基于Kurtogram与DSCN的滚动轴承故障诊断方法

doi:10.16265/j.cnki.issn 1003-3033.2021.06.013

摘要/Abstract

摘要： 为揭示不同轴承故障类型的特征,提高故障诊断的精度与效率,提出一种基于Kurtogram与深可分卷积神经网络(DSCN)相结合的轴承故障诊断方法。在利用原始振动信号生成Kurtogram的基础上,通过DSCN学习和识别不同故障模式下Kurtogram的图形特征,自动提取优势特征并进行故障分类。结果表明:相对于其他故障诊断方法,提出的方法在测试集上的识别精确度较高,可达到97.28%;同时,DSCN在降低参数量及提高训练速度上具有明显优势。

关键词: 滚动轴承, Kurtogram, 深可分卷积神经网络(DSCN), 故障诊断, 混淆矩阵

Abstract: In order to reveal fault characteristics of different bearing and improve accuracy and efficiency of fault diagnosis, a bearing fault diagnosis method based on Kurtogram and DSCN was proposed. After Kurtogram was generated with original vibration signals, its graphic features under different fault modes were studied and recognized by DSCN, and advantageous features were automatically extracted for fault classification. The results show that compared with other fault diagnosis methods, the proposed one has the highest recognition accuracy on test set, reaching as far as 97.28%, which also reflects obvious advantages of DSCN in reducing number of parameters and increasing training speed.

Key words: rolling bearing, Kurtogram, depthwise separable convolutional network (DSCN), fault diagnosis, confusion matrix

中图分类号:

古莹奎, 刘平, 林忠海, 邱光琦. 基于Kurtogram与DSCN的滚动轴承故障诊断方法[J]. 中国安全科学学报, 2021, 31(6): 99-105.

GU Yingkui, LIU Ping, LIN Zhonghai, QIU Guangqi. Fault diagnosis method of rolling bearing based on Kurtogram and DSCN[J]. China Safety Science Journal, 2021, 31(6): 99-105.

