Abstract
This paper develops a novel, effective, and practical decision aid tool for distribution system operators based on a many-objective evolutionary algorithm. Since the distribution network reconfiguration problem in the real world has many objectives including discrete objectives, non-convexity, and severe constraints, the distribution system operators have difficulty in finding a preferred solution by him/her with little effort. To realize an efficient decision making in such a complicated optimization problem, providing alternatives with him/her to be chosen among them is one of practical approaches from the viewpoint of the multi-objective optimization approach. In order to find well-diversified Pareto optimal solutions for the many-objective optimization problem with an unknown Pareto front, we propose novel algorithms specialized for the distribution system reconfiguration problem integrated with a promising many-objective evolutionary algorithm, i.e., non-dominated sorting genetic algorithm III (NSGA-III). The proposed method enables the distribution system operators to find Pareto optimal solutions within a practical computational time and also choose the most preferred solution among them. The developed algorithm is validated using a realistic distribution system model with many nodes, switches including remote switches, and SVRs on multiple feeders.
