An Adaptive Differential Evolution Algorithm with Hierarchical Mutation Strategy and Opposition Learning

In recent years, differential evolution algorithm (DE) has shown excellent performance in solving various optimization problems, therefore it has been extensively applied in many research and scientific fields. However,it's easy for DE to obtain the local optimal solution. In the cause of improving the convergence performance and global search ability of DE,an improved differential evolution algorithm is proposed in this paper. In this algorithm, the initial population is generated by the Halton sequence, and in the process of mutation and crossover, adaptive mutation operator and crossover operator are applied. Then on the basis of different adaptive values, the population is layered to two parts, the two parts take different mutation strategy. The proposed algorithm is compared with DE and other variants of DE in 10, 30, and 50 dimensions respectively by using a set of twenty-six benchmark functions. The experimental results indicate that the proposed algorithm can significantly improve global optimization performance.