부산대학교 도서관

In this article a methodology for optimal distribution system planning (DSP) considering automatic reclosers is presented. The methodology has two phases. An optimal configuration of the distribution system is found in phase I, by using a simulated annealing algorithm (SA). It considers costs of installing of new elements and upgrading of existing ones, and operative costs of energy technical losses. In phase II, a simultaneous location strategy of normally open and closed reclosers (NOR and NCR) is applied, in order to improve the power supply reliability indices and allow proper load transference between feeders under certain operative conditions. To solve this phase, the multi-objective evolutionary algorithm NSGA II (non-dominated genetic algorithm) is used. The proposed methodology introduces reliability topics to the DSP problem such as it allows incorporating cost variety, regulatory constraints and own constraints of the problem. Furthermore, it is easier to implement than traditional centralized models. To demonstrate the efficiency of the proposed methodology, a 54-nodes system is used.