A fuzzy color Petri Net for repair scheduling and dispatching of large-scale contingencies in distribution system

摘要

When large-scale contingencies in a distribution system occur, urgent repair must be carried out as soon as possible. The resources to be distributed for the contingency include human power, vehicles and tools for each accident. Efficient scheduling and dispatching for the resources distribution is a crucial and complex task. In this thesis, a fuzzy color Petri Net (FCPN) with rule-based system is proposed to assist the dispatchers in crew management in case of large-scale multiple outages happen. Fuzzy logic is exploited to determine the priority of repair sequences of the events. The proposed fuzzy membership functions include the functions of the types of key customers, the estimated time need for the repair, the amount of resources needed and the easiness of restoration after the repair. To increase the availability of the resources and meet the special requirements of events, fuzzy rules and fuzzy membership functions are proposed to determine the priority of each crew and vehicle to be dispatched to field. Besides, pre-scheduling for the outages that lack some resources temporarily is proposed. To verify the effect of the proposed methodology, partial of the practical Taipower distribution system in Taipei southern area is exploited . Case studies including multiple accidents in a feeder and in several feeders are simulated It is shown that the proposed algorithm is capable of obtaining relevant scheduling results with high efficiency.