A self-adaptive differential evolution algorithm with an external archive for unconstrained optimization problems

Abstract

Differential evolution algorithm (DE) has yielded promising results for solving nonlinear, non-differentiable and multi-modal optimization issues. Due to its simple structure, fast convergence and strong robustness, DE has received increasing attention and wide application in a variety of fields. We propose a novel differential evolution approach (SE-DE) which uses an external archive for opposition-based learning, by this way, more high quality solutions can be selected for candidate solutions. In addition, the mutation factor (F) is adaptively controlled based on the success of offspring/trial solutions generated. An optimization factor α is proposed to select the crossover strategy, a combination of binomial and exponential crossover can effectively balance the exploration and exploitation ability of the algorithm. The performance of SE-DE is compared with the other five DE algorithms including DE, SADE, ODE, NDE and MDE-pBX. The comparison is carried out for a set of 30-, 50- and 100-dimensional test functions from CEC2005. The results show that our algorithm is better than, or at least comparable to, the algorithms from other literature.