A Binary-Differential Evolution algorithm based on Ordering of individuals

Abstract

Differential Evolution (DE) algorithm has been shown to be powerful for many real optimization problems. In traditional DE and its variant algorithms, there are three or more candidates used in the step of mutation, which is one of the three typical operation steps: mutation, crossover and selection. To simplify the mutation, we propose a novel evolution algorithm which needs only two individuals in mutation, and is based on a ordering of the individuals. It is the Binary-Differential Evolution algorithm based on Ordering of individuals (OBDE), which shows two-candidate-mutation can accomplish function optimization as well. Experimental results show that OBDE has advantages such as faster computing speed and better precision, compared with traditional DE algorithm.