An improved multi-objective differential evolution algorithm with an adaptive crossover rate

Abstract

In order to properly solve multi-objective optimization problem, an improved multi-objective differential evolutionary algorithm with an adaptive crossover rate is proposed in this paper. To adjust the evolution adaptively, an adaptive crossover rate is integrated into the differential evolution. The new strategy can diverse pareto individuals and further to pareto front, which avoids the local convergence that traditional differential evolution always trapped in. In addition, combining with great ability of searching local optima of differential evolution, evolutionary speed and diversity can be simultaneously improved by the modified crossing operator. The simulation on these benchmark problems verifies the efficiency of the proposed algorithm with convergence metric and diversity metric, and the obtained results also reveal that the proposed method can be a promising alternative in solving multi-objective optimization problems. © 2013 Springer Science+Business Media New York.