Evolving non-dominated solutions in multi objective fault section estimation for automated distribution networks

Abstract

Multiobjective evolutionary algorithms (MOEAs) that use nondominated sorting have been criticized mainly for their computational complexity and nonelitism approach. In this paper, we suggest a non-dominated sorting based multiobjective evolutionary algorithm (MOEA), called nondominated sorting genetic algorithm-II (NSGA-II) for solving the fault section estimation problem in automated distribution networks, which alleviates all the above difficulties. Due to the presence of various conflicting objective functions, the fault location task is a multi-objective, optimization problem. The considered FSE problem should be handled using Multiobjective Optimization techniques since its solution requires a compromise between different criteria. In the adopted formulation, these criteria are fast and accurate estimation of the potential fault location. In contrast to the conventional Genetic algorithm (GA) based approach; NSGA-II does not require weighting factors for conversion of such a multi-objective optimization problem into an equivalent single objective optimization problem. Based on the simulation results on four different automated distribution networks, the performance of the NSGA-II based scheme has been found significantly better than that of a conventional GA based method.