A Hybrid Differential Evolution Method for Practical Engineering Problems

Abstract

This paper introduced an almost control parameters free modified differential evolution for global optimization problems. The modifications are derived from the mechanisms of particle swarm optimization viz., topologies, inertial weight, neighborhood best and individual best, with which each individual performed the mutation operator based on its current position, the neighborhood best and its individual best along with the inertial weight. And the crossover operator in traditional differential evolution is removed, while the selection operator was employed. The approach was employed for a tension/compression string design problem and an economic dispatch problem in power system. By comparisons with the other evolutionary algorithms, the proposed approach has shown its feasibility and effectiveness.