Multiobjective short-term planning of electric power distribution systems using NSGA-II

Abstract

This paper presents a mixed integer nonlinear programming multiobjective model for short-term planning of distribution networks that considers in an integrated manner the following planning activities: allocation of capacitor banks; voltage regulators; the cable replacement of branches and feeders. The objective functions considered in the proposed model are: to minimize operational and investment costs and minimize the voltage deviations in the the network buses, subject to a set of technical and operational constraints. A multiobjective genetic algorithm based on a Non-Dominated Sorting Genetic Algorithm (NSGA-II) is proposed to solve this model. The proposed mathematical model and solution methodology is validated testing a medium voltage distribution system with 135 buses. © 2013 Brazilian Society for Automatics - SBA.