Rolling Bearing Fault Diagnosis Based on SVD-GST Combined with Vision Transformer
<p>Transformer model structure.</p> "> Figure 2
<p>Transformer encoder model structure.</p> "> Figure 3
<p>Vision Transformer model framework.</p> "> Figure 4
<p>Overall model framework for rolling bearing fault diagnosis.</p> "> Figure 5
<p>Rolling bearing fault types: (<b>a</b>) inner ring fault; (<b>b</b>) rolling element fault; (<b>c</b>) cage fracture fault; (<b>d</b>) outer ring fault; and (<b>e</b>) normal state.</p> "> Figure 6
<p>The physical picture of the rolling bearing fault diagnosis experiment platform.</p> "> Figure 7
<p>Fault diagnosis experiment flowchart of rolling bearing.</p> "> Figure 8
<p>Singular value distribution curve of rolling bearing vibration signal.</p> "> Figure 9
<p>Singular value energy difference spectrum of rolling bearing vibration signal.</p> "> Figure 10
<p>Rolling bearing vibration signal before and after noise reduction display.</p> "> Figure 11
<p>Two-dimensional time–frequency diagram of rolling bearing: (<b>a</b>) cage fracture fault; (<b>b</b>) normal state; (<b>c</b>) inner ring fault; (<b>d</b>) rolling element fault; (<b>e</b>) outer ring fault.</p> "> Figure 12
<p>Loss value change curve.</p> "> Figure 13
<p>Accuracy transformation curve.</p> "> Figure 14
<p>Diagnosis results confusion matrix.</p> ">
Abstract
:1. Introduction
- (1)
- A rolling bearing fault diagnosis method based on SVD-GST combined with the Vision Transformer is proposed. A fault diagnosis experimental platform is built, and the model is verified to have high accuracy and feasibility through experiments.
- (2)
- In the process of using SVD noise reduction, the singular value energy difference spectrum is introduced to determine the order, which solves the problem of how to determine the effective order of the reconstruction matrix after the vibration signal of the rolling bearing is decomposed.
- (3)
- It is verified that the Vision Transformer model can mine more hidden fault information and reduce information loss for the two-dimensional vibration images of rolling bearings obtained using GST.
2. Principle Introduction
2.1. SVD
2.2. GST
2.3. Vision Transformer
3. Fault Diagnosis Model
3.1. Vision Transformer Model
3.2. Rolling Bearing Fault Diagnosis Model
4. Fault Diagnosis Platform Construction
5. Result Analysis
5.1. Rolling Bearing Vibration Signal Preprocessing
5.2. Two-Dimensional Time–Frequency Image Acquisition
5.3. Fault Diagnosis Model Analysis
5.4. Analysis of Fault Diagnosis Results
5.5. Comparison of Models
6. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Zhang, Q.; Gao, J.; Dong, H.; Mao, Y. WPD and DE/BBO-RBFNN for solution of rolling bearing fault diagnosis. Neurocomputing 2018, 312, 27–33. [Google Scholar] [CrossRef]
- Li, L.; Meng, W.; Liu, X.; Fei, J. Research on Rolling Bearing Fault Diagnosis Based on Variational Modal Decomposition Parameter Optimization and an Improved Support Vector Machine. Electronics 2023, 12, 1290. [Google Scholar] [CrossRef]
- Zhou, J.; Xiao, M.; Niu, Y.; Ji, G. Rolling Bearing Fault Diagnosis Based on WGWOA-VMD-SVM. Sensors 2022, 22, 6281. [Google Scholar] [CrossRef] [PubMed]
- Fuan, W.; Hongkai, J.; Haidong, S.; Wenjing, D.; Shuaipeng, W. An adaptive deep convolutional neural network for rolling bearing fault diagnosis. Meas. Sci. Technol. 2017, 28, 095005. [Google Scholar] [CrossRef]
- Meng, Z.; Zhan, X.; Li, J.; Pan, Z. An enhancement denoising autoencoder for rolling bearing fault diagnosis. Measurement 2018, 130, 448–454. [Google Scholar] [CrossRef]
- Li, X.; Jiang, H.; Wang, R.; Niu, M. Rolling bearing fault diagnosis using optimal ensemble deep transfer network. Knowl. Based Syst. 2021, 213, 106695. [Google Scholar] [CrossRef]
