Journal Home > Volume 2 , Issue 1

Monitoring the operating status of a High-Speed Train (HST) at any moment is necessary to ensure its security. Multi-channel vibration signals are collected by sensors installed on bogies and beneficial information are extracted to determine the running condition. Based on multi-view clustering and considering different views of complementary information, this study proposes a Multi-view Kernel Fuzzy C-Means (MvKFCM) model for condition recognition of the HST bogie. First, fast Fourier transform coefficients of HST vibration signals of all channels are extracted. Then, the fuzzy classification coefficient of every channel is calculated after clustering to select the appropriate channels. Finally, the selected channels are used to cluster by MvKFCM and the conditions of HST are determined. Experimental results show that the selection is effective to maintain rich feature information and remove redundancy. Furthermore, the condition recognition rate of MvKFCM is higher than that of single-view and four other multiple-view clustering algorithms.


menu
Abstract
Full text
Outline
About this article

Condition Recognition of High-Speed Train Bogie Based on Multi-View Kernel FCM

Show Author's information Qi RaoYan Yang( )Yongquan Jiang
School of Information Science and Technology, Southwest Jiaotong University, Chengdu 611756, China.
State Key Laboratory of Traction Power, Southwest Jiaotong University, Chengdu 611756, China.

Abstract

Monitoring the operating status of a High-Speed Train (HST) at any moment is necessary to ensure its security. Multi-channel vibration signals are collected by sensors installed on bogies and beneficial information are extracted to determine the running condition. Based on multi-view clustering and considering different views of complementary information, this study proposes a Multi-view Kernel Fuzzy C-Means (MvKFCM) model for condition recognition of the HST bogie. First, fast Fourier transform coefficients of HST vibration signals of all channels are extracted. Then, the fuzzy classification coefficient of every channel is calculated after clustering to select the appropriate channels. Finally, the selected channels are used to cluster by MvKFCM and the conditions of HST are determined. Experimental results show that the selection is effective to maintain rich feature information and remove redundancy. Furthermore, the condition recognition rate of MvKFCM is higher than that of single-view and four other multiple-view clustering algorithms.

Keywords: multi-view clustering, fuzzy clustering, High-Speed Train (HST), condition recognition

References(26)

