一种机械设备故障诊断的FD-Transformer方法

PDF(2094 KB)

振动与冲击 ›› 2023, Vol. 42 ›› Issue (8) : 326-333.

论文

赵志宏1，2，李春秀2，李乐豪2，杨绍普1

作者信息 +

A FD-Transformer method for fault diagnosis of mechanical equipment

ZHAO Zhihong1,2, LI Chunxiu2, LI Lehao2, YANG Shaopu1

Author information +

文章历史 +

摘要

随着机械设备故障诊断技术的发展，利用深度学习技术判断设备故障类型越来越引起人们的重视。目前，基于注意力机制的Transformer模型有着优于卷积神经网络(convolutional neural network,CNN)的特征提取能力且在自然语言处理及计算机视觉领域都得到了成功的应用。提出一种用于机械设备故障诊断的Transformer方法（FaultDiagnosis-Transformer，简称FD-Transformer），首先对原始振动信号利用Dropout技术进行数据增强，提高模型的泛化能力；然后利用多通道一维卷积进行数据处理并得到矩阵形式；接着利用Dense连接的Encoder结构来进行机械设备的故障特征提取，最后利用分类模块得到故障诊断结果。分别采用变转速轴承数据和轮对轴承数据对模型进行实验验证，实验结果表明，该模型在两种数据集上均达到了99%以上的故障识别率，与CNN相比可以更好地提取机械设备故障特征，有一定的工程应用价值。

Abstract

With the development of mechanical equipment fault diagnosis technology, people pay more and more attention to using deep learning technology to judge the type of equipment fault. At present, Transformer model based on attention mechanism has better feature extraction ability than Convolutional Neural Network, and has been successfully applied in the fields of natural language processing and computer vision. A Transformer method for mechanical equipment fault diagnosis (FD-Transformer) is proposed. Firstly, the original vibration signal is enhanced by dropout technology to improve the generalization ability of the model; Then the matrix form is obtained by multi-channel one-dimensional convolution; Then, the Encoder structure connected by Dense is used to extract the fault features of mechanical equipment. Finally, the fault diagnosis results are obtained by using the classification module. The experimental results show that the model achieves a fault recognition rate of more than 99% on both data sets. Compared with CNN, it can better extract the fault characteristics of mechanical equipment, and has certain engineering application value.

导出引用

赵志宏1，2，李春秀2，李乐豪2，杨绍普1. 一种机械设备故障诊断的FD-Transformer方法[J]. 振动与冲击, 2023, 42(8): 326-333

ZHAO Zhihong1,2, LI Chunxiu2, LI Lehao2, YANG Shaopu1. A FD-Transformer method for fault diagnosis of mechanical equipment[J]. Journal of Vibration and Shock, 2023, 42(8): 326-333

