高速列车的样本关联改进故障诊断方法

摘要/Abstract

摘要： 聚合经验模态分解和基于变量预测模型的模式识别的结合是一种有效的机械故障诊断方法。针对该方法在高速列车故障诊断时存在小样本方法不适用和识别率较低等不足，首先采用滑窗逐步回归法对基于变量预测模型进行了适应性改进，再利用样本间的关联性和连续性，将相邻样本纳入模式识别，并进行样本平滑性处理，从而有效提高了故障诊断识别率。实验分析结果表明，改进方法降低了对样本量的需求，故障识别率提高了20%以上。

关键词: 高速列车, 变量预测模型, 故障诊断, 小样本, 关联分析

Abstract: Ensemble empirical mode decomposition(EEMD) and variable predictive model based class discriminate(VPMCD) combined method was a effective method for mechanical fault diagnosis. For some insufficients such as the method might not apply small samples and with low recognition rates in diagnosing a high speed train, so the adaptability of VPMCD was improved by sliding window stepwise regression herein. Then, using relevancy and continuity of samples, recognition rate of this method was improved by pattern recognition including adjacent samples and smooth processing of samples. Experimental results indicate that the corrected method may reduce needs for samples, and increase recognition rate 20% at least.

Key words: high speed train, variable predictive model, fault diagnosis, limited sample, association analysis

中图分类号:

TH113

张楷;罗怡澜;邹益胜;王超;宋小欣. 高速列车的样本关联改进故障诊断方法[J]. 中国机械工程.

ZHANG Kai;LUO Yilan;ZOU Yisheng;WANG Chao;SONG Xiaoxin. Sample Correlation Improvement Based High Speed Train Fault Diagnosis Method[J]. China Mechanical Engineering.

参考文献

[1]赵志宏. 基于振动信号的机械故障特征提取与诊断研究[D].北京：北京交通大学,2012.
ZHAO Zhihong. Research on Vibration Signal Based Machinery Fault Feature Extraction and Diagnosis[D]. Beijing: Beijing Jiaotong University,2012.
[2]王靖. 列车轮对故障振动特性及诊断关键技术研究[D]. 长沙：中南大学, 2012.
WANG Jing. Research of the Key Technologies for the Train’s Wheel Set Fault Vibriation Characteristics Diagnosis Method[D]. Changsha: Central South University, 2012.
[3]丁夏完,刘金朝,王成国.基于EMD和自适应STFT的货车滚动轴承故障诊断[J]. 中央民族大学学报(自然科学版), 2006, 26(3):253-258.
DING Xiawan, LIU Jinzhao,WANG Chengguo.The EMD and Adaptive STFT Based Fault Diagnosis Method of Freight Locomative Rolling Bearing [J]. Journal of Central University for Nationalities (Natural Science Edition), 2006, 26(3): 253-258.
[4]陈特放,黄采伦,樊晓平. 基于小波分析的机车走行部故障诊断方法[J]. 中国铁道科学, 2005, 26(4):89-92.
CHEN Tefang, HUANG Cailun, FAN Xiaoping. The Wavelet Analysis Based Fault Diagnosis Method of Locomotive Running Gear Fault [J]. China Railway Science, 2005, 26(4):89-92.
[5]马莉,傅八路,王泽勇，等.列车车轮踏面擦伤信号处理算法研究[J]. 铁道技术监督,2009,37(3):5-7.
MA Li, FU Balu,WANG Zeyong, et al. The Study of Train Wheel Tread Scratches Signal Processing Algorithms[J]. Railway Quality Control ,2009,37(3):5-7.
[6]秦娜. 高速列车转向架故障的信息熵测度特征分析方法研究[D]. 成都:西南交通大学, 2014.
QIN Na. Research on High-speed Train Bogie Fault Data Feature Analysis Method Based on Information Entropy Measuerment[D]. Chengdu: Southwest Jiaotong University, 2014.
[7]WU Z H, HUANG N E. Nsemble Empirical Mode Decomposition: a Noise Assisted Data Analysis Method[J]. Advances in Adaptive Data Analysis, 2009,1(1):1-41.
[8]RAGHURAJ R, LAKSHMINARAYANAN S. Variable Predictive Models-A New Multivariate Classification Approach for Pattern Recogintion Applications[J]. Patten Recogintion, 2009,42(1):7-16.
[9]RAGHURAJ R, LAKSHMINARAYANAN S. Variable Predictive Model Based Classification Algorithm for Effective Separation of Protein Structural Classes[J]. Computational Biology and Chemistry, 2008, 32(4): 302-306.
[10]罗颂荣, 程军圣, 杨宇，等. 基于本征时间尺度分解和变量预测模型模式识别的机械故障诊断[J]. 振动与冲击, 2013, 32(13):43-48.
LUO Songrong, CHENG Junsheng, YANG Yu, et al. Machine Fault Diagnosis Method Using ITD and Variable Predictive Model-based Class Discrimination[J]. Journal of Vibration and Shock, 2013,32(13)：43-48.
[11]杨宇,李杰,潘海洋,等. 改进VPMCD方法在滚动轴承故障诊断中的应用[J]. 中国机械工程,2014,25(4): 491-496.
YANG Yu, LI Jie, PAN Haiyang, et al. Application of Improved VPMCD Approach in Roller Bearing Fault Diagnosis[J]. China Mechanical Engineering, 2014, 25(4): 491-496.
[12]杨宇,李紫珠,何知义,等. 基于ASTFA降噪和AKVPMCD的滚动轴承故障诊断方法[J]. 中国机械工程, 2015, 24(21): 2934-2940.
YANG Yu, LI Zizhu, HE Zhiyi, et al. The Rolling Bearing Fault Diagnosis Method Based on ASTFA De-noising and AKVPMCD[J]. China Mechanical Engineering, 2015, 24(21): 2934-2940.
[13]潘海洋,杨宇,李永国,等. 基于流形学习和改进VPMCD的滚动轴承故障诊断方法[J]. 振动工程学报, 2014, 27(6): 934-941.
PAN Haiyang,YANG Yu, LI Yongguo, et al. A Rolling Bearing Fault Diagnosis Method Based on LE and KVPMCD[J]. Journal of Vibration Engineering, 2014, 27(6): 934-941.
[14]杨宇，潘海洋，程军圣.基于Kriging函数的KVPMCD在滚动轴承故障诊断中的应用[J]. 中国机械工程，2014,25（16）：2131-2136.
YANG Yu, PAN Haiyang, CHENG Junsheng. Applation of KVPMCD Based on Kriging Function in Rolling Bearing Fault Diagnosis[J]. China Mechanical Engineering, 2014, 25（16）：2131-2136.
[15]朱明,吴思东,付克昌,等. 基于熵特征的高速列车故障诊断方法[J]. 振动、测试与诊断, 2015, 35(2):381-387.
ZHU Ming, WU Sidong, FU Kechang. A High-speed Train Fault Diagnosis Method Based on Entropy Feature[J].Journal of Vibration, Measurement & Diagnosis, 2015, 35(2): 381-387.
[16]何晓群，刘文卿. 应用回归分析[M]. 北京：中国人民大学出版社，2011.
HE Xiaoqun, LIU Wenqing. Applied Regression Analysis[M]. Beijing: China Renmin University Press,2011.