摘要
为提高蜉蝣算法的收敛速度,提升算法寻优能力,提出一种引入精英反向学习和柯西变异的混沌蜉蝣算法。利用Circle混沌映射序列优化初始种群使种群分布更加均匀,提高种群多样性。在蜉蝣更新阶段,对蜉蝣中的精英个体进行反向学习策略,防止算法陷入局部最优,提高算法收敛速度。为保证种群进化方向和扩大寻优范围,将自适应概率阈值和柯西变异的扰动机制相结合,对劣势蜉蝣个体附近生成更大的扰动。通过8个基准测试函数实验对比和Wilcoxon秩和检验,实验结果表明,混沌蜉蝣算法在收敛速度、求解精度以及稳定性等方面有较大提高。
To improve the convergence speed of the mayfly algorithm and the optimization ability of the algorithm,a chaotic mayfly algorithm based on elite reverse learning and Cauchy mutation was proposed.The initial population was optimized through the Circle chaotic map sequence,so that the population distribution was more uniform and the population diversity was improved.In the mayfly update stage,the opposition-based learning strategy was performed on the elite individuals in the mayfly to prevent the algorithm from falling into local optimum and improve the algorithm convergence speed.To ensure the evolution direction of the population and expand the scope of optimization,the adaptive probability threshold and the perturbation mechanism of Cauchy variation were combined to generate more significant perturbation near the inferior mayfly individuals.Results of simulation experiments and Wilcoxon test of 8 benchmark functions verify that the chaotic mayfly algorithm significantly improves the convergence speed,the solution accuracy,and the stability.
作者
张少丰
李书琴
ZHANG Shao-feng;LI Shu-qin(College of Information Engineering,Northwest A&F University,Yangling 712100,China)
出处
《计算机工程与设计》
北大核心
2024年第1期187-196,共10页
Computer Engineering and Design
基金
国家重点研发计划基金项目(2020YFD1100601)。
关键词
蜉蝣算法
混沌映射
精英反向学习
柯西变异
扰动机制
自适应
劣势蜉蝣
mayfly algorithm
chaotic mapping
elite reverse learning
Cauchy mutation
perturbation mechanism
self-adaptation
inferiority mayfly