real-time optimal scheduling model is established in this paper.The model fully considers the cost of energy acquisition in the past period and the operation benefit in the future period,so as to determine the mode of energy storage unit participating in the economic optimization of island MG in the current period.On this basis,each agent in the island MG can maximize its own operating efficiency by exchanging expected transaction power and electricity price with its neighboring agents.At last,under three different weather scenarios,the operation results of distributed economic optimal dispatch of islanded island MG are analyzed,and the validity of the proposed model and solution method is verified.