Improved based Differential Evolution Algorithm using New Environment Adaption Operator

In this work, a novel operator-based differential evolution (DE) algorithm has been proposed. The proposed approach has been inspired by the internal adaption (environment) of the search space. Therefore, maintaining environment (vectors) for the search space can be achieved by introducing the better fitness of candidate solution. In the proposed approach, candidate solutions are multiplied with the different parameter values, which depend on the nature of the problem and available counterbalancing resources. The proposed variant termed an internal adaption-based environment is considered in the existing mutation and crossover operators to provide more diversity for selecting the effective mutant solutions. In the experimental analysis, the proposed approach is compared with the five modern DE variants and tested on benchmark function (f1 to f24) on 20, and 40 dimensions. In addition, it is also verified by hypothesis testing in terms of the minimum error. From the obtained results, it is observed that the proposed algorithm is found to be a better target value in terms of the minimum number of function evaluation and statistical functions. It also validates that the proposed algorithm has achieved a reasonable convergence rate and diversity on 20 and 40 dimensions. © 2021, The Institution of Engineers (India).