参考文献

[1] 陈祥龙, 张兵志, 冯辅周, 等. 基于改进排列熵的滚动轴承故障特征提取[J]. 振动工程学报, 2018,31(5): 902-908.
CHEN Xianglong, ZHANG Bingzhi, FENG Fuzhou, et al. Fault feature extraction of rolling bearings based on an improved permutation entropy [J]. Journal of Vobration Engineering, 2018,31(5): 902-908.
[2] MOHAMMED O D, RANTATALO M, AIDANPAA J O. Dynamic modelling of a one-stage spur gear system and vibration-based tooth crack detection analysis [J]. Mechanical Systems and Signal Processing, 2015,54-55(3): 293-305.
[3] GU Yingkui, ZHOU Xiaoqing, YU Dongping, et al. Fault diagnosis method of rolling bearing using principal component analysis and support vector machine [J]. Journal of Mechanical Science and Technology, 2018,32(11): 5 079-5 088.
[4] 孟宗, 殷娜, 李晶. 基于信号稀疏表示和瞬态冲击信号多特征提取的滚动轴承故障诊断[J]. 计量学报, 2019,40(5): 855-8 601.
MENG Zong, YIN Na, LI Jing. Fault diagnosis of rolling bearing based on sparse representation of signals and transient impulse signal multifeature extraction [J]. Acta Metrology Sinica, 2019,40(5): 855-8 601.
[5] 隋文涛, 张丹, WANG Wilson. 基于EMD和MKD的滚动轴承故障诊断方法[J]. 振动与冲击, 2015,34(9): 55-59,64.
SUI Wentao, ZHANG Dan, WANG Wilson. Fault diagnosis of rolling element bearings based on EMD and MKD [J]. Journal of Vibration and Shock, 2015,34(9):55-59,64.
[6] 胡爱军, 赵军. 基于自适应最大相关峭度解卷积的滚动轴承多故障诊断[J]. 振动与冲击, 2019,38(22): 171-177.
HU Aijun, ZHAO Jun. Diagnosis of multiple faults in rolling bearings based on adaptive maximum correlated kurtosis deconvolution [J]. Journal of Vibration and Shock, 2019,38(22): 171-177.
[7] ANTONI J. Fast computation of the kurtogram for the detection of transient faults [J]. Mechanical Systems and Signal Processing, 2007,21(1): 108-124.
[8] ZHANG Yu, FAN Zhuoyou, GAO Xiaorong, et al. A fault diagnosis method of train wheelset rolling bearing combined with improved LMD and FK [J]. Journal of Sensors, 2019,18: 1-11.
[9] HU Yue, BAO Wenjie, TU Xiaotong, et al. An adaptive spectral kurtosis method and its application to fault detection of rolling element bearings [J]. IEEE Transactions on Instrumentation and Measurement, 2020,69(3): 739-750.
[10] WANG Lei, LIU Zhiwen, CAO Hongrui, et al. Subband averaging kurtogram with dual-tree complex wavelet packet transform for rotating machinery fault diagnosis [J]. Mechanical Systems and Signal Processing, 2020,142:106755.
[11] 孙文卿, 邓艾东, 邓敏强, 等. 基于随机森林和自编码的滚动轴承多视角特征融合[J]. 东南大学学报:英文版, 2019,35(3): 302-309.
SUN Wenqing, DENG Aidong, DENG Minqiang, et al. Multi-view feature fusion for rolling bearing fault diagnosis using random forest and autoencoder [J]. Journal of Southeast University: English Edition, 2019,35(3): 302-309.
[12] 古莹奎, 承姿辛, 朱繁泷. 基于主成分分析和支持向量机的滚动轴承故障特征融合分析[J]. 中国机械工程, 2015,26(20): 2 778-2 783.
GU Yingkui, CHENG Zixin, ZHU Fanlong. Rolling bearing fault feature fusion based on PCA and SVM [J]. China Mechanical Engineering, 2015,26(20): 2 778-2 783.
[13] ZHAO Zhuanzhe, XU Qingsong, JIA Minping. Improved shuffled frog leaping algorithm-based BP neural network and its application in bearing fault diagnosis [J]. Neural Computing and Applications, 2016,27(2):375-385.
[14] TANG Jiahui, WU Jimei, HU Bingbing, et al. A fault diagnosis method using Interval coded deep belief network [J]. Journal of Mechanical Science and Technology, 2020,34(5): 1 949-1 956.
[15] ZHU Haiping, CHENG Jiaxin, ZHANG Cong, et al. Stacked pruning sparse denoising autoencoder based intelligent fault diagnosis of rolling bearings [J]. Applied Soft Computing, 2020,88:106060.
[16] 佟瑞鹏, 崔鹏程. 基于深度学习的不安全因素识别和交互分析[J]. 中国安全科学学报, 2017,27(4): 49-54.
TONG Ruipeng, CUI Pengcheng. Unsafe factor recognition and interactive analysis based on deep learning [J].China Safety Science Journal, 2017,27(4): 49-54.
[17] LIU Han, ZHOU Jianzhong, XU Yanhe, et al. Unsupervised fault diagnosis of rolling bearings using a deep neural network based on generative adversarial networks [J]. Neurocomputing, 2018,315(13): 412-424.
[18] 唐波, 陈慎慎. 基于深度卷积神经网络的轴承故障诊断方法[J]. 电子测量与仪器学报, 2020,34(3): 88-93.
TANG Bo, CHEN Shenshen. Method of bearing fault diagnosis based on deep convolutional neural network [J]. Journal of Electronic Measurement and Instrumentation, 2020,34(3): 88-93.
[19] GU Yingkui, ZENG Lei, QIU Guangqi. Bearing fault diagnosis with varying conditions using angular domain resampling technology, SDP and DCNN [J]. Measurement, 2020,156:107616.
[20] 徐丹, 代勇, 纪军红. 基于卷积神经网络的驾驶人行为识别方法研究[J]. 中国安全科学学报, 2019, 29(10): 12-17.
XU Dan, DAI Yong, JI Junhong. Research on driver behavior recognition method based on convolutional neural network[J].China Safety Science Journal, 2019,29(10): 12-17.
[21] ANTONI J. The spectral kurtosis: a useful tool for characterising non-stationary signals [J]. Mechanical Systems and Signal Processing, 2006, 20(2): 282-307.