[1]
S. L. Sun, A survey of multi-view machine learning, Neural Comput. Appl., vol. 23, nos. 7&8, pp. 2031-2038, 2013.
[2]
J. L. Xu, J. W. Han, and F. P. Nie, Multi-view feature learning with discriminative regularization, in Proc. 26th Int. Joint Conf. Artificial Intelligence, Melbourne, Australia, 2017, pp. 3161-3167.
DOI
[3]
Y. T. Wang and L. H. Chen, Multi-view fuzzy clustering with minimax optimization for effective clustering of data from multiple sources, Expert Syst. Appl., vol. 72, pp. 457-466, 2017.
[4]
C. X. Yan, M. N. Luo, H. Liu, Z. H. Li, and Q. H. Zheng, Top-k multi-class SVM using multiple features, Inf. Sci., vol. 432, pp. 479-494, 2017.
[5]
J. Zhao, X. J. Xie, X. Xu, and S. L. Sun, Multi-view learning overview: Recent progress and new challenges, Inf. Fusion, vol. 38, pp. 43-54, 2017.
[6]
Z. Zheng, W. L. Jiang, H. S. Hu, Y. Zhu, and Y. Li, Research on rolling bearings fault diagnosis method based on EEMD morphological spectrum and kernel fuzzy C-means clustering, (in Chinese), Journal of Vibration Engineering, vol. 28, no. 2, pp. 324-330, 2015.
[7]
W. L. Jiang, Z. W. Wang, Y. Zhu, Z. Zheng, and B. Zhang, Fault recognition method for rolling bearing integrating VMD denoising and FCM clustering, J. Inf. Comput. Sci., vol. 12, no. 16, pp. 5967-5975, 2015.
[8]
C. Xu, D. C. Tao, and C. Xu, A survey on multi-view learning, arXiv preprint arXiv: 1304.5634, 2013.
[9]
J. J. Zhao, Y. Yang, T. R. Li, and W. D. Jin, Application of empirical mode decomposition and fuzzy entropy to high-speed rail fault diagnosis, in Foundations of Intelligent Systems, Z. K. Wen and T. R. Li, eds. Springer, 2014, pp. 93-103.
DOI
[10]
P. Yu, W. D. Jin, and N. Qin, High-speed train running gear fault feature extraction based on EEMD denoising and manifold learning, (in Chinese), Journal of the China Railway Society, vol. 38, no. 4, pp. 16-21, 2016.
[11]
C. Guo, Y. Yang, H. Pan, T. R. Li, and W. D. Jin, Fault analysis of high speed train with DBN hierarchical ensemble, in Proc. 2016 Int. Joint Conf. Neural Networks, Vancouver, Canada, 2016, pp. 2552-2559.
DOI
[12]
Y. B. Liu, Q. Qian, F. Liu, S. L. Lu, and Y. Y. Fu, Wayside acoustic fault diagnosis of train wheel bearing based on Doppler effect correction and fault-relevant information enhancement, Adv. Mech. Eng., vol. 9, no. 11, pp. 1-15, 2017.
[13]
D. Y. Chen, J. H. Lin, and Y. P. Li, Modified complementary ensemble empirical mode decomposition and intrinsic mode functions evaluation index for high-speed train gearbox fault diagnosis, J. Sound Vib., vol. 424, pp. 192-207, 2018.
[14]
C. Guo, Y. Yang, Y. Jiang, and T. Li, Condition analysis of high-speed train based on similarity ratio and MDBN, Comput. Sci. Eng., 2018, (in press).
[15]
W. C. Xiao, Y. Yang, H. J. Wang, T. R. Li, and H. L. Xing, Semi-supervised hierarchical clustering ensemble and its application, Neurocomputing, vol. 173, pp. 1362-1376, 2016.
[16]
Y. Zhao, Z. H. Guo, and J. M. Yan, Vibration signal analysis and fault diagnosis of bogies of the high-speed train based on deep neural networks, J. Vib., vol. 19, no. 4, pp. 2456-2474, 2017.
[17]
H. X. Hu, B. Tang, X. J. Gong, W. Wei, and H. H. Wang, Intelligent fault diagnosis of the high-speed train with big data based on deep neural networks, IEEE Trans. Industr. Inform., vol. 13, no. 4, pp. 2106-2116, 2017.
[18]
Y. Yang and H. Wang, Multi-view clustering: A survey, Big Data Mining and Analytics., vol. 1, no. 2, pp. 83-107, 2018.
[19]
G. Tzortzis and A. Likas, Kernel-based weighted multi-view clustering, in Proc. 2012 IEEE 12th Int. Conf. Data Mining, Brussels, Belgium, 2012, pp. 675-684.
DOI
[20]
H. Wang, F. P. Nie, and H. Huang, Multi-view clustering and feature learning via structured sparsity, in Proc. 30th Int. Conf. Machine Learning, Atlanta, GA, USA, 2013, pp. 352-360.
[21]
L. Du, P. Zhou, L. Shi, and H. M. Wang, Robust multiple kernel k-means using L21-norm, in Proc. 24th Int. Joint Conf. Artificial Intelligence, Buenos Aires, Argentina, 2015, pp. 3476-3482.
[22]
D. Graves and W. Pedrycz, Fuzzy C-means, Gustafson-kessel FCM, and kernel-based FCM: A comparative study, in Analysis and Design of Intelligent Systems using Soft Computing Techniques, P. Melin, O. Castillo, E. G. Ramírez, J. Kacprzyk, and W. Pedrycz, eds. Springer, 2007, pp. 140-149.
[23]
E. Trauwaert, On the meaning of Dunn’s partition coefficient for fuzzy clusters, Fuzzy Sets Syst., vol. 25, no. 2, pp. 217-242, 1988.
[24]
K. R. Muller, S. Mika, G. Ratsch, K. Tsuda, and B. Scholkopf, An introduction to kernel-based learning algorithms, IEEE Trans. Neural Netw., vol. 12, no. 2, pp. 181-201, 2001.
[25]
X. J. Chen, X. F. Xu, J. Z. Huang, and Y. M. Ye, TW-k-means: Automated two-level variable weighting clustering algorithm for multiview data, IEEE Trans. Knowl. Data Eng., vol. 25, no. 4, pp. 932-944, 2013.
[26]
X. Cai, F. P. Nie, and H. Huang, Multi-view k-means clustering on big data, in Proc. 23th Int. Joint Conf. Artificial Intelligence, Beijing, China, 2013, pp. 2598-2604.
Publication history
Copyright
Acknowledgements
Rights and permissions

Publication history

Received: 09 March 2018
Accepted: 12 April 2018
Published: 15 October 2018
Issue date: March 2019

Copyright

© The author(s) 2019

Acknowledgements

This work was supported in part by the National Natural Science Foundation of China (Nos. 61572407 and 61134002).

Rights and permissions

Return