Rolling bearing fault study based on an improved multi-scale convolutional recurrent neural network

PDF(2639 KB)

Journal of Vibration and Shock ›› 2024, Vol. 43 ›› Issue (20) : 94-105.

DONG Shaojiang，HUANG Xiang，XIA Zongyou，ZOU Song

Author information +

History +

Abstract

A novel fault diagnosis method is proposed, which combines a multi-scale convolutional neural network (MSCNN) with a bi-directional long short-term memory network (BiLSTM) using an attention mechanism. This approach addresses the issue of feature extraction in traditional fault diagnosis methods, which often result in limited representation of fault information and the inability to deeply explore fault characteristics under complex working conditions. Firstly, the method employs pooling layers and convolutional kernels of different sizes to capture multi-scale features from vibration signals. Then, a multi-head self-attention mechanism (MHSA) is introduced to automatically assign different weights to different parts of the feature sequence, further enhancing the ability to represent features. Additionally, the BiLSTM structure is used to extract the internal relationships between features before and after, enabling the progressive transmission of information. Finally, the maximum-kernel mean discrepancy (MK-MMD) is utilized to reduce the distribution differences between the source and target domains at various layers of the pre-trained model, and a small amount of labeled target domain data is used to further train the model. The experimental results show that the proposed method has an average accuracy of 98.43% and 97.66% on the JNU and PU open bearing datasets, respectively, and the method also shows a very high accuracy and fast convergence speed on the bearing fault dataset (CME) made by Chongqing Changjiang Bearing Co. and provides a practical basis for the effective diagnosis of vibration rotating component faults.

Key words

fault diagnosis / multi-scale convolutional neural network / bi-directional long short-term memory / multi-head self-attention / maximum-kernel mean discrepancy

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

DONG Shaojiang, HUANG Xiang, XIA Zongyou, ZOU Song. Rolling bearing fault study based on an improved multi-scale convolutional recurrent neural network[J]. Journal of Vibration and Shock, 2024, 43(20): 94-105

