Opposition Based Constriction Factor Particle Swarm Optimization for Economic Load Dispatch

Abstract

This paper makes an attempt to incorporate Opposition Based Learning (OBL) technique into the classic Particle Swarm Optimization (PSO) method modified by constriction factor. Aim of the work is to improve the convergence of PSO by avoiding premature convergence at local optima. The proposed OBL will help to improve the exploration as well as exploitation capability of the algorithm with the help of introducing the opposite particles into the search space and hence increasing the search space as well. In order to validate the proposed method, the economic load dispatch problem of electric power system is considered and the proposed method is validated on two test systems; 3 unit and 12 unit generating systems. The validation is done with Inertia Factor based PSO, Constriction factor based PSO and with Opposition based Constriction factor PSO for both 3 unit and 12 unit systems. The results are compared on the basis of fuel cost as well as the convergence rate of the algorithms. The Constriction factor based PSO gives minimum fuel cost and Opposition based Constriction factor PSO improves the result with a better convergence rate.