Transposition Point Optimization for Voltage Unbalance Mitigation in Electric Power Distribution Systems by Integer Form of Population-Based Incremental Learning

This paper proposes a transposition point optimization method for voltage unbalance mitigation in electric power distribution systems by integer form of population-based incremental learning. In power distribution systems, since voltage unbalance has become a problem due to the recent penetration of large amounts of low-voltage photovoltaic, a method for voltage unbalance mitigation is required. Transposition is one of the methods for voltage unbalance mitigation. It is necessary to determine how and at which points transposition should be performed in order to apply transposition. The problem can be formulated as a combinatorial optimization problem. Since the number of combinations in the combinatorial optimization problem increases exponentially as the system size expands, it is difficult to obtain an optimal solution in practical time. However, no research has been conducted on the problem until now. Evolutionary computation can be applied to the problem. The proposed integer form of population-based incremental learning based method is applied to a real-scale distribution model system based on a JST-CREST126 distribution line model. Effectiveness of the proposed method is verified by comparing it with the tabu search based method and the Mann–;Whitney U test.