基于定向采样和自适应选择的免疫算法
Immune algorithm based on directional sampling and adaptive selection
杨珍 1李婉晴 2张雄涛2
作者信息
- 1. 湖州师范学院信息工程学院,浙江湖州 313000;湖州学院电子信息学院,浙江湖州 313000
- 2. 湖州师范学院信息工程学院,浙江湖州 313000
- 折叠
摘要
针对算法容易陷入局部最优以及无法很好保持多样性等不足,提出一种基于定向采样和自适应选择的免疫算法(DSASIA).利用在线种群信息动态选择个体;采用自适应比例克隆方式平衡全局和局部搜索能力,确保收敛性;采取定向采样策略识别子代个体,保证多样性.在13个测试函数上与其它4种多目标优化算法进行对比,实验结果表明,DSA-SIA算法可以较快求出帕累托解集,且解具有更好的多样性和收敛性.
Abstract
Aiming at the shortcomings of the algorithm,such as easily falling into the local optimum and failing to maintain diver-sity,an immune algorithm based on directional sampling and adaptive selection(DSASIA)was proposed.The online population information was used to dynamically select individuals.The adaptive proportional cloning was used to balance the global and local search capabilities to ensure convergence.The directional sampling strategy was adopted to identify the offspring and ensure the diversity.Compared with the other four multi-objective optimization algorithms on 13 test functions,experimental results show that DSASIA algorithm can quickly find the Pareto solution set,and the solutions have better diversity and convergence.
关键词
多样性/定向采样/自适应选择/免疫/自适应比例克隆/收敛性/多目标优化Key words
diversity/directional sampling/adaptive selection/immune/adaptive proportional cloning/convergence/multi-objective optimization引用本文复制引用
出版年
2024
NETL
NSTL
万方数据