References

[1] Zhang J , Sun Y , Guo L ,et al.A new bearing fault diagnosis method based on modified convolutional neural networks [J].Chinese Journal of Aeronautics, 2019, 33(2). DOI:10.1016/j.cja.2019.07.011.
[2] ZHAO Minghang，ZHONG Shisheng，FU Xuyun，et al. Deep residual shrinkage networks for fault diagnosis[J]. IEEE Transactions on Industrial Informatics，2020，16(7)：4681-4690.
[3] 王海涛,郭一帆,史丽晨.基于SC-ResNeSt及频域格拉姆角场的滚动轴承故障诊断方法[J/OL].计算机集成制造系统:1-21[2024-02-04].http://kns.cnki.net/kcms/detail/11.5946.TP.20230207.1619.002.html.
Wang Haitao, Guo Yifan, Shi Lichen. Rolling bearing fault diagnosis method based on SC-ResNeSt and frequency domain Gram angle field [J/OL]. Computer Integrated Manufacturing System: 1-21 [2024-02-04]. http://kns. cnki.net/kcms/detail/11.5946.TP.20230207.1619.002.html.
[4] Shan S , Liu J , Wu S ,et al.A motor bearing fault voiceprint recognition method based on Mel-CNN model[J].Measurement, 2023, 207:112408-.
[5] 李喆,吐松江•卡日,范想等.基于双图转换和融合CRNN网络的轴承故障诊断[J].振动与冲击,2023,42(19):240-248.DOI:10.13465/j.cnki.jvs.2023.19.03.
Li Zhe, Tu Songjiang•Kari, Fan Xiang, et al. Bearing fault diagnosis based on dual graph conversion and fusion CRNN network [J]. Vibration and Impact, 2023, 42(19): 240-248. DOI: 10.13465 /j.cnki.jvs.2023.19.031.
[6] 余萍,赵康,曹洁.基于优化A-BiLSTM的滚动轴承故障诊断[J/OL].吉林大学学报(工学版).https://kns.cnki.net/kcms/detail/22.1341.T.20230327.0901.003.html.
Yu Ping, Zhao Kang, Cao Jie. Rolling bearing fault diagnosis based on optimized A-BiLSTM [J/OL]. Journal of Jilin University (Engineering Edition): 1-10 [2024-02-04]. http:// kns.cnki.net/kcms/detail/22.1341.T.20230327.0901.003.html.
[7] 陈保家,陈学力,沈保明,等.CNN-LSTM深度神经网络在滚动轴承故障诊断中的应用[J].西安交通大学学报, 2021(006):055.
Chen Baojia, Chen Xueli, Shen Baoming, et al. Application of CNN-LSTM deep neural network in rolling bearing fault diagnosis [J]. Journal of Xi'an Jiaotong University, 2021, 55(06): 28-36.
[8] H. Wang, M. Li, Z. Liu, X. Dai, R. Wang and L. Shi, "Rotary Machinery Fault Diagnosis Based on Split Attention Mechanism and Graph Convolutional Domain Adaptive Adversarial Network," in IEEE Sensors Journal, doi: 10.1109/JSEN.2023.3348597.
[9] 刘飞,陈仁文,邢凯玲,等.基于迁移学习与深度残差网络的滚动轴承快速故障诊断算法[J].振动与冲击, 2022, 41(3):11.
Liu Fei, Chen Renwen, Xing Kailing, et al. Fast fault diagnosis algorithm for rolling bearings based on transfer learning and deep residual network [J]. Vibration and Impact, 2022, 41(03): 154-164. DOI: 10.13465/j. cnki.jvs.2022.03.019.
[10] 陈祝云,钟琪,黄如意,等.基于增强迁移卷积神经网络的机械智能故障诊断[J].机械工程学报, 2021, 57(21):10.DOI:10.3901/JME.2021.21.096.
Chen Zhuyun, Zhong Qi, Huang Ruyi, et al. Mechanical intelligent fault diagnosis based on enhanced migration convolutional neural network [J]. Transactions of the Chinese Society of Mechanical Engineering, 2021, 57(21): 96-105.
[11] Fan H , Xue C , Zhang X ,et al.Vibration Images-Driven Fault Diagnosis Based on CNN and Transfer Learning of Rolling Bearing under Strong Noise[J].Shock and Vibration, 2021.DOI:10.1155/2021/6616592.
[12] Hou W , Zhang C , Jiang Y ,et al.A new bearing fault diagnosis method via simulation data driving transfer learning without target fault data[J].Measurement, 2023.
[13] Vaswani A , Shazeer N , Parmar N ,et al.Attention Is All You Need[J].arXiv, 2017.DOI:10.48550/arXiv.1706.03762.
[14] Wang P , Xiong H , He H .Bearing fault diagnosis under various conditions using an incremental learning-based multi-task shared classifier[J].Knowledge-based systems, 2023.
[15] Ma R , Han T , Lei W .Cross-domain meta learning fault diagnosis based on multi-scale dilated convolution and adaptive relation module[J].Knowledge-Based Systems, 2023, 261:110175-.
[16] 江南大学轴承数据集[EB/OL].https://github.com/ClarkGab leWang/JNU-Bearing-Dataset.
[17] 帕德伯恩大学轴承数据集[EB/OL].http://groups.uni-pader born.de/kat/BearingDataCenter/.
[18] Wang Q , Wu B , Zhu P ,et al.ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).IEEE, 2020.DOI:10.1109/42600.2020.
[19] Hu J , Shen L , Sun G ,et al.Squeeze-and-Excitation Networks[C]//IEEE Transactions on Pattern Analysis and Machine Intelligence.IEEE, 2017.DOI:10.1109/2019.2913372.
[20] 唐世钰,童靳于,郑近德等.改进的深度残差收缩网络轴承故障诊断方法[J].振动与冲击,2023,42(18):217-224+285.DOI:10.13465/j.cnki.jvs.2023.18.024.
Tang Shiyu, Tong Jinyu, Zheng Jinde, etc. Improved deep residual shrinkage network bearing fault diagnosis method [J]. Vibration and Impact, 2023, 42(18): 217-224+285.DOI: 10.13465/j. cnki.jvs.2023.18.024.
[21] Shen J , Qu Y , Zhang W ,et al.Wasserstein Distance Guided Representation Learning for Domain Adaptation[J]. 2017.DOI:10.48550/arXiv.1707.01217.