期刊导航

论文摘要

改进的克隆选择算法与SPEA相结合的进化算法

Evolutionary Algorithm Based on Improved Clonal Selection Algorithm and SPEA

作者:杨观赐(中国科学院 成都计算机应用研究所);马鑫(贵州大学 教育部现代制造技术重点实验室);李少波(中国科学院 成都计算机应用研究所;贵州大学 教育部现代制造技术重点实验室);钟勇(中国科学院 成都计算机应用研究所);于丽娅(贵州大学 教育部现代制造技术重点实验室)

Author:Yang Guanci(Chengdu Inst. of Computer Applications,Chinese Academy of Sci.);Ma Xin(Key Lab. of Advanced Manufacturing Technol., Guizhou Univ.);Li Shaobo(Chengdu Inst. of Computer Applications,Chinese Academy of Sci.;Key Lab. of Advanced Manufacturing Technol., Guizhou Univ.);Zhong Yong(Chengdu Inst. of Computer Applications,Chinese Academy of Sci.);Yu Liya(Key Lab. of Advanced Manufacturing Technol., Guizhou Univ.)

收稿日期:2010-09-17          年卷(期)页码:2011,43(5):109-113

期刊名称:工程科学与技术

Journal Name:Advanced Engineering Sciences

关键字:多目标优化;进化算法;克隆选择;基因挖掘;遗传信息

Key words:multiobjective optimization;evolutionary algorithms;clonal selection;genes mining;genetic Information

基金项目:教育部新世纪优秀人才支持计划资助项目(NCET09-0094);贵州省科学技术基金资助项目(黔科合J字[2010]2095号)

中文摘要

为了使进化过程中子代的繁殖能够像生物繁殖那样继承进化信息,通过挖掘抗体中优秀决定基并生成记忆集、增加高斯变异、用变异抗体群中亲和度高的抗体按概率替换记忆抗体群中低亲和度抗体等策略,提出了一种改进的克隆选择算法(ICSA)。将ICSA与SPEA相结合,形成了一种改进的克隆选择算法与强度Pareto进化算法相结合的新型的进化算法(ICSA-SPEA)。ICSA-SPEA通过克隆选择替代选择、交叉、重组等遗传操作。用一组多目标0/1背包问题测试算法性能的统计结果表明,改进的算法可以有效保持种群多样性,具有良好的收敛精度与准确度。

英文摘要

In order to inherit evolutionary information as living beings during offspring generation, a kind of improved clonal selection algorithm (ICSA) was put forward based on mining excellent gene schema to fill a memory pool from antibody set, applying Gaussian mutation operator, and replacing low affinity antibody with high affinity antibody with probability from mutation antibody population during updating memory antibody population. By combining ICSA with strength Pareto evolutionary algorithm (SPEA), a new kind of evolutionary algorithm (ICSA-SPEA) was proposed, which replaces the genetic operation such as selection, crossover and recombinant with clonal selection. The testing of a multi-objective 0/1 knapsack problem showed that ICSA-SPEA has ability to maintain the diversity of population, and is capable of finding out the well distributed non-dominated solutions approximating to Pareto front.

关闭

Copyright © 2020四川大学期刊社 版权所有.

地址:成都市一环路南一段24号

邮编:610065