基于变分模态分解与流形学习的滚动轴承故障特征提取方法

PDF(1470 KB)

振动与冲击 ›› 2018, Vol. 37 ›› Issue (23) : 133-140.

论文

戚晓利，叶绪丹，蔡江林,郑近德，潘紫微，张兴权

作者信息 +

Fault feature extraction method of rolling bearings based on VMD and manifold learning

QI Xiaoli, YE Xudan, CAI Jianglin, ZHENG Jinde, PAN Ziwei,ZHANG Xingquan

Author information +

文章历史 +

摘要

提出了一种基于变分模态分解（VMD）和局部切空间排列算法（LTSA）相结合的滚动轴承早期故障诊断方法。首先利用VMD算法分解圆柱滚子轴承不同运行状态下的振动信号，通过求取瞬时频率均值并绘制特征曲线筛选出与原始信号最为相关的几个分量，然后，提取有效模态分量的时域指标和小波包频带分解能量所构成的频域指标，两者结合初步提取高维故障特征后，再应用LTSA对故障特征进行二次提取，最后输入到K-means分类器进行故障类型识别。通过对圆柱滚子轴承故障诊断的对比实验分析，发现：（1）与时频特征+LTSA、EMD+LTSA 特征提取方法相比，VMD+LTSA方法在分类效果和识别精度上更具优势；（2）LTSA算法相比较于PCA、LPP、LE、ISOMAP和LLE这5种算法，其降维后的特征故障敏感性最好。研究结果表明本文所提出的方法在圆柱滚子轴承故障诊断方面具有一定的优越性。

Abstract

A method for early fault diagnosis of rolling bearings based on the variational mode decomposition (VMD) combined with the local tangent space alignment (LTSA) algorithm was proposed.Firstly,the VMD algorithm was used to decompose vibration signals of a rolling bearing under different operational conditions,and several components most correlated to the original signal were screened out by solving the instantaneous frequency mean and plotting the feature curve.Then,the time domain index of the effective modal component and the frequency domain index formed with the wavelet packet frequency band decomposition energy were extracted.After these two indexes were combined to primarily extract higher dimensional fault features,LTSA was used to extract fault features again.Finally,the extracted fault features were inputted into a K-means classifier to recognize fault types.Through the contrastive test analysis of cylindrical roller bearings’fault diagnosis,it was shown that compared with the method of time frequency features extraction+LTSA and the method of EMD+LTSA,the method of VMD+LTSA has more advantages in classification effect and recognition accuracy; the LTSA algorithm has the best sensitivity to fault features after dimension reduction compared to the algorithms of PCA,LPP,LE,ISOMAP and LLE; the proposed method has a certain superiority in the fault diagnosis of cylindrical roller bearings.

导出引用

戚晓利，叶绪丹，蔡江林,郑近德，潘紫微，张兴权. 基于变分模态分解与流形学习的滚动轴承故障特征提取方法[J]. 振动与冲击, 2018, 37(23): 133-140

QI Xiaoli, YE Xudan, CAI Jianglin, ZHENG Jinde, PAN Ziwei,ZHANG Xingquan. Fault feature extraction method of rolling bearings based on VMD and manifold learning[J]. Journal of Vibration and Shock, 2018, 37(23): 133-140

