The elaboration of plans for service restoration in a fast way is necessary to deal with situations that leave network regions out-of-service. These situations may be caused by faults in the distribution network or by isolation of circuit zones for maintenance task. The methodologies developed for elaboration of plans for service restoration require network simplifications to work with large-scale real distribution systems. In general, these simplifications restrict the achievement of an adequate plan for energy restoration. This project proposes the development and implementation of a new technique for elaboration of plans for service restoration, based on Genetic Algorithms (GAs), that can deal with large-scale real distribution systems, without requiring network simplifications, that is, considering a large number of (or all) lines, buses, loads and switches of the system. In order to improve the performance (running time) of the GA proposed, it will be investigated the use of the data structure for graphs, called Node-Depth Representation (NDR), to computationally represent distribution systems. Preliminary studies have shown the GA with NDR should enable a better investigation of the search space, requiring a very reduced running time.
News published in Agência FAPESP Newsletter about the scholarship: