求解多目标优化问题的GTSPA混合算法  

GTSPA Hybrid Algorithm for Solving Multi-objective Optimization Problem

作　　者：叶雪梅[1] 田甜[1] 陈柏松[1] 

机构地区：[1]第二炮兵工程学院计算机室,陕西西安710025

出　　处：《微电子学与计算机》2010年第6期167-169,173,共4页Microelectronics & Computer

摘　　要：为了有效求解多目标优化问题,文中结合遗传算法、禁忌搜索算法以及粒子群算法的优势,设计了一种新型的混合算法GTSPA,并对所设计的GTSPA混合算法的总体步骤进行了描述.通过从解的质量、算法的收敛性以及算法的复杂度这几方面对算法进行分析,可以发现:GTSPA混合算法是以较大概率收敛于全局最优的,具有很好的收敛性,并且运算相当快、计算复杂度并不高,具有很好的优势.In order to solve multi-objective optimization problem effectively,this article designs a new hybrid algorithm GTSPA which combines the advantages and the ideology of the three algorithm — GA,TSA and PSO.And then the paper depicts the steps of the algorithm.Through the analysis of the quality of the answer,the astringency and the complication of the algorithm,we can find：the probability of the global optimal solution is high,the astringency is fine and GTSPA has low complexity of computer.

关 键 词：多目标优化问题 遗传算法 禁忌算法 粒子群算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]