参考文献

[1] X.Yan, M. Jia, A novel optimized SVM classifification algorithm with multi-domain feature and its application to fault diagnosis of rolling bearing, Neurocomputing 313 (2018) 47–64.
[2] H.Shao,H.Jiang,X.Zhang,M.Niu,Rolling bearing fault diagnosis using an optimization deep belief network, Meas. Sci. Technol. 26 (11) (2015).
[3] 赵志宏,杨绍普.一种基于样本熵的轴承故障诊断方法[J].振动与冲击,2012,31(06):136-140+154.
Zhao ZhiHong,Yang ShaoPu. Sample entropy-based roller bearing fault diagnosis method[J]. Journal of Vibration and Shock, 2012,31(06):136-140+154.
[4] 李兵,韩睿,何怡刚,等.改进随机森林算法在电机轴承故障诊断中的应用[J].中国电机工程学报,2020,40(04): 1310-1319+1422.
LI Bing, HAN Rui, He Yigang, et al.Applications of the Improved Random Forest Algorithm in Fault Diagnosis of Motor Bearings[J]. Proceedings of the CSEE,2020,40(04): 1310-1319+1422.
[5] 雷亚国，贾峰，周昕，等．基于深度学习理论的机械装备大数据健康监测方法[J]．机械工程学报，2015， 51(21)：49-56．
LEI Yaguo, JIA Feng,ZHOU Xin, et al.A Deep Learning-based Method for Machinery Health Monitoring with Big Data[J]． Journal of Mechanical Engineering,2015, 51(21)：49-56．
[6]   汪祖民,张志豪,秦静,等.基于卷积神经网络的机械
故障诊断技术综述[J/OL].计算机应用:1-10[2022-01-11].http://kns.cnki.net/kcms/detail/51.1307.tp.20211026.1657.010.html.
WANG Zumin, ZHANG Zhihao,QIN Jing, et al.Review of Mechanical Fault Diagnosis Technology Based on Convolutional Neural Network[J/OL].Journal of Computer Applications :1-10[2022-01-11].http://kns.cnki.net/kcms/detail/51.1307.tp.20211026.1657.010.html.
[7] 谢远东,雷文平,韩捷,等.全矢RNN的轴承故障诊断研究[J].机械设计与制造,2021(09):27-31.DOI:10.19356/j.cnki.1001-3997.2021.09.007.
XIE Yuandong,LEI Wenping,HAN Jie,et al.Research On Bearing Fault Diagnosis Using Full Vector RNN[J].Mechanical design and manufacturing engineering,2021(09):27-31.DOI:10.19356/j.cnki.1001-3997.2021.09.007.
[8] A. Vaswani, N. Shazeer, N. Parmar,et al, Attention is all you need, in: I. Guyon, U. von Luxburg, S. Bengio, H. M. Wallach, R. Fergus, S. V. N. Vishwanathan, R. Garnett (Eds.), Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017, December 4-9, 2017, Long Beach, CA, USA, 2017, pp. 5998–6008.
[9] J. F. Kolen, S. C. Kremer, Gradient Flow in Recurrent Nets: The Difficulty of Learning LongTerm Dependencies, in: A Field Guide to Dynamical Recurrent Networks.[S.l.]: Wiley-IEEE Press,2010.
[10] A. Gulati, J. Qin, C.-C. Chiu, et al. Conformer: Convolution-augmented Transformer for Speech Recognition, ArXiv200508100 Cs Eess (2020).
[11] B. Wu, C. Xu, X. Dai,et al. Visual Transformers: Token-based Image Representation and Processing for Computer Vision, ArXiv200603677 Cs Eess (2020).
[12] 刘建伟,刘俊文,罗雄麟.深度学习中注意力机制研究进展[J].工程科学学报,2021,v.43;No.331(11):1499-1511.
LIU Jianwei,LIU Junwen,LUO Xionglin.Research progress in attention mechanism in deep learning[J]. Chinese Journal of Engineering, 2021,v.43;No.331 (11):1499-1511.
[13] X. Li, W. Zhang, Q. Ding, Understanding and  improving deep learning-based rolling bearing fault diagnosis with attention mechanism, Signal Process. 161 (2019) 136–154.
[14] 朱浩,宁芊,雷印杰,等.基于注意力机制-Inception-CNN模型的滚动轴承故障分类[J].振动与冲击,2020,39(19):84-93.
ZHU Hao,NING Qian,LEI Yinjie,et al.Fault classificationof rolling bearing based on attention mechanism －inception －CNN Model[J].Journal of Vibration and Shock,2020,39(19):84-93.
[15] 王太勇,王廷虎,王鹏,等.基于注意力机制BiLSTM的设备智能故障诊断方法[J].天津大学学报(自然科学与工程技术版),2020,53(06):601-608.
WANG Taiyong,WANG Yanhu,WANGpeng,et al.An Intelligent Fault Diagnosis Method Based on Attention-Based Bidirectional LSTM Network[J]. Jouranl of Tianjin University (Science and Tenchology), 2020,53(06):601-608.
[16] Dosovitskiy A, Beyer L, Kolesnikov A, et al. An image is worth 16x16 words: Transformers for image recognition at scale[C]// Published as a Conference Paper at ICLR 2021.[S.l.]:ICLR,2021.
[17]   A. Dosovitskiy, L. Beyer, A. Kolesnikov,et al. An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale, ArXiv201011929 Cs (2020).
[18] HUANG G, LIU Z, LAURENS V D M, et al. Densely connected convolutional networks [C]// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2017: 4700-4708.
[19] H. Huang and N. Baddour, “Bearing vibration data collected under timevarying rotational speed conditions,” Data in Brief, 21, pp. 1745-1749, (2018).