A Staged Rush Repair Strategy Of Distribution Networks Based On The Multi-agent System

The distribution network is complicated, it often causes many faults when happyingnatural disasters. Relying on the repair dispatching experience, it is difficult to develop themost effective repair scheme. In view of the feature that the rush-repair faults after thedisaster is different from the general., it emphasizes the repair fastly and timeliness, As ifit can’t recover the power fastly, and it is the prerequisite of the disaster repair. Thereforeit is important to develop a emergency platform of optimal repair scheduling with limitedrepair persons and resources, improving the post-disaster grid power supply security.Based on accessing a large number of domestic and foreign literature in this paper,we do further research on the multi-faults rush repair strategies after disaster, as follows:First of all, on the basis of analyzing the distribution network multiple faultrepairing，using the improved discrete bacterial colony chemotaxis optimization algorithmfor multiple fault model optimization and thus establishing scientific and reasonable repairscheme. Forward reference populations and the directional variation are put forward toincrease the search space. The proposed method can speed up the distribution networkcomputing speed and obtain the global optimal solution.Secondly, according to the actual situation that important load get the priority whendistribution networks lose power. Taking into the concern in economy and recovery time, astaged rush-repair optimization model of multi-teams cooperation with distributedgenerators was developed. The proposed model could make full use of the capacity of thedistributed power, To improve the rush-repair efficiency, the interconnection switcheswere taken as the invented faults, operations by the interconnection switches andtemporary power supply by the generated cars were used too. In the model, DGs contactswitches and emergency generator car with each other to ensure the important load forrapid restoration of power supply.Finally, on the basis of the above work, the multi agent is applied to the distributionnetwork fault repairing, which is composed of control center agent, repair team agent andinformation acquisition agent. As long as the voltage, capacity and other indexes can meetthe constraint demands, then the emergency repair strategy can be established quickly. Multi-agent system takes unexpected new failure into consideration, and control centeragent update repair strategy on condition that normal communication is obtained,otherwise repair squad agent modify network topology according to the team estimatedrepair time, and update repair strategies.