基于多域信息融合的深度学习轴承故障诊断方法

PDF(4531 KB)

振动与冲击 ›› 2024, Vol. 43 ›› Issue (23) : 47-55.

论文

基于多域信息融合的深度学习轴承故障诊断方法

葛卓1,2，夏华猛1，王凯亮1，徐增丙3，丁改革3

作者信息 +

Bearing fault diagnosis method based on deep learning of multi-domain information fusion

GE Zhuo1,2, XIA Huameng1, WANG Kailiang1, XU Zengbing3, DING Gaige3

Author information +

文章历史 +

摘要

针对单一振动信号包含故障信息易被隐藏以及单一深度学习模型诊断能力不强导致轴承故障诊断精度低的问题，提出了一种基于多域信息融合的深度学习故障诊断方法。利用变分模态分解方法（variational mode decomposition，VMD）将原始振动信号分解为多个IMF分量，同时对每个IMF分量进行快速傅里叶变换（fast Fourier transformation，FFT）转化为频域样本；然后将多个IMF分量和其对应频域样本分别输入至多个深度度量学习(deep metric learning，DML)模型和深度置信网络(deep belief network，DBN)模型分别进行初步诊断分析，并利用简单软投票法对这些初步诊断结果进行融合从而获取最终诊断结果。最后通过对不同轴承故障的诊断试验分析，结果表明本文提出的方法不仅具有较好的诊断效果，而且诊断性能分别优于基于时域和基于频域的信息融合诊断方法。

Abstract

Aiming at the problem of a single vibration signal containing fault information being easily hidden and the weak diagnostic ability of a single deep learning model leading to low accuracy in bearing fault diagnosis, a deep learning fault diagnosis method based on multi-domain information fusion is proposed in this paper. Variational Mode Decomposition (VMD) method is adopted to decompose the original vibration signal into multiple IMF components, while fast Fourier transformation FFT transforming each IMF component into frequency domain samples. After that, multiple IMF components and their corresponding frequency domain samples are inputted into multiple deep metric learning (DML) models and deep belief network DBN models for preliminary diagnostic analysis, respectively. And then a simple soft voting method is used to fuse these preliminary diagnostic results to obtain the final diagnostic result. Finally, through the analysis of bearing fault diagnosis experiments, the results show that the proposed method not only has good diagnostic performance, but also outperforms information fusion diagnosis methods based on time domain and frequency domain, respectively.

导出引用

葛卓1, 2, 夏华猛1, 王凯亮1, 徐增丙3, 丁改革3. 基于多域信息融合的深度学习轴承故障诊断方法[J]. 振动与冲击, 2024, 43(23): 47-55

GE Zhuo1, 2, XIA Huameng1, WANG Kailiang1, XU Zengbing3, DING Gaige3. Bearing fault diagnosis method based on deep learning of multi-domain information fusion[J]. Journal of Vibration and Shock, 2024, 43(23): 47-55

