结合VMD和Volterra预测模型的轴承振动信号特征提取

PDF(960 KB)

振动与冲击 ›› 2018, Vol. 37 ›› Issue (3) : 129-135.

论文

张云强1，张培林1，王怀光1，杨玉栋2，吴定海1

作者信息 +

Feature extraction method for rolling bearing vibration signals based on VMD and Volterra prediction model

ZHANG Yun-qiang1, ZHANG Pei-lin1, WANG Huai-guang1, YANG Yu-dong2, WU Ding-hai1

Author information +

文章历史 +

摘要

针对滚动轴承振动信号的非线性和非平稳性特点，提出了一种结合变分模式分解（VMD）和Volterra预测模型的轴承振动信号特征提取方法。利用VMD良好的非平稳信号分解能力将轴承振动信号分解成有限个平稳的本征模式函数（IMF）分量，然后对各IMF分量进行相空间重构，在重构的相空间内建立Volterra自适应预测模型，最后根据类内类间距准则对模型参数进行优选，用于描述轴承振动信号。对4种状态的滚动轴承振动信号进行了分析，优选的特征参数表现出较好的分类性能。实验结果表明，所提方法能有效提取振动信号中的非线性和非平稳特征，从而提高滚动轴承故障诊断精度。

Abstract

Aiming at nonlinear and non-stationary characteristics of rolling bearing vibration signals, a feature extraction method for rolling bearing vibration signals based on variational mode decomposition(VMD) and Volterra prediction model was proposed. VMD with a good ability of non-stationary signal decomposition was utilized to decompose a rolling bearing vibration signal into finite stationary intrinsic mode functions (IMFs). Then phase space reconstruction was conducted for these IMFs. Volterra prediction model was established in the reconstructed phase space. With the class distance within class criterion, the model’s parameters were optimized to describe the bearing vibration signal. Four different states of rolling bearing vibration signals were analyzed, the optimized feature parameters had a better classification performance. Test results indicated that the proposed method can effectively be used to extract nonlinear and non-stationary features of vibration signals, and improve the fault diagnosis accuracy of rolling bearings.

导出引用

张云强1，张培林1，王怀光1，杨玉栋2，吴定海1. 结合VMD和Volterra预测模型的轴承振动信号特征提取[J]. 振动与冲击, 2018, 37(3): 129-135

ZHANG Yun-qiang1, ZHANG Pei-lin1, WANG Huai-guang1, YANG Yu-dong2, WU Ding-hai1. Feature extraction method for rolling bearing vibration signals based on VMD and Volterra prediction model[J]. Journal of Vibration and Shock, 2018, 37(3): 129-135

