结合模糊聚类和合作竞争博弈的优化方法  

Optimization Method Based on Fuzzy Clustering and Cooperative Competition Game

作　　者：陈加明[1] 宋小文[2] 徐浩[2] 

机构地区：[1]杭州职业技术学院,杭州310000 [2]浙江大学,杭州310027

出　　处：《中国机械工程》2015年第16期2179-2182,共4页China Mechanical Engineering

摘　　要：针对多目标优化问题转化为博弈问题的难点,即博弈问题的各博弈方战略集的确定问题,利用均匀设计试验法、回归分析和F检验为模糊聚类提供数据基础,将多目标优化设计与博弈分析结合起来,提出了一种多目标优化方法。以减速器优化设计为例,对该方法的性能进行了分析与验证,并与传统单目标优化方法进行对比分析。结果表明,该方法收敛速度快,计算效率高,无需人为确定各目标权重,具有工程应用价值。Aiming at the difficulty of transforming multi-objective optimization problem to game problem which was about determining the strategy set for each partition of game players,a multi-objective optimization method was proposed by using uniform design experimentation,regression analysis and F test statistic to ob-tain data bases for fuzzy clustering and combining cooperative competition game.Taking reducer design opti-mization as an example,performance of the method was analyzed and verified.Compared with the traditional single-objective optimization method,the experimental results show that the multi-objective method has fast convergence rate and high computational efficiency,and it is unneccessary to determine the weight of each tar-get,which has engineering application values.

关 键 词：模糊聚类 合作竞争博弈 均匀设计 回归分析 

分 类 号：TG122[金属学及工艺—金属学]