The Service Restoration In The Distribution Network Based On The Mobile Multi-agent

With the promotion and development of smart grid, distribution network structure isincreasingly complex because of the integration of distributed generators and micro grid.The traditional optimization algorithm can not fully meet the needs of distributionnetwork fault recovery reconstruction, so new methods that adapt to complex distributionnetwork are particularly important and in urgent need. In this paper, based on thebackground and current research situation of the distribution network fault recovery, theshortcomings of research on distribution network fault recovery is studied. The concreteresearch content is as followings:Firstly, topology analysis and power flow calculation are introduced which arefrequently used in distribution network fault recovery process. Based on the comparisonof depth first search algorithm and breadth-first search algorithm and the radialdistribution network structure, this paper adopt the breadth-first search algorithm fortopology analysis. The back/forward sweep algorithm is used for power flow calculationbased on the comparison of several different power flow calculation methods.Secondly, mobile multi-agent based distribution network fault recovery strategyconsidering distributed generators is proposed. The scheme implement fault recoverytasks by building control center agents and subtasks agents. Ant colony algorithm isembedded into subtasks agent to realize the optimization of distribution network faultrecovery, and its search efficiency is improved through the subtasks agent module ofdynamic alliance. The simulation result demonstrates the effectiveness of the proposedstrategy.Finally, this article considers kinds of distributed generators supply integrated intodistribution network in the form of micro grid. The random power model is convertedinto day parting or constant power scene model based on the analyzed of kinds ofdistributed generators. This paper set continuous fault and discontinuous fault whichconsidering different failure background. Using the evolutionary algorithm based onmobile multi-agent more for multi-objective optimization of the distribution network. Finally the effectiveness of the proposed algorithm is verified through case studiesconsidering micro grid.