基于角度邻域的多目标差分进化算法

发布时间：2018-02-08 17:13

本文关键词： 差分进化角度邻域外部存档多目标优化　出处：《控制理论与应用》2017年01期 　论文类型：期刊论文

【摘要】：针对如何实现差分进化算法求解多目标优化问题,提出了一种基于角度邻域的多目标差分进化算法,通过在选择操作中引入弱支配概念,实现了对多目标优化问题的求解.该算法通过计算目标空间中个体与权重向量的夹角来确定每个个体的邻域,并在此基础上引入了基于角度邻域的变异策略,使个体的变异在邻域内进行,保证进化方向.此外,该算法创建了一个外部存档用来保存进化过程中的非支配解,并定期对外部存档进行维护,大大改善了解集的分布性.大量的数值仿真实验结果表明通过角度确定邻域的方法比通过欧氏距离确定邻域的方法更加有效,算法所得解集的收敛性和分布性也均明显优于基于分解的差分多目标进化算法(multiobjective evolutionary algorithm based on decomposition and differential evolution,MOEA/D DE)和非支配排序算法Ⅱ(nondominated sorting genetic algorithm II,NSGA).
[Abstract]:In this paper, a multi-objective differential evolutionary algorithm based on angle neighborhood is proposed to solve the multi-objective optimization problem. The concept of weak dominance is introduced into the selection operation. The algorithm determines the neighborhood of each individual by calculating the angle between individual and weight vector in the target space, and then introduces a mutation strategy based on angle neighborhood. In addition, the algorithm creates an external archive to preserve the non-dominant solution in the evolution process and regularly maintains the external archive. A large number of numerical simulation results show that the method to determine neighborhood by angle is more effective than that by Euclidean distance. The convergence and distribution of the solution set obtained by the proposed algorithm are also superior to those of the multiobjective evolutionary algorithm based on decomposition and differential evolutional MOEA-DED and the nondominated sorting genetic algorithm II NSGAA.
【作者单位】：燕山大学电气工程学院工业计算机控制工程河北省重点实验室;唐山学院计算机科学与技术系;燕山大学国家冷轧板带装备及工艺工程技术研究中心;
【基金】：国家自然科学基金项目(U1260203) 河北省高等学校创新团队领军人才培育计划项目(LJRC013) 国家冷轧板带装备及工艺工程技术研究中心开放课题(2012005)资助~~
【分类号】：TP18

【相似文献】