An adaptive artificial electric field algorithm for continuous optimization problems

Abstract

The comprehensive learning strategy is a meticulous method for enhancing the optimization ability of population‐based optimization algorithms. This article proposes an adaptive artificial electric field algorithm (iAEFA), which is developed by embedding a comprehensive learning strategy into AEFA. The proposed algorithm utilizes a novel adaptive approach for developing a better learning strategy in which an agent's velocity is updated using the comprehensive influence of the entire population. The developed scheme has shown a stronger potential to discover better candidate solutions in each iteration. The objective of the proposed method is to develop an efficient optimizer for continuous optimization problems. The performance of the proposed iAEFA is evaluated using a set of 13 classical benchmark test problems and the CEC 2019 (100‐digit challenge) benchmark functions. The experimental results are compared to seven state‐of‐the‐art optimization algorithms. Using the Wilcoxon signed‐rank test, the statistical significance of the results is confirmed. This article also discusses the theoretical convergence of the proposed algorithm, along with other significant findings about the proposed scheme. The experimental results and the theoretical analysis shows that the proposed scheme can be an excellent choice for the function optimization task compared to other existing algorithms.