Hybrid Algorithm based on Differential Immune Clone with Orthogonal design method

Abstract

A novel Hybrid Algorithm called Hybrid Algorithm based on Differential Immune Clone with Orthogonal design method (OHADIC) is proposed in this paper, which can avoid the decrease of population diversity and accelerate the convergence rate in evolutionary process. The novel algorithm adopts several main operators to evolve two populations; they are clone reproduction and selection, differential mutation, crossover and selection. Moreover, the orthogonal design method is not only to be used to design orthogonal crossover, but also is adapted to scheme orthogonal local search. In experiments, a wide range of benchmark functions is used to validate the novel hybrid algorithm. Performance comparisons with other well-known differential evolution algorithms including DE, JADE and SADE are also presented, and it is shown that OHADIC has better performance in optimizing these functions.