Feature Extraction of Impulse Faults for Vibration Signals Based on Sparse Non-Negative Tensor Factorization

被引:10
|
作者
Liang, Lin [1 ,2 ]
Wen, Haobin [1 ]
Liu, Fei [1 ]
Li, Guang [1 ]
Li, Maolin [3 ]
机构
[1] Xi An Jiao Tong Univ, Sch Mech Engn, Xian 710049, Shaanxi, Peoples R China
[2] Xi An Jiao Tong Univ, Key Lab Educ Minist Modern Design & Rotor Bearing, Xian 710049, Shaanxi, Peoples R China
[3] Xi An Jiao Tong Univ, Engn Workshop, Xian 710049, Shaanxi, Peoples R China
来源
APPLIED SCIENCES-BASEL | 2019年 / 9卷 / 18期
基金
中国国家自然科学基金; 中国博士后科学基金;
关键词
sparse non-negative tensor factorization; feature extraction; impulse fault; phase space reconstruction; time-frequency distribution; ALGORITHMS; DIAGNOSIS; MATRIX;
D O I
10.3390/app9183642
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
The incipient damages of mechanical equipment excite weak impulse vibration, which is hidden, almost unobservable, in the collected signal, making fault detection and failure prevention at the inchoate stage rather challenging. Traditional feature extraction techniques, such as bandpass filtering and time-frequency analysis, are suitable for matrix processing but challenged by the higher-order data. To tackle these problems, a novel method of impulse feature extraction for vibration signals, based on sparse non-negative tensor factorization is presented in this paper. Primarily, the phase space reconstruction and the short time Fourier transform are successively employed to convert the original signal into time-frequency distributions, which are further arranged into a three-way tensor to obtain a time-frequency multi-aspect array. The tensor is decomposed by sparse non-negative tensor factorization via hierarchical alternating least squares algorithm, after which the latent components are reconstructed from the factors by the inverse short time Fourier transform and eventually help extract the impulse feature through envelope analysis. For performance verification, the experimental analysis on the bearing datasets and the swashplate piston pump has confirmed the effectiveness of the proposed method. Comparisons to the traditional methods, including maximum correlated kurtosis deconvolution, singular value decomposition, and maximum spectrum kurtosis, also suggest its better performance of feature extraction.
引用
收藏
页数:26
相关论文
共 50 条
  • [1] A Non-negative Tensor Factorization Approach to Feature Extraction for Image Analysis
    Ang, Andersen Man Shun
    Hung, Yeung Sam
    Zhang, Zhiguo
    2016 IEEE INTERNATIONAL CONFERENCE ON DIGITAL SIGNAL PROCESSING (DSP), 2016, : 168 - 171
  • [2] Spatial feature extraction non-negative tensor factorization for hyperspectral unmixing
    Wang, Jin-Ju
    Wang, Ding-Cheng
    Huang, Ting-Zhu
    Huang, Jie
    APPLIED MATHEMATICAL MODELLING, 2022, 103 : 18 - 35
  • [3] Supervised non-negative tensor factorization for automatic hyperspectral feature extraction and target discrimination
    Anderson, Dylan
    Bapst, Aleksander
    Coon, Joshua
    Pung, Aaron
    Kudenov, Michael
    ALGORITHMS AND TECHNOLOGIES FOR MULTISPECTRAL, HYPERSPECTRAL, AND ULTRASPECTRAL IMAGERY XXIII, 2017, 10198
  • [4] Non-negative Matrix Semi-tensor Factorization for Image Feature Extraction and Clustering
    Ben, Chi
    Wang, Zhiyuan
    Yang, Xuejun
    Yi, Xiaodong
    PROCEEDINGS OF 2016 CHINESE INTELLIGENT SYSTEMS CONFERENCE, VOL I, 2016, 404 : 381 - 393
  • [5] A Feature Extraction Method for Remnant Particles based on Non-negative Tensor Factorization in Aerospace Electronic Equipments
    Chen, Rui
    Wang, Shujuan
    Zhai, Guofu
    Yi, Shen
    2012 IEEE INTERNATIONAL INSTRUMENTATION AND MEASUREMENT TECHNOLOGY CONFERENCE (I2MTC), 2012, : 1650 - 1653
  • [6] Non-negative tensor factorization for vibration-based local damage detection
    Gabor, Mateusz
    Zdunek, Rafal
    Zimroz, Radoslaw
    Wodecki, Jacek
    Wylomanska, Agnieszka
    MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2023, 198
  • [7] Sparse non-negative tensor factorization using columnwise coordinate descent
    Liu, Ji
    Liu, Jun
    Wonka, Peter
    Ye, Jieping
    PATTERN RECOGNITION, 2012, 45 (01) : 649 - 656
  • [8] FLEXIBLE HALS ALGORITHMS FOR SPARSE NON-NEGATIVE MATRIX/TENSOR FACTORIZATION
    Cichocki, Andrzej
    Phan, Anh Huy
    Caiafa, Cesar
    2008 IEEE WORKSHOP ON MACHINE LEARNING FOR SIGNAL PROCESSING, 2008, : 73 - 78
  • [9] DETECTION OF PUMP FAULTS BASED ON SOUND SIGNALS USING NON-NEGATIVE MATRIX FACTORIZATION
    Vinaya, Anindita Adikaputri
    Nurmaulidah, Fitri
    Arifianto, Dhany
    Arifianti, Qurrotin Ayunina Maulida Okta
    JURNAL TEKNOLOGI-SCIENCES & ENGINEERING, 2020, 82 (02): : 7 - 14
  • [10] Non-negative matrix factorization based text mining: Feature extraction and classification
    Barman, P. C.
    Iqbal, Nadeem
    Lee, Soo-Young
    NEURAL INFORMATION PROCESSING, PT 2, PROCEEDINGS, 2006, 4233 : 703 - 712