参考文献

[1] 徐振辉, 马立元. 滚动轴承的故障特征提取[J]. 兵工自动化, 2004, 23(1): 46-48.
XU Zhen-hui, MA Li-yuan. Picking up fault character of rolling bearings[J]. Ordnance Industry Automation, 2004, 23(1): 46-48.
[2] 王宏超, 陈进, 董广明. 基于最小熵解卷积与稀疏分解的滚动轴承微弱故障特征提取 [J]. 机械工程学报, 2013, 49(1): 88-94.
WANG Hong-chao, CHEN Jin, DONG Guang-ming. Fault diagnosis method for rolling bearing's weak fault based on minimum entropy deconvolution and sparse decomposition[J]. Journal of Aechanical Engineering, 2013, 49(1): 88-94.
[3] 赵联春, 马家驹, 范树迁等. 滚动轴承振动分析中的AR模型研究[J]. 中国机械工程, 2004, 15(3): 210-213.
ZHAO Lian-chun, MA Jia-ju, FAN Shu-qian, et al. Research on AR model in vibration analysis of rolling bearing[J]. China Mechanical Engineering, 2004, 15(3): 210-213.
[4] 孙学斌. AR模型和SVM在机床滚动轴承故障诊断中的应用[J]. 机械工程与自动化, 2010, 44(2): 132- 134.
SUN Xue-bin. Application of AR model and SVM in machine rolling bearing fault diagnosis[J]. Mechanical Engineering & Automation, 2010, 44(2): 132-134.
[5] 刘颖, 严军. 基于时间序列ARMA模型的振动故障预测 [J]. 化工自动化及仪表, 2011, 38(7): 841-843.
LIU Ying, YAN Jun. Vibration faults prediction based on time series auto regressive moving average(ARMA) model [J]. Control and Instruments in Chinese Industry, 2011, 38(7): 841-843.
[6] 范庚, 马登武. 基于EMD和RVM-AR的航空发动机磨损故障预测模型[J]. 计算机测量与控制, 2013, 21(3): 1746-1749.
FAN Geng, MA Deng-wu. Aero-engine wear faults prediction based on EMD and RVM-AR[J]. Computer Measurement & Control, 2013, 21(3): 1746-1749.
[7] 王俨剀, 马进锐, 廖明夫等. 发动机振动趋势预测模型研究Research on trend prediction model of engine vibration[J]. 振动、测试与诊断Journal of Vibration, Measurement & Diagnosis, 2014, 34(3): 517-523.
WANG YAN-kai, MA Jin-rui, LIAO Ming-fu, et al. Research on trend prediction model of engine vibration[J]. Journal of Vibration, Measurement & Diagnosis, 2014, 34(3): 517-523.
[8] Huang N E, Shen Z, Long S R, et al. The empirical model decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Processing of Royal Society of London A, 1998, 454: 903-995.
[9] CHENG Jun-sheng, YU De-jie, YANG Yu. A fault diagnosis approach for roller bearings based on EMD method and AR model[J]. Mechanical System and Signal Processing, 2006, 20(2): 3650-362.
[10] 孟宗, 顾海燕. 应用经验模态分解下的AR模型提取旋转机械故障特征 [J]. 燕山大学学报, 2011, 35(4): 342-326.
MENG Zong, GU Hai-yan. Research on fault feature extraction of rotating machine based on empirical mode decomposition and AR model[J]. Journal of Yanshan University, 2011, 35(4): 342-326.
[11] 陈冬青, 许红波, 王新华. 基于ARMA模型关联维数与LSSVM的轴承损伤评定[J]. 起重运输机械, 2015, (5): 51-55.
CHEN Dong-qing, XU Hong-bo, WANG Xinhua. Bearing damage assessment based on the correlation dimension of ARMA model and LSSVM[J]. Hoisting and Conveying Machinery, 2015, (5): 51-55.
[12] Tang Hao, Liao Y H, Cao J Y，et al. Fault diagnosis approach based on Volterra model[J]. Mechanical System and Signal Processing, 2010,24: 1099-1113.
[13] 裘焱, 吴亚峰, 李野. 应用EMD分解下的Volterra模型提取机械故障特征 [J]. 振动与冲击, 2010, 29(6): 59-61.
QIU Yan, WU Ya-feng, LI Ye. Applying EMD decomposition of the Volterra model to extract mechanical fault feature[J]. Journal of Vibration and Shock, 2010, 29(6): 59-61.
[14] Konstantin D, Dominique Z. Variational Mode Decomposition[J]. IEEE Transactions on signal processing, 2014, 62(3): 531-544.
[15] Salim L. Comparative study of signal denoising by wavelet threshold in empirical and variational mode decomposition domains[J]. Healthcare Technology Letters, 2014,1 (3): 104-109.
[16] WANG Yan-xue, Richard M, XIANG Jia-wei, et al. Research on variational mode decomposition and its application in detecting rub-impact fault of the rotor system[J]. Mechanical Systems and Signal Processing, 2015, (60-61): 243-251.
[17] Aneesh C, Sachin K, Hisham P M, et al. Performance comparison of variational mode decomposition over empirical wavelet transform for the classification of power quality disturbances using support vector machine[J]. Procedia Computer Science, 2015, (46): 372 -380.
[18] 韩敏. 混沌时间序列预测理论与方法[M]. 北京:中国水利水电出版社, 2007.
[19] Chow T W S, Tan H Z. HOS-based nonparametric and parametric methodologies for machine fault detection[J]. IEEE Transactions on Industrial Electronics, 2000, 47(5): 1051-1059.
[20] Peng H C, Long F H, Chris D. Feature selection based on mutual information: criteria of max- dependency, max-relevance, and min-redundancy[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(8): 1226-1238.