- Jia, F.; Lei, Y.; Guo, L.; Lin, J.; Xing, S. A neural network constructed by deep learning technique and its application to intelligent fault diagnosis of machines. Neurocomputing 2018, 272, 619–628. [Google Scholar] [CrossRef]
- Cheng, C.; Wang, J.; Chen, H.; Chen, Z.; Luo, H.; Xie, P. A Review of Intelligent Fault Diagnosis for High-Speed Trains: Qualitative Approaches. Entropy 2021, 23, 1. [Google Scholar] [CrossRef] [PubMed]
- Duan, L.; Xie, M.; Wang, J.; Bai, T. Deep learning enabled intelligent fault diagnosis: Overview and applications. J. Intell. Fuzzy Syst. 2018, 35, 5771–5784. [Google Scholar] [CrossRef]
- Li, X.; Zhang, W.; Ding, Q.; Sun, J.Q. Multi-layer domain adaptation method for rolling bearing fault diagnosis. Signal Process. 2019, 157, 180–197. [Google Scholar] [CrossRef]
- Liu, X.; Zhao, X.; He, K. Feasibility Study of the GST-SVD in Extracting the Fault Feature of Rolling Bearing under Variable Conditions. Chin. J. Mech. Eng. 2022, 35, 1–14. [Google Scholar] [CrossRef]
- Pan, H.; He, X.; Tang, S.; Meng, F. An improved bearing fault diagnosis method using one-dimensional CNN and LSTM. Stroj. Vestn./J. Mech. Eng. 2018, 64, 443–452. [Google Scholar]
- Huang, M.; Huang, T.; Zhao, Y.; Dai, W. Fault diagnosis of rolling bearing based on empirical mode decomposition and convolutional recurrent neural network. In Conference Series: Materials Science and Engineering; IOP Publishing: Bristol, UK, 2021; Volume 1043, p. 42015. [Google Scholar]
- Lu, Z.; Qin, Y.; Cheng, X.; Zhang, S.; Zeng, Y. Bearing Fault Diagnosis Method of Bearing Based on LSTM Auto-Encoder. In International Conference on Electrical and Information Technologies for Rail Transportation; Springer: Singapore, 2021; pp. 582–591. [Google Scholar]
- Tang, X.; Xu, Z.; Wang, Z. A Novel Fault Diagnosis Method of Rolling Bearing Based on Integrated Vision Transformer Model. Sensors 2022, 22, 3878. [Google Scholar] [CrossRef]
- Ding, Y.; Jia, M.; Miao, Q.; Cao, Y. A novel time–frequency Transformer based on self–attention mechanism and its application in fault diagnosis of rolling bearings. Mech. Syst. Signal Process. 2022, 168, 108616. [Google Scholar] [CrossRef]
- Wang, J.; Mo, Z.; Zhang, H.; Miao, Q. A deep learning method for bearing fault diagnosis based on time-frequency image. IEEE Access 2019, 7, 42373–42383. [Google Scholar] [CrossRef]
- Kim, J.H. Time frequency image and artificial neural network based classification of impact noise for machine fault diagnosis. Int. J. Precis. Eng. Manuf. 2018, 19, 821–827. [Google Scholar] [CrossRef]
- Yang, B.; Liu, R.; Chen, X. Fault diagnosis for a wind turbine generator bearing via sparse representation and shift-invariant K-SVD. IEEE Trans. Ind. Inform. 2017, 13, 1321–1331. [Google Scholar] [CrossRef]
- Yang, Z.-X.; Zhong, J.-H. A Hybrid EEMD-Based SampEn and SVD for Acoustic Signal Processing and Fault Diagnosis. Entropy 2016, 18, 112. [Google Scholar] [CrossRef]
- Cheng, H.; Zhang, Y.; Lu, W.; Yang, Z. A bearing fault diagnosis method based on VMD-SVD and Fuzzy clustering. Int. J. Pattern Recognit. Artif. Intell. 2019, 33, 1950018. [Google Scholar] [CrossRef]
- Tian, Y.; Ma, J.; Lu, C.; Wang, Z. Rolling bearing fault diagnosis under variable conditions using LMD-SVD and extreme learning machine. Mech. Mach. Theory 2015, 90, 175–186. [Google Scholar] [CrossRef]
- Ma, Q.; Huang, D.; Yang, J. Adaptive stochastic resonance in second-order system with general scale transformation for weak feature extraction and its application in bearing fault diagnosis. Fluct. Noise Lett. 2018, 17, 1850009. [Google Scholar] [CrossRef]
- Qin, Y.; Shi, X. Fault Diagnosis Method for Rolling Bearings Based on Two-Channel CNN under Unbalanced Datasets. Appl. Sci. 2022, 12, 8474. [Google Scholar] [CrossRef]
