Application of simple examination and retrieval procedure in evolutionary algorithms for feeder reconfiguration

Abstract

This paper presents an effective approach based on simple examination and retrieval procedure (SERP) that can be used in evolutionary algorithms, such as genetic algorithm (GA) and binary particle swarm optimization (BPSO), for solving distribution feeder reconfiguration problems. The proposed approach is able to improve the search efficiency by identifying and modifying the illegal chromosomes that represent switching operation plans. The concept of simple examination is done by checking the numbers of normally opened switches on a branchless line segment in distribution systems. By applying the proposed approach, the number of illegal chromosomes during the regeneration process in evolutionary algorithms can be reduced. The idea of retrieval procedure is done by trying to use the information from historical local optimal solutions to repair the illegal chromosomes as many as possible. A 33-Bus distribution system and a practical distribution system of Taiwan Power Company (TPC) are used in this paper to verify and validate the effectiveness of the proposed procedure. The simulations consider the distribution system with and without distributed generation (DG) for power losses problems. Simulation results show that integration of the proposed procedure to GA and PSO provides more efficient and stable solutions for feeder reconfiguration problems. Copyright © 2010 John Wiley & Sons, Ltd.