参考文献

[1] LI Y F，LIANG X H，ZUO M J．Diagonal slice spectrum assisted optimal scale morphological filter for rolling element bearing fault diagnosis[J]. Mechanical Systems and Signal Processing, 2017, 85: 146－161．
[2] 余永华,杨建国,胡闹. 智能机舱关键部件状态监测诊断技术研究[J].船舶,2018,29(z1):98-105.
YU Y H, YANG J G, Nonsense. Research on condition monitoring and diagnosis technology of Key Components of intelligent engine room [J]. Ships,2018,29(z1):98-105.
[3] 吴春志，江鹏程，冯辅周，等．基于一维卷积神经网络的齿轮箱故障诊断[J]．振动与冲击，2018，37(22):56－61．
WU C Z, JIANG P C, FENG F Z, et al. Gear box Fault Diagnosis based on one-dimensional Convolutional Neural Network [J]. Journal of Vibration and Shock, 2018,37 (22):56-61
[4] 李恒，张氢，秦仙蓉，等．基于短时傅里叶变换和卷积神经网络的轴承故障诊断方法[J]．振动与冲击，2018，37(19):124－131.
LI H, ZHANG H, QIN X R, et al. Bearing fault diagnosis Method based on short-time Fourier Transform and Convolutional Neural Network [J]. Journal of Vibration and Shock, 2018,37 (19):124-131.
[5] 马伦，康建设，孟妍，等．基于Morlet小波变换的滚动轴承早期故障特征提取研究[J]．仪器仪表学报2013，34(4):920－926．
MA L, KANG J S, MENG Y, et al. Research on early fault Feature Extraction of rolling bearing based on Morlet Wavelet transform [J]. Chinese Journal of Scientific Instrument, 2013,34 (4):920-926.
[6] 李伟,邹本健,杨宁等.基于深度置信网络铝合金加筋板冲击损伤识别[J].振动.测试与诊断,2023, 43(1):88-95+199.
LI W, ZOU B J, YANG N et al. Impact Damage Identification of Aluminum Alloy stiffeners based on Deep confidence Network [J]. Vibration. Test and Diagnosis,2023, 43(1):88-95+199.
[7] 赵志宏,李春秀,窦广鉴等.基于MTF-CNN的轴承故障诊断研究[J].振动与冲击, 2023, 42(02):126-131.
ZHAO Z H, LI C X, DOU G J et al. Re search on Bearing Fault Diagnosis based on MTF-CNN [J]. Journal of Vibration and Shock, 2023, 42 (02):126-131.
[8] 李小娟,徐增丙,熊文等.基于深度度量学习的轴承故障诊断方法[J].振动与冲击,2020,39(15):25-31.
LI X J, XU Z B, XIONG W et al. Bearing Fault Diagnosis Method Based on Deep Metric Learning [J]. Journal of Vibration and Shock,20,39(15):25-31.
[9] 韩特,刘超,沈长青，等.深度嵌入度量学习的机械跨工况故障识别方法[J].振动工程学报,2023,36(2): 565-573.
HAN Te, LIU Chao, SHEN Changqing, et al. Deep embedding metric learning for machinery fault identification across different working conditions[J]. Journal of Vibration Engineering,2023,36(2):565-573.
[10] 贾峰,李世豪,沈建军，等.采用深度迁移学习与自适应加权的滚动轴承故障诊断[J].西安交通大学学报,2022,56(8):1-10.
JIA Feng，LI Shihao，SHEN Jianjun，et al. Fault diagnosis of rolling bearings using deep transfer learning and adaptive weighting[J]. Journal of Xi'an Jiaotong University,2022,56(8):1-10.
[11] ZHAO M H，KANG M，TANG B P，et al．Deep residual networks with dynamically weighted wavelet coefficients for fault diagnosis of planetary gearboxes［J］．IEEE Transactions on Industrial Electronics，2018，65(5) : 4290－4300．
[12] MA S J, LIU W K, CAI W, et al．Lightweight deep residual cnn for fault diagnosis of rotating machinery based on depthwise separable convolutions［J］. IEEE Access，2019，7:57023－57036．
[13] APPANA D K，ALEXANDER P，JONG-MYON K. Reliable fault diagnosis of bearings with varying rotational speeds using envelope spectrum and convolution neural networks[J]. Soft Computing，2018, 22: 6719－6729．
[14] SADOUGHI M，HU C．Physics-based convolutional neural network for fault diagnosis of rolling element bearings [J]. IEEE Sensors Journal, 2019, 19( 11) :4181－4192．
[15] WANG J. J, MA Y. L, HUANG Z. G, et al．Performance analysis and enhancement of deep convolutional neural network: application to gearbox condition monitoring ［J］．Business and Information Systems Engineering， 2019, 61(3):311－326．
[16] The Case Western Reserve University Bearing Data Center．Bearing Data Center Fault Test Data (1998-10-04). http://csegroups.case.edu /bearingdatacenter/pages/ download- data -file．
[17] DAGA A P, FASANA A , MARCHESIELLO S , et al. The Politecnico di Torino rolling bearing test rig: Description and analysis of open access data[J]. Mechanical Systems and Signal Processing, Volume 120,2019,Pages 252-273,ISSN 0888-3270, https://doi.org/10.1016/j.ymssp.2018.10.010.