参考文献

[1] 郑近德, 程军圣, 杨宇. 基于LCD和排列熵的滚动轴承故障诊断[J]. 振动、测试与诊断, 2014, 34(5):802-806.
Zheng Jin-de, Cheng Jun-heng, Yang Yu A rolling bearing fault diagnosis method based on LCD and
permutation entropy[J]. Journal of Vibration and Shock,
2005, 24 (1): 85-88.
[2] Zhu K, Song X, Xue D. Incipient fault diagnosis of roller bearings using empirical mode decomposition and correlation coefficient[J]. Journal of Vibroengineering, 2013, 15(2):597-603.
[3] 张淑清, 翟欣沛, 董璇, 等. EMD及Duffing振子在小电流系统故障选线方法中的应用[J]. 中国电机工程学报, 2013, 33(10):161-167.
Zhang Su qing, Zhai Xin Pei, Dong Xuan, et al. Application of EMD and Duffing oscillator to fault line detection in un-effectively grounded system[J]. Proceedings of the CSEE, 2013, 33(10):161-167.
[4] 陈亚农, 郜普刚, 何田, 等. 局部均值分解在滚动轴承故障综合诊断中的应用[J]. 振动与冲击, 2012, 31(3):73-78.
Chen Ya-nong, Gao Pu-gang, He Tian, et al. Roller bearing comprehensive fault diagnosis based on LMD[J].Journal of Vibration and Shock, 2012, 31(3):73-78.
[5] Wang Y, He Z, Zi Y. A Comparative Study on the Local Mean Decomposition and Empirical Mode Decomposition and Their Applications to Rotating Machinery Health Diagnosis[J]. Journal of Vibration & Acoustics, 2010, 132(2):613-624.
[6] 程军圣, 张亢, 杨宇,等. 局部均值分解与经验模式分解的对比研究[J]. 振动与冲击, 2009, 28(5):13-16.
CHENG Jun-Sheng, ZHANG Kang, YANG Yu. Comparison of local mean decomposition and empirical mode decomposition [J].Journal of Vibration and Shock, 2009, 28(5):13-16.
[7] 刘长良, 武英杰, 甄成刚. 基于变分模态分解和模糊C均值聚类的滚动轴承故障诊断[J]. 中国电机工程学报, 2015, 35(13):3358-3365.
Liu Changliang, Wu Yingjie, Zhen Chenggang. Rolling bearing fault diagnosis based on variational mode decomposition and fuzzy c means clustering [J]. Proceedings of the CSEE, 2015, 35(13):3358-3365.
[8] Dragomiretskiy K, Zosso D. Variational Mode Decomposition[J]. IEEE Transactions on Signal Processing, 2014, 62(3):531-544.
[9] An X, Tang Y. Application of variational mode decomposition energy distribution to bearing fault diagnosis in a wind turbine[J]. Transactions of the Institute of Measurement & Control, 2016, 5(2):753-772.
[10] Wang Y, Markert R, Xiang J, et al. Research on variational mode decomposition and its application in detecting rub-impact fault of the rotor system[J]. Mechanical Systems & Signal Processing, 2015, 60–61:243-251.
[11] 马增强, 李亚超, 刘政, 等. 基于变分模态分解和 Teager 能量算子的滚动轴承故障特征提取[J]. 振动与冲击, 2016, 35(13)：134-139
MA Zeng-qiang, LI Ya-chao, LIU Zheng, et al.Rolling Bearings’ Fault Feature extration based on variational mode decomposition and Teager energy operator [J]. Journal of Vibration and Shock, 2016, 35 (13)：134-139
[12] 万鹏, 王红军, 徐小力. 局部切空间排列和支持向量机的故障诊断模型[J]. 仪器仪表学报, 2012, 33(12):2789-2795.
WAN Peng, WANG Hong-jun, XU Xiao-li. Fault diagnosis model based on local tangent space alignment and support vector machine [J]. Chinese Journal of Scientific Instrument, 2012, 33 (12): 2789-2795.
[13] 王斐, 房立清, 吕岩. 局部特征尺度分解和局部切空间排列在故障特征频率提取中的应用[J]. 中国机械工程, 28(05): 565-575.
WANG Fei, FANG Liqing, Lv Yan. Fault Frequency Extracting Methods Based on LCD and LTSA[J]. ChinaMechanical Engineering, 28(05): 565-575.
[14] 向丹, 葛爽. 基于EMD样本熵-LLTSA的故障特征提取方法[J]. 航空动力学报, 2014, 29(7):1535-1542.
Xiang Dan, Ge Shuang. Method of fault feature extraction based on EMD sample entropy and LLTSA [J]. Journal of Aerospace Power, 2014, 29(7):1535-1542.
[15] 苏祖强, 汤宝平, 姚金宝. 基于敏感特征选择与流形学习维数约简的故障诊断[J]. 振动与冲击, 2014, 33(3):70-75.
Su Zu-qiang, Tang Bao-ping, Yao Jin-bao. Fault diagnosis method based on sensitive feature selection and manifold learning dimension reduction [J]. Journal of Vibration and Shock, 2014, 33(3):70-75.