Rolling Bearing Fault Diagnosis based on CEEMD-WVD Multi-scale Time-frequency Image
-
摘要: 针对一般EMD-WVD方法在提取时频图像信息不充分的问题,提出一种基于CEEMD-WVD多尺度时频图像的滚动轴承故障诊断方法。该方法采用互补集合经验模态分解(CEEMD)方法对故障振动信号进行分解,自适应地获得不同频段的固有模态函数(IMF)分量;选取前几个高频信号IMF模态分量,运用Wigner-Ville分布(WVD)对各IMF分量分别做时频分析,进一步转化成对应的多尺度的时频图像;然后提取各尺度时频图像的局部二进制(LBP)纹理特征,并利用其特征训练SVM分类器;最后用训练好的分类器对不同的轴承故障振动信号进行故障识别。实验结果表明,该方法有较强的自适应性且能生成高分辨率图像,故障识别率高,在凯斯西储大学(CWRU)的滚动轴承数据库上进行5类故障的实验,诊断正确率为99.75%。Abstract: Aiming at the problem that the general EMD-WVD method has insufficient information for extracting time-frequency image, a new fault diagnosis method for rolling bearing based on CEEMD-WVD multi-scale time-frequency image is proposed. The complementary ensemble empirical mode decomposition (CEEMD) method was adopted in order to decompose the fault vibration signal and adaptively obtained the intrinsic mode function (IMF) components of different frequency bands. By selecting IMF modal components of the first few high frequency signals, time-frequency analysis was performed on each IMF component using Wigner-Ville distribution (WVD), and further converted it into a corresponding multi-scale time-frequency image. Then the local binary (LBP) texture features of each time-frequency image were extracted, and the SVM classifier was trained with these features. Finally, the trained classifier was used to recognize faults of different bearing vibration signals. The experimental results show that the method has strong adaptability and can generate high-resolution images, and the fault recognition rate is high. The five types of faults are tested on the rolling bearing database of Case Western Reserve University (CWRU), and the diagnostic accuracy rate is up to 99.75%.
-
Key words:
- CEEMD-WVD /
- multi-scale /
- adaptive /
- time-frequency image /
- fault diagnosis /
- rolling bearing /
- IMF modal component /
- SVM classifier
-
表 1 故障类别及其描述
类别 名称 样本数 描述 1 Ball_DE 480 (0.007 in和0.014 in)
驱动端滚动体故障2 InnerRace_DE 480 (0.007 in和0.014 in)
驱动端内圈故障3 OuterRace_DE_@3 480 (0.007 in和0.014 in)
驱动端外圈3点位置故障4 OuterRace_DE_@6 480 (0.007 in和0.014 in)
驱动端外圈6点位置故障5 OuterRace_DE_@12 480 (0.007 in和0.014 in)驱动端外圈12点位置故障 表 2 各对比实验结果
实验序号 图像生成方式 特征提取方法 准确率/% 1 WVD HOG 92.67 2 WVD LBP 97.91 3 WT+WVD LBP 98.17 4 CEEMD+WVD(N) LBP 99.25 5 CEEMD+WVD LBP 99.75 表 3 实验名称及方法
实验号 实验方法 1 WVD+LBP 2 WT+WVD+LBP 3 CEEMD+WVD (N)+LBP 4 CEEMD+WVD+LBP -
[1] Antoni J, Randall R B. The spectral kurtosis:application to the vibratory surveillance and diagnostics of rotating machines[J]. Mechanical Systems and Signal Processing, 2006, 20(2):308-331 http://cn.bing.com/academic/profile?id=ba5efec12b698e2d6c171ffc46b5a63a&encoded=0&v=paper_preview&mkt=zh-cn [2] Bartelmus W, Zimroz R. Vibration condition monitoring of planetary gearbox under varying external load[J]. Mechanical Systems and Signal Processing, 2009, 23(1):246-257 http://cn.bing.com/academic/profile?id=1e66172e527f58dfd8475177ac5ba636&encoded=0&v=paper_preview&mkt=zh-cn [3] Lei Y G, Lin J, He Z J, et al. Application of an improved kurtogram method for fault diagnosis of rolling element bearings[J]. Mechanical Systems and Signal Processing, 2011, 25(5):1738-1749 http://d.old.wanfangdata.com.cn/Periodical/zdycj201723028 [4] Wang G F, Luo Z G, Qin X D, et al. Fault identification and classification of rolling element bearing based on time-varying autoregressive spectrum[J]. Mechanical Systems and Signal Processing, 2008, 22(4):934-947 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=cd570e6b6a35f34daab6afbf8c979be3 [5] 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[J]. Proceedings:Mathematical, Physical and Engineering Sciences, 1998, 454(1971):903-995 http://cn.bing.com/academic/profile?id=46b548aa17d217d7dd38acbf82205b2d&encoded=0&v=paper_preview&mkt=zh-cn [6] Yang W X, Tavner P J. Empirical mode decomposition, an adaptive approach for interpreting shaft vibratory signals of large rotating machinery[J]. Journal of Sound and Vibration, 2009, 321(3-5):1144-1170 