多目标优化非支配集构造方法的研究进展  

Research advance of multi-objective optimization non-dominated set construction methods

作　　者：李志强[1] 蔺想红[1] 

机构地区：[1]西北师范大学计算机科学与工程学院,兰州730070

出　　处：《计算机工程与应用》2013年第19期31-35,共5页Computer Engineering and Applications

基　　金：国家自然科学基金(No.61165002);甘肃省自然科学基金(No.1010RJZA019)

摘　　要：多目标优化非支配集的构造是多目标进化算法研究领域的一个重要步骤,旨在研究用多目标进化算法解决多目标优化问题的效率。对多目标优化问题进行了描述并且给出了求解算法的一般框架,结合研究现状讨论了目前该领域几种主要的基于Pareto非支配集的构造算法,以及它们的计算时间复杂度;总结并展望了该领域未来的发展趋势。Constructing the multi-objective optimization non-dominated set is an important step in the Multi-Objective Evolu-tionary Algorithm（MOEA）. It aims to study the operational efficiency to solve Multi-objective Optimization Problem（MOP） by MOEA. Firstly, the MOP is described as well as the basic framework of solving algorithm is given. Next, several non-domi-nated set building methods based on Pareto are discussed including their computational complexity. Finally, the future trends of this research filed are concluded and prospected.

关 键 词：多目标进化算法(MOEA) 多目标优化问题(MOP) 非支配集 PARETO前沿 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]