文献详情
Bearing diagnostics: A method based on differential geometry
文献类型期刊
作者Tian, Ye[1];Wang, Zili[2];Lu, Chen[3];Wang, Zhipeng[4]
机构
通讯作者Lu, C (reprint author), Beihang Univ, Sch Reliabil & Syst Engn, Beijing 100191, Peoples R China.
来源信息年:2016  卷:80  页码范围:377-391  
期刊信息MECHANICAL SYSTEMS AND SIGNAL PROCESSING影响因子和分区  ISSN:0888-3270
关键词Bearings; Noise reduction; Feature extraction; Fault diagnosis; Health assessment; Differential geometry
增刊正刊
摘要The structures around bearings are complex, and the working environment is variable. These conditions cause the collected vibration signals to become nonlinear, non-stationary, and chaotic characteristics that make noise reduction, feature extraction, fault diagnosis, and health assessment significantly challenging. Thus, a set of differential geometry-based methods with superiorities in nonlinear analysis is presented in this study. For noise reduction, the Local Projection method is modified by both selecting the neighborhood radius based on empirical mode decomposition and determining noise subspace constrained by neighborhood distribution information. For feature extraction, Hessian locally linear embedding is introduced to acquire manifold features from the manifold topological structures, and singular values of eigenmatrices as well as several specific frequency amplitudes in spectrograms are extracted subsequently to reduce the complexity of the manifold features. For fault diagnosis, information geometry-based support vector machine is applied to classify the fault states. For health assessment, the manifold distance is employed to represent the health information; the Gaussian mixture model is utilized to calculate the confidence values, which directly reflect the health status. Case studies on Lorenz signals and vibration datasets of bearings demonstrate the effectiveness of the proposed methods. (C) 2016 Elsevier Ltd. All rights reserved.
收录情况SCIE(WOS:000379556200022)  EI(20162302459360)  
所属部门可靠性与系统工程学院
DOI10.1016/j.ymssp.2016.05.010
学科工程:机械
百度学术Bearing diagnostics: A method based on differential geometry
语言外文
被引频次2
人气指数40
浏览次数40
基金Fundamental Research Funds for the Central Universities [YWF-16-BJ-J-18]; National Natural Science Foundation of China [51575021]; Technology Foundation Program of National Defense [Z1320138002]
全部评论(0 条评论)
作者其他论文

Light and pH Cooperative Nanofluidic Diode Using a Spiropyran-Functionalized Single Nanochannel.

Zhang, Minghui;Hou, Xu;Wang, Jingtao,等.ADVANCED MATERIALS.2012,24(18),2424-2428.

Health assessment and fault diagnosis for centrifugal pumps using Softmax regression.

Ma, Jian;Lu, Chen;Zhang, Wenjin,等.JOURNAL OF VIBROENGINEERING.2014,16(3),1464-1474.

Satellite Navigation Regional Augmentation System performance analysis based on BDS.

Wang, Zhipeng;Zhu, Yanbo;Xue, Rui,等.27th International Technical Meeting of the Satellite-Division of the Institute-of-Navigation (ION GNSS).2014,3389-3398.

Chaotic Parallel Support Vector Machine and Its Application for Fault Diagnosis of Hydraulic Pump.

Wang, Zili;Wang, Zhipeng.IEEE International Conference on Prognostics and Health Management (PHM).2013.

System maintenance drive simulation method in intermittent operations under PHM supervision.

Cui, Yiqian;Shi, Junyou;Wang, Zili,等.2012 3rd Annual IEEE Prognostics and System Health Management Conference, PHM-2012.2012.

登录