面向云制造的资源集群聚类方法研究  

Research on clustering method of resource service cluster in cloud manufacturing

作　　者：苑明海[1] 李子晨 黄涵钰 裴凤雀 俞红焱 YUAN Minghai;LI Zichen;HUANG Hanyu;PEI Fengque;YU Hongyan(College of Mechanical Electrical Engineering,Hohai University,Changzhou 213000,CHN)

机构地区：[1]河海大学机电工程学院,江苏常州213000

出　　处：《制造技术与机床》2023年第10期41-47,共7页Manufacturing Technology & Machine Tool

基　　金：教育部人文社会科学规划基金(21YJA630111);江苏省自然科学基金项目(BK20201162);常州市科技计划项目(CM20223014和CJ20220207)。

摘　　要：针对云制造平台海量多样的服务资源分类界限模糊的问题,分析了云服务和制造资源之间的关系,提出云制造下混合式资源服务聚集模型。此外,文章基于k-means聚类算法建立了聚类有效性评估函数;针对k-means聚类算法对初始聚簇中心敏感易陷入局部最优的缺点,引入蛙跳算法确定初始聚簇中心,利用反向解扩大初始蛙群的搜索范围,结合最优解均值改进族群最差蛙的优化,提高族群的信息共享能力,结合改进后的蛙跳算法和k-means迭代,提出一种基于蛙跳算法改进的k-means聚类算法。最后,以两种数据集和云平台上同类机床资源为例,验证了所提聚类算法的有效性和可行性。In view of the fuzzy classification boundaries of massive and diverse cloud manufacturing(CMfg)service resources,this paper analyzes the relationship between cloud services and manufacturing resources,and establishes a CMfg hybrid service aggregation model based on the service resources aggregation type.In addition,this paper establishes a clustering validity evaluation function based on k�means clustering algorithm.Aiming at the disadvantage that k-means clustering algorithm is sensitive to the initial clustering center,the shuffled frog leaping algorithm(SFLA)is introduced to determine the initial clustering center.The inverse solution is used to expand the search range of the initial frog population,and the optimization of the worst frog population is improved by combining the mean value of the optimal solution.Based on the improved leapfrog algorithm and k-means iteration,an improved k�means clustering algorithm based on leapfrog algorithm is proposed.Finally,the validity of the algorithm is verified by the Iris test data set and a self-constructed data set(Self-cd),and the feasibility of the algorithm is illustrated by the application of lathe resource on the CMfg platform.

关 键 词：云制造 制造资源 聚类方法 K-MEANS 蛙跳算法 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]