首页 | 本学科首页   官方微博 | 高级检索  
     

基于邻域差分和协方差信息的单目标进化算法
引用本文:李学强,黄翰,郝志峰.基于邻域差分和协方差信息的单目标进化算法[J].软件学报,2018,29(9):2606-2615.
作者姓名:李学强  黄翰  郝志峰
作者单位:东莞理工学院 计算机与网络安全学院,广东 东莞 523000,华南理工大学 软件学院,广东 广州 510006,佛山科学技术学院 数学与大数据学院,广东 佛山 528000
基金项目:国家自然科学基金(61370102);广东省杰出青年自然科学基金(2014A030306050);教育部-中国移动科研基金(MCM20160206);广东高层次人才特殊支持计划(2014TQ01X664).
摘    要:复杂的单目标优化问题是进化计算领域的一个研究热点问题.已有差分进化和协方差进化被认为是处理该问题的较有效的方法,其中差分信息类似于梯度可以有效的指导算法朝着最优解方向搜索,而协方差则是基于统计的方式来生成较优的子代种群.本文引入了协方差信息对差分算子进行改进,提出了一种基于邻域差分和协方差信息的进化算法(DEA/NC)来处理复杂的单目标优化问题.算法对现有差分算子中通常采用的随机选点或结合当前最优解进行差分的方式进行了分析,当随机选择的差分个体间的差异较大时,差分信息不能作为一种局部的梯度信息来指导算法的搜索;而结合最优解的差分信息又会使得种群朝着当前最优解的方向搜索,导致种群快速的陷入局部最优.基于此,本文采用了邻域差分的方式来提高差分算子的有效性,同时避免种群的多样性丢失.另外,引入了协方差来度量个体变量间的相关度,并利用相关度来优化差分算子.最后,算法对cec2014中的单目标优化问题进行了测试,并将实验结果与已有的较好的差分进化算法进行了比较,实验结果表明了本算法的有效性.

关 键 词:单目标优化  进化算法  差分进化  协方差进化  多样性保持
收稿时间:2017/4/26 0:00:00
修稿时间:2017/7/10 0:00:00

Evolutionary Algorithm for Single-Objective Optimization Based on Neighborhood Difference and Covariance Information
LI Xue-Qiang,HUANG Han and HAO Zhi-Feng.Evolutionary Algorithm for Single-Objective Optimization Based on Neighborhood Difference and Covariance Information[J].Journal of Software,2018,29(9):2606-2615.
Authors:LI Xue-Qiang  HUANG Han and HAO Zhi-Feng
Affiliation:School of Information Technology, Guangdong Medical University, Dongguan 523808, China,School of Software Engineering, South China University of Technology, Guangzhou 510006, China and School of Mathematics and Big Data, Foshan University, Foshan 528000, China
Abstract:Complex single-objective optimization problem is a hot topic in the field of evolutionary computation. Differential evolution and covariance evolution are considered to be two of the most effective algorithms for this problem, the difference information similar to the gradient can effectively guide the algorithm towards the optimal solution direction, and the covariance is based on statistics to generate an offspring population. In this paper, the covariance information is introduced to improve the difference operator, then an evolutionary algorithm based on neighborhood difference and covariance information (DEA/NC) is proposed to deal with complex single-objective optimization problem. The two commonly used difference operators generated by random selection individuals and combined by the current optimal solution are analyzed, on the first approach, the difference information can not be used as a local gradient information to guide the search of the algorithm when the euclidean distance between randomly selected individuals is large; and the second approach will make the population search in the direction of the current optimal solution, which will lead the population to quickly fall into local optimum. Based on this, the neighborhood difference method is proposed to improve the effectiveness of the differential operator, while avoiding the diversity of population loss. In addition, the covariance is introduced to measure the correlation between individual variables, and the correlation is used to optimize the difference operator. Finally, the algorithm tests the single-objective optimization problem in cec2014, and compares the results with the existing differential evolution algorithms. The experimental results show the effectiveness of the proposed algorithm.
Keywords:single-objective optimization  evolutionary algorithm  differential evolution  the covariance evolution  diversity maintenance
点击此处可从《软件学报》浏览原始摘要信息
点击此处可从《软件学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号