摘要
针对传统樽海鞘群算法寻优精度低、易于陷入局部最优的问题,提出基于混沌映射与动态学习的自适应樽海鞘群算法。引入改进混沌Tent映射实现种群初始化,确保更加均匀的搜索空间;设计基于Logistic映射的领导者更新机制,有效增强种群多样性;利用基于动态学习的追随者更新机制,使算法跳出局部最优,提升全局搜索能力;设计领导者/追随者规模的自适应调整机制,有效均衡种群的局部开发和全局勘探能力。实验结果表明,该算法在收敛速度、寻优精度及寻优成功率上均有大幅提升。
The traditional salp swarm algorithm has lower optimization precision and is easy to fall into local optimum.To solve this problem,a self-adaptive salp swarm algorithm based on chaotic mapping and dynamic learning was proposed.An improved chaotic Tent mapping was introduced to the population initialization,which guaranteed uniform search space.A leader update mechanism based on Logistic mapping was designed,which effectively enhanced population diversity.A new follower update mechanism based on dynamic learning was designed,which made the algorithm jump out of a local optimum and improve the global searching ability.A self-adaptive adjustment mechanism of the leader/follower population size was designed,which effectively balanced the individual local exploitation and the global exploration ability.Results show that the proposed algorithm performs better on the optimization precision,the convergence speed and the optimization success rate.
作者
卓然
王未卿
ZHUO Ran;WANG Wei-qing(Department of Basic Education,Zhejiang Vocational College of Special Education,Hangzhou 310023,China;DonLinks School of Economics and Management,University of Science and Technology Beijing,Beijing 100083,China)
出处
《计算机工程与设计》
北大核心
2021年第7期1963-1972,共10页
Computer Engineering and Design
基金
国家自然科学基金项目(71901025)。
关键词
樽海鞘群算法
混沌映射
动态学习
收敛速度
局部最优
salp swarm algoirthm
chaotic mapping
dynamic learning
convergence speed
local optimum
