欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
提出一种双链结构的多目标进化算法（DCMOEA）.该算法采用双链结构表示个体,执行过程中无需设置外部归档集合,并采用ε支配策略保持解群的多样性.DCMOEA与MOEA/D、NSGA-II、SPEA2和PAES一同在4个2-目标ZDT函数和4个3-目标DTLZ问题上进行实验,并从算法所获解集的收敛性、分布均匀性和宽广性3个方面进行比较,仿真实验结果表明了DCMOEA的综合性能最好,是一种颇具竞争力的多目标进化算法.
英文摘要：
A multi-objective evolutionary algorithm based on double chalns（DCMOEA） is proposed, which is characterized with populated individuals based on double chains, no external archive is needed, and the ε dominance mechanism is adopted for preserving population diversity. The DCMOEA is compared with MOEA/D, NSGA-II, SPEA2 and PAES simultaneously on the platform employing four 2-objective ZDT test functions and four 3-objective DTLZ instances from three aspects including convergence, spacing and maximum spread. Experimental results show that the DCMOEA has the best comprehensive performance among five multi-objective evolutionary algorithms, and is a promising multi-objective evolutionary algorithm.
同期刊论文项目
同项目期刊论文
期刊信息
位置：