[1]高智阔,陈未如,彭弗楠.基于网格投影的超多目标进化算法[J].计算机技术与发展,2023,33(05):22-28.[doi:10. 3969 / j. issn. 1673-629X. 2023. 05. 004]
　GAO Zhi-kuo,CHEN Wei-ru,PENG Fu-nan.Many-objective Evolutionary Algorithm Based on Grid Projection[J].,2023,33(05):22-28.[doi:10. 3969 / j. issn. 1673-629X. 2023. 05. 004]
点击复制

基于网格投影的超多目标进化算法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
33
期数:
2023年05期
页码:
22-28
栏目:
分布与并行计算
出版日期:
2023-05-10

文章信息/Info

Title:
Many-objective Evolutionary Algorithm Based on Grid Projection
文章编号:
1673-629X(2023)05-0022-07
作者:
高智阔1; 2 ; 陈未如1; 2 ; 彭弗楠1; 2
1. 沈阳化工大学 计算机科学与技术学院,辽宁 沈阳 110142;
2. 辽宁省化工过程工业智能化技术重点实验室,辽宁 沈阳 110142
Author(s):
GAO Zhi-kuo1; 2 ; CHEN Wei-ru1; 2 ; PENG Fu-nan1; 2
1. School of Computer Science and Technology,Shenyang University of Chemical Technology,Shenyang 110142,China;
2. Liaoning Province Key Laboratory of Industrial Intelligence Technology on Chemical Process,Shenyang 110142,China
关键词:
超多目标优化; 进化算法; 目标空间; 网格; 投影; 评价指标
Keywords:
many-objective optimization; evolutionary algorithm; objective space; grid; projection; performance indicators
分类号:
TP301. 6
DOI:
10. 3969 / j. issn. 1673-629X. 2023. 05. 004
摘要:
针对超多目标优化问题求解困难的问题,研究如何得到收敛性和分布性较优的解集,提出了一种基于网格投影的超多目标进化算法-GPEA。 该算法根据决策需求将超多目标优化问题的目标空间进行分解,得到投影维目标空间和自由维目标空间;再将投影维目标空间分割为若干投影格,将自由维目标空间分段成若干自由格。 算法在每个投影格上进行种群进化,并根据个体相对投影格的位置采用两测度策略筛选个体。 第一测度是对落入到投影格内的个体使用非支配排序和自由维目标空间个体筛选策略,选择收敛性和分布性较优的个体作为候选种群。 当落入到投影格内的个体数量不足时,进行第二测度筛选,根据个体相对投影格的距离排队,选择相对较近的个体并入到候选种群中。 分析了算法的性能,通过对标准测试函数在不同目标下的求解,实验证明基于网格投影的超多目标进化算法能够有效地求解超多目标优化问题。
Abstract:
Aiming at the difficult problem of solving many-objective optimization problems,how to obtain a solution set with better convergence and distribution is studied, and a many - objective evolutionary algorithm based on grid projection ( GPEA ) is proposed.According to the decision-making requirements,the objective space is decomposed into the projection-dimensional space and the free-dimensional space,then the projected dimension space is divided into several projection grids,and the free dimension space is segmentedinto several free grids. The GPEA performs population evolution on each projected grid, and uses a two - measure strategy to selectindividuals according to their position relative to the projected grid. The first measure is to use nondominated sorting and a free -dimensional space individual selecting strategy for those falling into the projection gird,and the individuals with better convergence anddistribution are selected as a candidate population. When the number of individuals falling into the projected grid is insufficient, thesecond measure is performed,and the individuals relatively close to the projected grid are selected to be incorporated into the candidatepopulation. The performance of the algorithm is analyzed. By solving the standard test function under different objectives,the experimentproves that the GPEA can effectively solve the many-objective optimization problem.

相似文献/References:

[1]樊玮 朱贺.软件结构化测试用例自动生成方法[J].计算机技术与发展,2010,(05):26.
　FAN Wei,ZHU He.Summary of Method for Software Structured Testing Case Automatic Generation[J].,2010,(05):26.
[2]廖锋 高兴宝.差分演化算法在约束优化问题中的应用[J].计算机技术与发展,2010,(05):187.
　LIAO Feng,GAO Xing-bao.Application of Differential Evolution Algorithms on Constraint Optimization Problems[J].,2010,(05):187.
[3]陈玲 严南南.基于多Agent和进化算法的作业线调度研究[J].计算机技术与发展,2009,(11):210.
　CHEN Ling,YAN Nan-nan.Crane Scheduling Research Based on Multi- Agent and Evolutionary Algorithms[J].,2009,(05):210.
[4]周国亮 刘希玉.一种基于进化算法的BP神经网络优化方法及应用[J].计算机技术与发展,2007,(08):58.
　ZHOU Guo-liang,LIU Xi-yu.A Method of BP NN Optimization Based on Evolution Algorithm and Its Application[J].,2007,(05):58.
[5]常新功 李宏.基于近似个体协同的进化子结构发现[J].计算机技术与发展,2010,(09):106.
　CHANG Xin-gong,LI Hong.Evolutionary Algorithm for Substructure Discovery Based on Approximate Individual Cooperation[J].,2010,(05):106.
[6]周飞,罗杰.基于远缘杂交的精英进化算法[J].计算机技术与发展,2013,(02):93.
　ZHOU Fei,LUO Jie.Elite Evolutionary Algorithm Based on Distant Hybridization[J].,2013,(05):93.
[7]李杭涓,崔志华,谢丽萍.基于权重学习的高维多目标进化算法[J].计算机技术与发展,2021,31(01):18.[doi:10. 3969 / j. issn. 1673-629X. 2021. 01. 004]
　LI Hang-juan,CUI Zhi-hua,XIE Li-ping.Research on Multi-objective Evolution Algorithm Based on Weight Learning[J].,2021,31(05):18.[doi:10. 3969 / j. issn. 1673-629X. 2021. 01. 004]

更新日期/Last Update: 2023-05-10