基于多视图加权聚类集成的高速列车工况识别

doi:10.3969/j.issn.0253-2778.2018.01.005

中国科学技术大学学报 ›› 2018, Vol. 48 ›› Issue (1): 35-41.DOI: 10.3969/j.issn.0253-2778.2018.01.005

基于多视图加权聚类集成的高速列车工况识别

饶齐，杨燕*，滕飞

西南交通大学信息科学与技术学院，四川成都 611756

收稿日期:2017-05-29 修回日期:2017-06-22 出版日期:2018-01-01 发布日期:2018-01-01
通讯作者: 杨燕
作者简介:饶齐，男，1992年生，硕士生，研究方向:数据挖掘. E-mail：rich7777@126.com
基金资助:
国家自然科学基金项目(61572407, 61134002)，国家科技支撑计划课题(2015BAH19F02)资助.

Condition recognition of high-speed train based on multi-view weighted clustering ensemble

RAO Qi, YANG Yan*, TENG Fei

School of Information Science and Technology, Southwest Jiaotong University, Chengdu 611756, China）

Received:2017-05-29 Revised:2017-06-22 Online:2018-01-01 Published:2018-01-01

摘要/Abstract

摘要： 随着中国高速列车行业的快速发展，高速列车运行所产生的安全隐患问题引发了更多的关注．由于利用传感器所采集到的高速列车监测数据具有非线性、非平稳的特点，导致故障工况难以识别，为此提出一种基于加权非负矩阵的多视图聚类集成模型（weighted non-negative matrix factorization, WNMF）来对车体走行部的故障工况进行识别．首先，对振动信号进行频域、时频域、时域的分析，通过快速傅里叶变换、小波包能量、经验模态分解的近似熵和模糊熵、机械统计特征四个方面提取特征向量，构建四个特征视图；其次进行K-means聚类，得到每个视图的结果；再通过聚类成员的贡献度和相似度分别求取各视图的两种权值；最后进行加权的非负矩阵分解集成．实验结果表明，该模型能够有效地识别高速列车的故障工况．

关键词: 高速列车, 工况识别, 多视图, 聚类集成, 特征提取

Abstract: With the rapid development of China's high-speed train industry, some safety problems arising from the high-speed train operation are attracting more attention. Since the monitoring signals of the high-speed trains collected by sensors are nonlinear and non-stationary, it is difficult to identify the fault conditions of high-speed train. Therefore, in this paper, a multi-view clustering ensemble model based on weighted non-negative matrix factorization (WNMF) is proposed to it. Firstly, the vibration signals are analyzed the frequency domain, time-frequency domain and time domain. And the multi-views are obtained by extracting the eigenvector from the four aspects of the vibration signal, which are fast Fourier transform, wavelet packet energy, approximate entropy and fuzzy entropy of empirical mode decomposition, and the mechanical statistical characteristics. And then the clustering result of each view is obtained by the K-means. Secondly, two kinds of weight of the views are generated respectively by the contribution and the similarity of the clustering partitions. Finally, the output results of multiple clustering and the weights are combined for WNMF to ensemble. The experimental results show that the model can better identify fault conditions of high-speed trains.

Key words: high-speed train, condition recognition, multi-view, clustering ensemble, feature extraction

中图分类号:

TP181

饶齐，杨燕*，滕飞. 基于多视图加权聚类集成的高速列车工况识别[J]. 中国科学技术大学学报, 2018, 48(1): 35-41.

RAO Qi, YANG Yan*, TENG Fei. Condition recognition of high-speed train based on multi-view weighted clustering ensemble[J]. Journal of University of Science and Technology of China, 2018, 48(1): 35-41.