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=e02bc12c029f5b23f3dced2c104d680a [7] Shiau Y H, Wu M C. Detecting characteristics of information masked by a laser-triggered microwave system via Hilbert-Huang transform[J]. Optics Communications, 2010, 283(9):1909-1916 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=57bcb8475fbd7dc0bbf790dd195756dc [8] Wu Z H, Huang N E. Ensemble empirical mode decomposi-tion:a noise-assisted data analysis method[J]. Advances in Adaptive Data Analysis, 2009, 1(1):1-41 http://d.old.wanfangdata.com.cn/Periodical/dianzixb201305033 [9] Yeh J R, Shieh J S, Huang N E. Complementary ensemble empirical mode decomposition:a novel noise enhanced data analysis method[J]. Advances in Adaptive Data Analysis, 2010, 2(2):135-156 http://d.old.wanfangdata.com.cn/NSTLQK/NSTL_QKJJ0219457365/ [10] 蔡艳平, 李艾华, 石林锁, 等.基于EMD-WVD振动谱时频图像SVM识别的内燃机故障诊断[J].内燃机工程, 2012, 33(2):72-78, 85 http://d.old.wanfangdata.com.cn/Periodical/nrjgc201202012Cai Y P, Li A H, Shi L S, et al. IC engine fault diagnosis method based on EMD-WVD vibration spectrum time-frequency image recognition by SVM[J]. Chinese Internal Combustion Engine Engineering, 2012, 33(2):72-78, 85(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/nrjgc201202012 [11] 蔡艳平, 李艾华, 王涛, 等.基于EMD-Wigner-Ville的内燃机振动时频分析[J].振动工程学报, 2010, 23(4):430-437 http://d.old.wanfangdata.com.cn/Periodical/zdgcxb201004012Cai Y P, Li A H, Wang T, et al. I.C. engine vibration time-frequency analysis based on EMD-Wigner-Ville[J]. Journal of Vibration Engineering, 2010, 23(4):430-437(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/zdgcxb201004012 [12] 牟伟杰, 石林锁, 蔡艳平, 等.基于EMD-WVD与LNMF的内燃机故障诊断[J].振动与冲击, 2016, 35(23):191-196, 202 http://d.old.wanfangdata.com.cn/Periodical/zdycj201623030Mu W J, Shi L S, Cai Y P, et al. IC engine fault diagnosis method based EMD-WVD and LNMF[J]. Journal of Vibration and Shock, 2016, 35(23):191-196, 202(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/zdycj201623030 [13] 董利超, 郭兴明, 郑伊能.基于CEEMD的心音信号小波包去噪算法研究[J].振动与冲击, 2019, 38(9):192-198, 222 http://d.old.wanfangdata.com.cn/Periodical/zdycj201909026Dong L C, Guo X M, Zheng Y N. Wavelet packet de-noising algorithm for heart sound signals based on CEEMD[J]. Journal of Vibration and Shock, 2019, 38(9):192-198, 222(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/zdycj201909026 [14] 王亚萍, 匡宇麒, 葛江华, 等.CEEMD和小波半软阈值相结合的滚动轴承降噪[J].振动、测试与诊断, 2018, 38(1):80-86 http://d.old.wanfangdata.com.cn/Periodical/zdcsyzd201801012Wang Y P, Kuang Y Q, Ge J H, et al. De-noising method for bearing vibration signal based on CEEMD and wavelet semi-soft threshold[J]. Journal of Vibration, Measurement & Diagnosis, 2018, 38(1):80-86(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/zdcsyzd201801012 [15] 胡剑超, 练继建, 马斌, 等.基于CEEMD和小波包阈值的组合降噪及泄流结构的模态识别方法[J].振动与冲击, 2017, 36(17):1-9 http://d.old.wanfangdata.com.cn/Periodical/zdycj201717001Hu J C, Lian J J, Ma B, et al. A de-noising and modal identification combined method based on CEEMD and wavelet packet threshold for flood discharge structures[J]. Journal of Vibration and Shock, 2017, 36(17):1-9(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/zdycj201717001 [16] Liu J T, Zhou K F, Hu Y. EMD-WVD method based high-frequency current analysis of low voltage arc[C]//Proceedings of the 2018 Condition Monitoring and Diagnosis. Perth: IEEE, 2018 [17] Chai Y Z, Zhang X D. EMD-WVD time-frequency distribution for analysis of multi-component signals[C]//Proceedings of the Fourth International Conference on Wireless and Optical Communications. Beijing: SPIE, 2016 [18] Ojala T, Pietikäinen M, Harwood I. A comparative study of texture measures with classification based on featured distributions[J]. Pattern Recognition, 1996, 29(1):51-59 http://cn.bing.com/academic/profile?id=af5c928d762e12c9ee1905236a07c386&encoded=0&v=paper_preview&mkt=zh-cn [19] Bearing data center, case western reserve university[EB/OL]. 2013. http://csegroups.case.edu/bearingdatacenter/home.