Integrated methodology between metaheuristic techniques for optimal vegetation maintenance in distribution systems

Abstract

Tree pruning underneath overhead distribution networks is a practice done by utilities, because the vegetation contact with the energized lines represents one of the major causes of interruption electric energy supply. The problem of vegetation maintenance can be studied through two mathematical models. The first model corresponds to a model where the non-served energy level NSEL is minimized; the second model minimizes the average percentage of violation of the security zone APSZ, which refers to a minimum distance between the energized lines and the vegetation. When maintenance cost MC function is added in both models, these are converted into multi-objective models. In this article, it is implemented a quality and quantity improvement methodology of a solutions front that represents a set of vegetation maintenance schedules. This methodology is performed as a hybrid between the genetic algorithm proposed by Chu & Beasley GACB and the non-dominated sorting genetic algorithm NSGA II.