基于高斯映射和小孔成像学习策略的鲸鱼优化算法  

Whale optimization algorithm based on Gauss map and small hole imaging learning strategy

作　　者：徐航 张达敏 王依柔 宋婷婷 王栎桥 Xu Hang;Zhang Damin;Wang Yirou;Song Tingting;Wang Liqiao(College of Big Data&Information Engineering,Guizhou University,Guiyang 550025,China)

机构地区：[1]贵州大学大数据与信息工程学院,贵阳550025

出　　处：《计算机应用研究》2020年第11期3271-3275,共5页Application Research of Computers

基　　金：贵州省自然科学基金资助项目(黔科合基础[2017]1047号)。

摘　　要：针对鲸鱼优化算法(WOA)容易陷入局部最优解、收敛速度慢等缺陷,提出一种基于小孔成像反向学习策略的鲸鱼优化算法。首先采用高斯映射生成的混沌序列取代原始算法中随机产生的初始种群,增加种群的多样性;其次,提出了一种小孔成像反向学习策略,并结合最优最差反向学习思想,增加了寻优位置的多样性,提高了算法跳出局部最优的能力;最后,在算法中加入了一个非线性收敛因子和一个对数形式的概率阈值,在保留鲸鱼算法优点的前提下,协调了算法的全局搜索和局部开发能力。通过对10个基准函数进行仿真测试,实验结果表明改进算法在收敛速度和收敛精度等方面有明显的提高。Aiming at the defects that the whale optimization algorithm(WOA)is easy to fall into the local optimal solution and the convergence speed is slow,this paper proposed a whale algorithm based on the small hole imaging reverse learning strategy to solve the defects.Firstly,this paper introduced the chaotic sequence generated by Gauss map to replace the original algorithm for increasing the diversity of the population.Secondly,it introduced a small hole imaging reverse learning strategy,on this basis,by combining with the optimal worst reverse learning method,which increased the diversity of the optimal position and improved the ability to jump out of local optimum.Finally,it added a nonlinear convergence factor and a logarithmic probability threshold in the algorithm to balance the exploration and exploitation of the algorithm while preserving the advantages of the whale algorithm.Through the simulation test of 10 benchmark functions,the experimental results show that the improved algorithm has obvious improvement in convergence speed and convergence precision performance.

关 键 词：鲸鱼优化算法 高斯映射 小孔成像反向学习 概率阈值 非线性收敛因子 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]