| Autor: |
Xi, Mingyang, Song, Qixian, Xu, Min, Zhou, Zhaorong |
| Zdroj: |
International Journal of Machine Learning & Cybernetics; Apr2023, Vol. 14 Issue 4, p1333-1364, 32p |
| Abstrakt: |
As one novel meta-heuristic algorithm, African Vultures Optimization Algorithm (AVOA) has been proved to be efficient in solving continuous optimization problems. However, many real-world optimization problems are in the discrete form, and the continuous characteristics of AVOA make it unsuitable for solving discrete optimization problems. Therefore, this article proposes Binary African Vultures Optimization Algorithm (BAVOA) to solve various optimization problems, especially discrete optimization problems. In BAVOA, the X-shaped transfer function is firstly adopted to convert the continuous search space into the binary search space, and then the opposition-based learning strategy and the improved multi-elite strategy are utilized to enhance the optimization ability of BAVOA. Moreover, the performance of BAVOA is evaluated by twenty-three benchmark functions with the relevant Wilcoxon rank sum tests, and the effectiveness of BAVOA is demonstrated by four engineering design problems and one combinational optimization problem. The results demonstrate that BAVOA outperforms eight well-known algorithms in addressing various optimization problems. Source codes of BAVOA are publicly available at: https://www.mathworks.com/matlabcentral/fileexchange/115350-binary-african-vultures-optimization-algorithm [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
Full text from SpringerLink