吉林大学学报(理学版) ›› 2024, Vol. 62 ›› Issue (6): 1419-1425.

• • 上一篇    下一篇

适应度步长的全局局部协同优化算法

初雅莉1, 韩旭明2,3, 王晏泽2, 吕帅2   

  1. 1. 长春工业大学 数学与统计学院, 长春 130012;  2. 暨南大学 信息科学技术学院, 广州 510632;3. 可信人工智能教育部工程研究中心, 广州 510632
  • 收稿日期:2023-06-06 出版日期:2024-11-26 发布日期:2024-11-26
  • 通讯作者: 韩旭明 E-mail:hanxuming@jnu.edu.cn

Global-Local Cooperative Optimization Algorithm with Fitness Step Size

CHU Yali1, HAN Xuming2,3, WANG Yanze2, LV Shuai2   

  1. 1. School of Mathematics and Statistics, Changchun University of Technology, Changchun 130012, China;
    2. College of Information Science and Technology, Jinan University, Guangzhou 510632, China;
    3. Engineering Research Center of Trustworthy AI of Ministry of Education, Guangzhou 510632, China
  • Received:2023-06-06 Online:2024-11-26 Published:2024-11-26

摘要: 针对现有优化算法求解精度低的问题, 提出一种适应度步长的全局局部协同优化算法. 该算法通过均衡化个体适应度, 动态分配每次迭代中个体的全局搜索步长和局部搜索步长, 实现了算法在解空间内全局搜索和局部搜索的有效协同, 进而提升了求解精度. 实验结果表明, 该算法在基准函数测试中展现了较高的精度和良好的稳定性, 并通过仿真实验验证了其在解决复杂工程优化问题中的有效性.

关键词: 优化算法, 均衡化适应度值, 适应度步长, 协同搜索, 工程优化

Abstract: Aiming at  the problem of low solution precision in existing optimization algorithms, we proposed a global-local cooperative optimization algorithm with fitness step size.  The algorithm achieved  effective collaboration between global and local search in  the solution space by balancing  individual fitness and dynamically allocating global and local search step sizes during each iteration, thereby enhancing the solution precision. Experimental results show that  the proposed algorithm has  high precision and stability in benchmark function tests, and its effectiveness in solving complex engineering optimization problems is verified through simulation experiments.

Key words: optimization algorithm, normalized fitness value, fitness step size, cooperative search, engineering optimization

中图分类号: 

  • TP301