一种基于截断机制的稳态优化算法求解多目标优化问题  

A STEADY-STATE OPTIMIZATION ALGORITHM BASED ON TRUNCATION MECHANISM FOR THE MULI-OBJECTIVE OPTIMIZATION PROBLEM

作　　者：荆东星[1] 张清安[1] Jing Dongxing;Zhang Qing’an(Department of Computer Science,Xiangxi Vocational and Technical College for Nationalities,Xiangxi 416000,Hunan,China)

机构地区：[1]湘西民族职业技术学院计算机系,湖南湘西416000

出　　处：《计算机应用与软件》2018年第11期283-288,295,共7页Computer Applications and Software

基　　金：湖南省教育厅科学研究项目(17c15)

摘　　要：近年来,多目标优化问题的求解已成为优化领域的一个重要研究方向。为设计一种优化算法使得解集快速收敛并均匀分布于问题的最优Pareto面上,提出一种基于截断机制的稳态优化算法求解多目标优化问题。与传统的稳态算法(ε-MOEA)相比,该算法具有无参、解集数量可控以及良好广泛性等特点。实验结果表明,该算法在多目标优化问题中表现良好,与几个经典算法相比具有一定的竞争力。In recent years,the solution of multi-objective optimization problem has become an important research direction in the field of optimization.To design an optimization algorithm to make the solution set converge quickly and uniformly distribute on the optimal Pareto of the problem,a steady-state optimization algorithm based on truncation mechanism was proposed to solve the multi-objective optimization problem in this paper.Compared with the traditional steady-state algorithm,ε-MOEA,the proposed algorithm had the features of no-references,controllable number of solution sets and good universality.The experimental results show that the algorithm performs well in multi-objective optimization and is competitive with several classical algorithms.

关 键 词：多目标优化问题 优化算法 稳态算法 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]