- Liu, X.; He, Y.; Wang, L. Adaptive Transfer Learning Based on a Two-Stream Densely Connected Residual Shrinkage Network for Transformer Fault Diagnosis over Vibration Signals. Electronics 2021, 10, 2130. [Google Scholar] [CrossRef]
- Vaswani, A.; Shazeer, N.; Parmar, N.; Uszkoreit, J.; Jones, L.; Gomez, A.N.; Polosukhin, I. Attention is all you need. arXiv 2017, arXiv:1706.03762v7. [Google Scholar]
- Han, K.; Xiao, A.; Wu, E.; Guo, J.; Xu, C.; Wang, Y. Transformer in transformer. Adv. Neural Inf. Process. Syst. 2021, 34, 15908–15919. [Google Scholar]
- Dosovitskiy, A.; Beyer, L.; Kolesnikov, A.; Weissenborn, D.; Zhai, X.; Unterthiner, T.; Houlsby, N. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv 2020, arXiv:2010.11929. [Google Scholar]
- Xu, Z.; Tang, X.; Wang, Z. A Multi-Information Fusion ViT Model and Its Application to the Fault Diagnosis of Bearing with Small Data Samples. Machines 2023, 11, 277. [Google Scholar] [CrossRef]
- Liang, P.; Yu, Z.; Wang, B.; Xu, X.; Tian, J. Fault transfer diagnosis of rolling bearings across multiple working conditions via subdomain adaptation and improved vision transformer network. Adv. Eng. Inform. 2023, 57, 102075. [Google Scholar] [CrossRef]
- Chen, Z.; Chen, J.; Liu, S.; Feng, Y.; He, S.; Xu, E. Multi-channel Calibrated Transformer with Shifted Windows for few-shot fault diagnosis under sharp speed variation. ISA Trans. 2022, 131, 501–515. [Google Scholar] [CrossRef] [PubMed]
- Jin, Y.; Hou, L.; Chen, Y. A time series transformer based method for the rotating machinery fault diagnosis. Neurocomputing 2022, 494, 379–395. [Google Scholar] [CrossRef]
- Xu, Y.; Li, Y.; Wang, Y.; Zhong, D.; Zhang, G. Improved few-shot learning method for transformer fault diagnosis based on approximation space and belief functions. Expert Syst. Appl. 2021, 167, 114105. [Google Scholar] [CrossRef]
Patch Size | Layers | Hidden Size D | MLP Size | Heads | Params |
---|---|---|---|---|---|
12 | 768 | 3072 | 12 | 86 M |
Category | Rolling Bearing Status | Motor Speed (Hz) | Brake Load (A) | Length | Number of Data Sets |
---|---|---|---|---|---|
1 | cage fracture fault | 30 | 0 | 1024 | 1000 |
2 | normal state | 30 | 0 | 1024 | 1000 |
3 | inner ring fault | 30 | 0 | 1024 | 1000 |
4 | rolling element fault | 30 | 0 | 1024 | 1000 |
5 | outer ring fault | 30 | 0 | 1024 | 1000 |
Gearbox Status | Number of Training Sets | Number of Validation Sets | Number of Test Sets | Label |
---|---|---|---|---|
cage fracture fault | 700 | 200 | 100 | 0 |
normal state | 700 | 200 | 100 | 1 |
inner ring fault | 700 | 200 | 100 | 2 |
rolling element fault | 700 | 200 | 100 | 3 |
outer ring fault | 700 | 200 | 100 | 4 |
total number of samples | 3500 | 1000 | 500 |
Fault Diagnosis Model | 10 Average Accuracy % | Standard Deviation % |
---|---|---|
SVD-GST-2DCNN | 95.24 | 1.2933 |
STFT-CNN-LSTM | 92.50 | 0.6520 |
SVD-GST-LSTM | 94.28 | 1.7863 |
GST-ViT | 91.06 | 0.9834 |
SVD-GST-ViT | 98.52 | 0.4266 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Xie, F.; Wang, G.; Zhu, H.; Sun, E.; Fan, Q.; Wang, Y. Rolling Bearing Fault Diagnosis Based on SVD-GST Combined with Vision Transformer. Electronics 2023, 12, 3515. https://doi.org/10.3390/electronics12163515
Xie F, Wang G, Zhu H, Sun E, Fan Q, Wang Y. Rolling Bearing Fault Diagnosis Based on SVD-GST Combined with Vision Transformer. Electronics. 2023; 12(16):3515. https://doi.org/10.3390/electronics12163515
Chicago/Turabian StyleXie, Fengyun, Gan Wang, Haiyan Zhu, Enguang Sun, Qiuyang Fan, and Yang Wang. 2023. "Rolling Bearing Fault Diagnosis Based on SVD-GST Combined with Vision Transformer" Electronics 12, no. 16: 3515. https://doi.org/10.3390/electronics12163515