Dynamic motion based evolutionary algorithm for enhancement of the search capability for global search space

The stagnation problem in evolutionary algorithms reduces the optimization algorithm’s performance after a certain number of iterations, leading to diversity loss. In this article, five new mutation operators using Random, Random-Best, Best, Current Random, and Current-Best-Random selection policies based on dynamic motion strategy according to feasible environments from global search space are added to the Differential Evolution DE) process. The proposed strategy supports the diversity of the problem’s nature while maintaining the global search and preserving environmental balance. The validation process has involved two tests. First, the testing and comparative study results showed that the proposed method performed satisfactorily at the target value of \(10^{-8}\) ${10}^{-8}$ for the bare minimum of statistical and function evaluations. In the second validation test, the proposed operators pass the Wilcoxon rank-sum test for the lowest error in each function (p = 0.05). The proposed approaches show an improvement in the search capability of evolutionary optimization algorithms.