Feature fusion using kernel joint approximate diagonalization of eigen-matrices for rolling bearing fault identification

被引:56
作者
Liu, Yongbin [1 ,2 ]
He, Bing [1 ]
Liu, Fang [1 ,2 ]
Lu, Siliang [1 ,2 ]
Zhao, Yilei [1 ]
机构
[1] Anhui Univ, Coll Elect Engn & Automat, Hefei 230601, Anhui, Peoples R China
[2] Anhui Univ, Natl Engn Lab Energy Saving Motor & Control Techn, Hefei 230601, Peoples R China
基金
中国国家自然科学基金;
关键词
NONLINEAR DIMENSIONALITY REDUCTION; DIAGNOSIS; CLASSIFICATION; MANIFOLD; WAVELET; PCA; LDA;
D O I
10.1016/j.jsv.2016.09.018
中图分类号
O42 [声学];
学科分类号
070206 ; 082403 ;
摘要
Fault pattern identification is a crucial step for the intelligent fault diagnosis of real-time health conditions in monitoring a mechanical system. However, many challenges exist in extracting the effective feature from vibration signals for fault recognition. A new feature fusion method is proposed in this study to extract new features using kernel joint approximate diagonalization of eigen-matrices (KJADE). In the method, the input space that is composed of original features is mapped into a high-dimensional feature space by nonlinear mapping. Then, the new features can be estimated through the eigen-decomposition of the fourth-order cumulative kernel matrix obtained from the feature space. Therefore, the proposed method could be used to reduce data redundancy because it extracts the inherent pattern structure of different fault classes as it is nonlinear by nature. The integration evaluation factor of between-class and within-class scatters (SS) is employed to depict the clustering performance quantitatively, and the new feature subset extracted by the proposed method is fed into a multi-class support vector machine for fault pattern identification. Finally, the effectiveness of the proposed method is verified by experimental vibration signals with different bearing fault types and seventies. Results of several cases show that the [(JADE algorithm is efficient in feature fusion for bearing fault identification. (C) 2016 Elsevier Ltd. All rights reserved.
引用
收藏
页码:389 / 401
页数:13
相关论文
共 36 条
  • [1] Stellar spectral subclasses classification based on Isomap and SVM
    Bu, Yude
    Chen, Fuqiang
    Pan, Jingchang
    [J]. NEW ASTRONOMY, 2014, 28 : 35 - 43
  • [2] BLIND BEAMFORMING FOR NON-GAUSSIAN SIGNALS
    CARDOSO, JF
    SOULOUMIAC, A
    [J]. IEE PROCEEDINGS-F RADAR AND SIGNAL PROCESSING, 1993, 140 (06) : 362 - 370
  • [3] High-order contrasts for independent component analysis
    Cardoso, JF
    [J]. NEURAL COMPUTATION, 1999, 11 (01) : 157 - 192
  • [4] A novel damage detection approach by using Volterra kernel functions based analysis
    Cheng, C. M.
    Peng, Z. K.
    Dong, X. J.
    Zhang, W. M.
    Meng, G.
    [J]. JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS, 2015, 352 (08): : 3098 - 3112
  • [5] Evaluation of principal component analysis and neural network performance for bearing fault diagnosis from vibration signal processed by RS and DF analyses
    de Moura, E. P.
    Souto, C. R.
    Silva, A. A.
    Irmao, M. A. S.
    [J]. MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2011, 25 (05) : 1765 - 1772
  • [6] A fusion feature and its improvement based on locality preserving projections for rolling element bearing fault classification
    Ding, Xiaoxi
    He, Qingbo
    Luo, Nianwu
    [J]. JOURNAL OF SOUND AND VIBRATION, 2015, 335 : 367 - 383
  • [7] Dong S.J., 2014, MATH PROBL ENG, V2014, P805
  • [8] Dong SJ, 2015, J CHIN SOC MECH ENG, V36, P299
  • [9] Bearing fault classification using ANN-based Hilbert footprint analysis
    Dubey, Rahul
    Agrawal, Dheeraj
    [J]. IET SCIENCE MEASUREMENT & TECHNOLOGY, 2015, 9 (08) : 1016 - 1022
  • [10] Analog circuit fault diagnosis using LDA and OAOSVM approach
    Guo, Ke
    Zhu, Yi
    San, Ye
    [J]. MECHATRONICS AND INTELLIGENT MATERIALS II, PTS 1-6, 2012, 490-495 : 1130 - +