Enhanced differential evolution using random-based sampling and neighborhood mutation

Differential evolution (DE) is a simple and efficient global optimization algorithm. When differential evolution is applied in complex optimization problems, it has the shortages of prematurity and stagnation. An enhanced differential evolution using random sampling and neighborhood mutation to solve the above problems is proposed in this paper. The proposed enhanced DE is called random-based differential evolution with neighborhood mutation (NRDE). Random-based sampling is an improvement of center-based sampling. In NRDE, random-based sampling as the new mutation operator to generate the random-based individuals and the designed neighborhood mutation operator are used to search in the neighborhood created by the centers of the population and the sub-population. This paper compares other state-of-the-art evolutionary algorithms with the proposed algorithm, NRDE. Experimental verifications are conducted on 24 benchmark functions and the CEC' 05 competition, including detailed analysis for NRDE. The results clearly show that NRDE outperforms other evolutionary algorithms in terms of the solution accuracy and the convergence rate.