基于多域空间状态矩阵奇异值与局部保持投影的滚动轴承故障特征提取方法  

Fault Feature Extraction Method of Rolling Element Bearing Based on Singular Values of Multiple-domain Spatial Condition Matrix and Locality Preserving Projection

作　　者：俞昆 谭继文[1] 李善[1] 

机构地区：[1]青岛理工大学机械工程学院,山东青岛266525

出　　处：《机床与液压》2017年第9期167-174,共8页Machine Tool & Hydraulics

基　　金：国家自然科学基金资助项目(51075220);高等学校博士学科点专项科研基金项目(20123721110001);青岛市科技计划基础研究资助项目(12-1-4-4-(3)-JCH)

摘　　要：针对滚动轴承振动信号复杂且难以从中提取有效故障特征的问题,提出了一种总体经验模态分解(EEMD)、奇异值分解(SVD)和局部保持投影(LPP)相结合的故障特征提取方法。首先,对振动信号进行EEMD分解,利用EEMD分解后的固有模态分量(IMF)分别构造时域、频域和时频域空间状态矩阵;其次,利用SVD提炼时域、频域和时频域空间状态矩阵中的故障信息,筛选其中累加百分比大于90%的奇异值组成多域有效奇异值数组,构造多域奇异值特征矩阵;然后,利用LPP约简多域奇异值特征矩阵,提取低维、高区分度的故障特征;最后,利用支持向量机(SVM)对提出的故障特征提取方法进行评估。实验结果证明了该方法提取的故障特征可有效反映滚动轴承的故障状态。A combined method was proposed which employs ensemble empirical mode decomposition （EEMD） , singular value decomposition （SVD） and locality preserving projection （LPP） to extract useful fault feature from the complex vibration signals of roll-ing element bearings with problems of extraction in difficulty. Firstly, the vibration signals were decomposed with EEMD into a set of intrinsic mode functions （IMFs） , which then were utilized to construct the time-domain and frequency-domain spatial condition ma-trix, as well as the time-frequency domain spatial condition matrix. Secondly, SVD was used to extract the fault information of multiple- domain spatial condition matrix and among which selected the effective SVs which cumulative percentage were greater than 90% consti-tuted the multiple-domain SV feature matrix. Thirdly, LPP was used to extract the low-dimension and high-separability fault features from multiple-domain SV feature matrix. Finally, support vector machine （SVM） was used to evaluate the fault feature extraction meth-od proposed. The experimental result illustrate that the fault feature extracted according to this method can effectively reflect the fault patterns of rolling element bearings.

关 键 词：总体经验模态分解 奇异值分解 空间状态矩阵 局部保持投影 特征提取 

分 类 号：TH165.3[机械工程—机械制造及自动化]