一种结合谱聚类与关联规则的轴承故障诊断方法  

Bearing Fault Diagnosis Method Combining Spectral Clustering and Association Rules

作　　者：徐秀芳[1] 徐丹妍 徐森[1] 郭乃瑄 许贺洋 XU Xiufang;XU Danyan;XU Sen;GUO Naixuan;XU Heyang(School of Information Engineering,Yancheng Institute of Technology,Yancheng 224051,China)

机构地区：[1]盐城工学院信息工程学院,江苏盐城224051

出　　处：《计算机测量与控制》2023年第1期51-58,共8页Computer Measurement &Control

基　　金：国家自然科学基金项目(62076215);江苏省高等学校自然科学研究面上项目(21KJD520006);2021年度未来网络科研基金(FNSRFP-2021-YB-46);盐城工学院研究生培养创新工程项目(SJCX21_XZ018);横向项目合同编号(2022032809);教育部产学研合作项目(202102594034)。

摘　　要：针对现阶段机械设备轴承故障诊断方法难以挖掘隐含特征、诊断精准度低等问题,将谱聚类(spectral clustering, SC)算法与关联规则算法Apriori相结合,提出SC-Apriori算法;首先根据美国西储大学轴承数据中心网站公开发布的轴承故障数据集,选取0负载下的数据,计算得到滚动轴承振动信号的9个时域特征和3个频域特征;其次使用Pearson相关系数进行特征筛选,留下9个有效特征,再利用SC-Apriori算法挖掘出训练数据集中轴承不同特征数据之间的关联关系,并引入提升度来去除冗余的关联规则,进而构建一个规则库;再将测试数据进行处理,并与已建立的规则库进行比对,根据匹配率来判断其故障类型;在测试数据上的实验结果表明,与已有算法相比,文章设计的SC-Apriori算法挖掘出的规则数量大幅减少,匹配速度更快,且匹配效果更好。Aimed at the problems of difficulty in mining implicit features and low diagnostic accuracy of existing mechanical equipment bearing fault diagnosis methods, a SC-Apriori algorithm was proposed by combining the spectral clustering(SC) algorithm with the association rule(Apriori) algorithm. Firstly, based on the bearing fault dataset publicly released on the website of Bearing Data Centre of Western Reserve University, the unload data were selected to calculate and obtain nine time-domain features and three frequency-domain features of the rolling bearing vibration signal. Secondly, the Pearson correlation coefficient was used to filter the features and reserve nine effective features, and then the SC-Apriori algorithm was used to mine the association relationship between the different features of bearings in the training dataset. and the boosting was introduced to remove the redundant association rules and construct a rule base. Then the test data were processed and compared with the established rule base to judge their fault types by the matching rate. Experimental results on the test data show that compared with existing algorithms, the SC-Apriori algorithm has the advantages of mining a significantly reduced number of rules, fast matching speed and better matching effect.

关 键 词：轴承故障诊断 数据挖掘 关联规则 谱聚类算法 提升度 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]