基于粗糙集的耦合任务集聚类方法  

A Coupled Task Set Clustering Method Based on Rough Set

作　　者：高维[1] 张剑[1] 黎荣[1] 张义军[1] 

机构地区：[1]西南交通大学机械工程学院,成都610031

出　　处：《机械制造》2017年第7期20-24,29,共6页Machinery

基　　金：国家自然科学基金资助项目(编号:5157051514)

摘　　要：针对耦合任务集识别方法无法处理与评价数据的模糊性和不确定性信息这一问题,提出了基于粗糙集的耦合任务集聚类方法。采用粗糙集方法对评价数据矩阵进行处理,将其转化为粗糙数形式的任务信息依赖度矩阵,然后通过改进的任务关联度算法获得任务关联度矩阵,再对任务关联度矩阵进行聚类,最后通过阈值划分得到耦合任务集。通过高速列车转向架构架的设计,验证了这一方法的有效性。Aiming at the problem that the recognition method for coupled task set can not deal with the ambiguity and uncertainty information of the evaluate data, a coupled task set clustering method based on rough set was proposed. It adopted the rough set method to process the evaluation data matrix, and transformed it into the task information dependency matrix in the form of roughness. Then the task dependency matrix was obtained by the improved task dependency algorithm, the task dependency matrix was clustered, and finally the coupled task set was obtained by threshold division. The effectiveness of this method was verified by the TGV bogie design.

关 键 词：粗糙集 耦合任务集 聚类 

分 类 号：TH122[机械工程—机械设计及理论] TP391[自动化与计算机技术—计算机应用技术]