自适应协同学习水波优化算法  

Adaptive Cooperative Learning Water Wave Optimization

作　　者：顾启元[1] 王俊祥[1] GU Qi-yuan;WANG Jun-xiang(College of Software Engineering,Chongqing University of Arts and Sciences,Chongqing 402160,China)

机构地区：[1]重庆文理学院软件工程学院

出　　处：《小型微型计算机系统》2019年第9期1858-1863,共6页Journal of Chinese Computer Systems

基　　金：重庆市教委科学技术研究项目(cstc2017jcyjAX0045)资助;重庆市永川区科技项目(Ycstc,2017nc2001)资助

摘　　要：水波优化算法(Water Wave Optimization,WWO)是一种新型群体智能搜索技术,具有种群规模小、操作简易等优点.但依然存在收敛速度慢、收敛精度不高等缺陷.为了改善WWO优化性能,提出一种自适应协同学习水波优化算法.算法中采用双种群进化结构实现主种群勘探和子种群开采的协同学习.主种群采用一种自适应学习策略,在维持种群多样性同时有效增强个体学习的效率.主群和子群的交互机制,可以使子群摆脱局部最优,提高算法的收敛精度.复杂多模基准测试函数的仿真结果表明本文算法在收敛精度和收敛速度上都有显著提高.Water Wave Optimization( WWO) is an novel swarm intelligence searching technique. Despite of its advantages with small swarm size and easy implementation,it still risks the lowsearching precision and the slow convergence. In order to improve the optimization performance of WWO,an adaptive cooperative learning water wave( ACLWWO) optimization is developed in which the structure of double swarms is used to balance the global exploration and local exploitation. In the master swarm,an adaptive learning is presented to maintain the swarm diversity and enhance the learning efficiency of the individual. Meanwhile,the interaction mechanism of two swarms can prevent the slave swarm getting into the premature convergence,which can improve the convergence precision. The experimental results show the performance of the proposed algorithm has been improved significantly in terms of convergence speed and solution accuracy.

关 键 词：水波优化算法 收敛速度 收敛精度 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]