参考文献

［1］
GUO C, YANG Y, PAN H, et al. Fault analysis of high speed train with DBN hierarchical ensembles[C]// Proceedings of the International Joint Conference on Neural Networks. Vancouver：IEEE Press, 2016: 2552-2559.
[2] HUANG N E, SHEN Z, LONG S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[C]// Proceedings of the Royal Society of London A: Mathematical, Physical and Engineering Sciences. London: The Royal Society, 1998, 454(1971): 903-995.
[3] 李智敏, 苟先太, 秦娜, 等. 高速列车振动监测信号的频率特征[J]. 仪表技术与传感器, 2015, (5): 99-103.

[4] ZHAO J, YANG Y, LI T, et al. Application of Empirical Mode Decomposition and Fuzzy Entropy to High-Speed Rail Fault Diagnosis[M]. Foundations of Intelligent Systems. Springer Berlin Heidelberg, 2014: 93-103.
[5] 刘林艳, 金炜东, 余志斌. 基于小波分析的高速列车车体运行状态估计[J]. 计算机应用研究, 2013, 30(10): 2948-2950.
[6] XU C, TAO D, XU C. A survey on multi-view learning[EB/OL]. [2017-05-10] https://arxiv.org/pdf/1304.5634v1.pdf.
[7] ROKACH L. A survey of clustering algorithms[J]. Data Mining and Knowledge Discovery Handbook, 2009, 16(3): 269-298.
[8] STREHL A, GHOSH J. Cluster ensembles: A knowledge reuse framework for combining partitionings[C]// Proceedings of the 8th National Conference on Artificial Intelligence. Edmonton, Canada: ACM Press, 2002: 93-99.
[9] GREENE D, CUNNINGHAM P. A matrix factorization approach for integrating multiple data views[C]// Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Springer, 2009: 423-438.
[10] 黄采伦, 樊晓平, 陈春阳, 等. 基于小波系数提取及离散余弦包络分析的机车牵引齿轮故障诊断方法[J]. 铁道学报, 2008, 30(2): 98-102.
[11] 张乾. 基于振动信号的轴承状态监测和故障诊断方法研究[D]. 长沙: 中南大学, 2012.
[12] SRA S, DHILLON I S. Nonnegative matrix approximation: Algorithms and applications[EB/OL]. [2017-04-28]http://www.cs.utexas.edu/ftp/techreports/tr06-27.pdf.
[13] 李乐, 章毓晋. 非负矩阵分解算法综述[J]. 电子学报, 2008, 36(4): 737-743.
[14] 杨燕, 靳蕃, KAMEL M. 聚类有效性评价综述[J]. 计算机应用研究, 2008, 25(6): 1630-1632.
[15] CAI X, NIE F, HUANG H. Multi-view k-means clustering on big data[C]// Proceedings of the 23th International Joint Conference on Artificial Intelligence. Beijing: IEEE Press2013: 2598-2604.
[16] XIA R, PAN Y, DU L, et al. Robust multi-view spectral clustering via low-rank and sparse decomposition[C]// Proceedings of the 28th AAAI Conference on Artificial Intelligence. Québec, Canada: ACM Press, 2014: 2149-2155.
[17] KUMAR A, RAI P, DAUMé H. Co-regularized multi-view spectral clustering[C]// Proceedings of the 24th International Conference on Neural Information Processing Systems. Granada, Spain: ACM Press, 2011: 1413-1421.

()
(

[1]	马玉莲, 崔文泉. 一种预测高频价格的端到端双目标多任务方法[J]. 中国科学技术大学学报, 2021, 51(3): 246-258.
[2]	崔文泉，陈伟，程浩洋. 非均衡数据情形的一种协同正则化多视图半监督学习分类器[J]. 中国科学技术大学学报, 2020, 50(5): 596-604.
[3]	张会敏，杨明，吕静. 基于自适应核联合稀疏表示的多特征高光谱图像分类[J]. 中国科学技术大学学报, 2018, 48(4): 298-306.

基于多视图加权聚类集成的高速列车工况识别

Condition recognition of high-speed train based on multi-view weighted clustering ensemble

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

本文评价