An Improved Elephant Clan Optimization Algorithm for Global Function Optimization

Abstract The elephant clan optimization algorithm (ECO) is a novel metaheuristic inspired by modeling the most basic individual and collective behavior of elephants. However, it suffers from the problems of easily falling into local optimum as well as insufficient convergence speed and convergence precision. To further improve the convergence performance of ECO, an improved elephant clan optimization algorithm (IECO) is proposed in this paper. The population initialization method with additional autonomous movement strategy, the Euclidean distance-based population partitioning method and the early maturity suppression mechanism proposed to improve the population diversity and the ability of the algorithm to jump out of the local optimum. An improved individual population update strategy balances the algorithm's convergence speed and variety. Finally, the enhanced substitution improves the convergence speed while maintaining population diversity and improves the algorithm's robustness to different optimization problems. The experimental results on the CEC2013 test set show that the IECO algorithm has significant advantages in terms of convergence speed, convergence accuracy, and stability compared with the original ECO algorithm and four other excellent algorithms.