KPCA和耦合隐马尔科夫模型在轴承故障诊断中的应用  

Application of KPCA and coupled hidden Markov model in bearing fault diagnosis

作　　者：刘韬[1,2] 陈进[2] 董广明[2] 

机构地区：[1]昆明理工大学机电工程学院,昆明650500 [2]上海交通大学机械系统与振动国家重点实验室,上海200240

出　　处：《振动与冲击》2014年第21期85-89,共5页Journal of Vibration and Shock

基　　金：国家自然科学基金项目(51035007;51175329;51105243)

摘　　要：针对多通道数据的有效融合能够更加准确地诊断轴承的故障,提出了一种基于KPCA和耦合隐马尔可夫模型(CHMM)的轴承故障诊断方法。首先,分别对轴承各通道的振动信号进行特征提取,获得特征向量。然后采用KPCA对各通道的特征向量分别进行特征约减,获取主要的信息成分。最后,利用CHMM对多通道信息进行融合和故障诊断。通过对滚动轴承在正常、内圈故障、外圈故障和滚动体故障状态下实验数据的分析表明,该方法能够更加有效地诊断轴承的故障。The fusion of multi-channel bearing monitoring information can obtain more accurate results in bearing fault diagnosis.Here,a rolling element bearing fault diagnosis scheme based on KPCA and coupled hidden Markov model （CHMM） was presented.At first,the features were extracted from bearing vibration signals of multi-channel, respectively.Then,the KPCA was utilized to reduce the feature dimensions.At last,the new KPCA features were input into a CHMMto be fused and to diagnose bearing faults.The data acquired from bearings’states under normal conditions, and states with inner race faults,outer race faults and rolling body faults were analyzed.The results demonstrated the effectiveness and validity of the proposed method.

关 键 词：核主成分分析 耦合隐马尔可夫模型 滚动轴承 故障诊断 

分 类 号：TH165.3[机械工程—机械制造及自动化] TN911[电子电信—通信与信息系统]