Multi-Mode Data Generation and Fault Diagnosis of Bearings Based on STFT-SACGAN
Abstract
:1. Introduction
2. The Brief Theories of GANs
2.1. GAN
2.2. SGAN
2.3. ACGAN
3. The Proposed SACGAN
3.1. Architecture of SACGAN
3.2. Structure of Generator
3.3. Structure of the Discriminator
3.4. Loss Functions of SACGAN
4. Fault Diagnosis Based on STFT-SACGAN
4.1. Data Processing
4.2. Fault Diagnosis Flow Based on STFT-SACGAN
- (1)
- Vibration signals of bearings with various fault modes are collected, converted into time-frequency images by STFT, and normalized into the interval [−1,1];
- (2)
- The noise vector z and the label vector c are input to the generator to obtain generated samples;
- (3)
- Labeled real samples, unlabeled real samples, and generated samples from the generator are fed into the discriminator to obtain discrimination and classification results.
- (4)
- Calculate the loss of the generator and the discriminator;
- (5)
- Fix the generator’s weight parameters and optimize the discriminator’s weight parameters;
- (6)
- Fix the discriminator’s weight parameters and optimize the generator’s weight parameters;
- (7)
- Repeat steps 2–6 until the number of iterations is satisfied;
- (8)
- Save the trained model, use the generator to generate multi-mode fault samples, and use the discriminator for fault diagnosis of test signals.
5. Case Study
5.1. Case 1 Multi-Mode Data Generation and Fault Diagnosis of Bearings
5.1.1. Introduction and Preprocessing of Bearing Data
5.1.2. Quality Evaluation and Comparison of Multi-Mode Generated Samples
5.1.3. Fault Recognition with Different Label Ratios
5.2. Case 2 Multi-Mode Data Generation and Fault Diagnosis of Bearings
5.2.1. Introduction and Preprocessing of Bearing Data
5.2.2. Multi-Mode Fault Sample Generation and Fault Recognition in Case 2
6. Conclusions
Author Contributions
Funding
Data Availability Statement
Conflicts of Interest
References
- Lei, Y.G.; Yang, B.; Jiang, X.W.; Jia, F.; Li, N.P.; Nandi, A. Applications of machine learning to machine fault diagnosis: A review and roadmap. Mech. Syst. Signal Process. 2020, 138, 106587–106625. [Google Scholar] [CrossRef]
- Sun, Y.J.; Wang, J.; Wang, X.H. Fault diagnosis of mechanical equipment in high energy consumption industries in China: A review. Mech. Syst. Signal Process. 2023, 186, 109833–109865. [Google Scholar] [CrossRef]
- Zhao, Z.B.; Li, T.F.; Wu, J.Y.; Sun, C.; Wang, S.B.; Yan, R.Q.; Chen, X.F. Deep learning algorithms for rotating machinery intelligent diagnosis: An open source benchmark study. ISA Trans. 2020, 107, 224–255. [Google Scholar] [CrossRef] [PubMed]
- Cen, J.; Yang, Z.H.; Liu, X.; Xiong, J.B.; Chen, H.H. A Review of Data-Driven Machinery Fault Diagnosis Using Machine Learning Algorithms. J. Vib. Eng. Technol. 2022, 10, 2481–2507. [Google Scholar] [CrossRef]
- Jalayer, M.; Orsenigo, C.; Vercellis, C. Fault detection and diagnosis for rotating machinery: A model based on convolutional LSTM, Fast Fourier and continuous wavelet transforms. Comput. Ind. 2021, 125, 103378–103393. [Google Scholar] [CrossRef]
- Przystupa, K.; Ambrożkiewicz, B.; Litak, G. Diagnostics of Transient States in Hydraulic Pump System with Short Time Fourier Transform. Adv. Sci. Technol. Res. J. 2020, 14, 178–183. [Google Scholar] [CrossRef]
- Lepicka, M.; Górski, G.; Gradzka-Dahlke, M.; Litak, G.; Ambrożkiewicz, B. Analysis of tribological behaviour of titanium nitride-coated stainless steel with the use of wavelet-based methods. Arch. Appl. Mech. 2021, 91, 4475–4483. [Google Scholar] [CrossRef]
- Kibrete, F.; Woldmichael, D.E. Applications of Artificial Intelligence for Fault Diagnosis of Rotating Machines: A Review. Inst. Comput. Sci. Soc. Inform. Telecommun. Eng. 2023, 455, 41–62. [Google Scholar]
- Zhu, Z.Q.; Lei, Y.B.; Qi, G.Q.; Chai, Y.; Mazur, N.; An, Y.; Huang, X.H. A review of the application of deep learning in intelligent fault diagnosis of rotating machinery. Measurement 2023, 206, 112346–112369. [Google Scholar] [CrossRef]
- Zhao, Z.B.; Wu, J.Y.; Li, T.F.; Sun, C.; Yan, R.Q.; Chen, X.F. Challenges and Opportunities of AI-Enabled Monitoring, Diagnosis & Prognosis: A Review. Chin. J. Mech. Eng. 2021, 34, 16–44. [Google Scholar]
- Zhang, T.C.; Chen, J.L.; Li, F.D.; Zhang, K.Y.; Lv, H.X.; He, S.L.; Xu, E.Y. Intelligent fault diagnosis of machines with small & imbalanced data: A state-of-the-art review and possible extensions. ISA Trans. 2021, 119, 152–171. [Google Scholar] [PubMed]
- Zhao, Z.B.; Zhang, Q.Y.; Yu, X.L.; Sun, C.; Wang, S.B.; Yan, R.Q.; Chen, X.F. Applications of Unsupervised Deep Transfer Learning to Intelligent Fault Diagnosis: A Survey and Comparative Study. IEEE Trans. Instrum. Meas. 2021, 70, 3525828–3525855. [Google Scholar] [CrossRef]
- Pan, T.Y.; Chen, J.L.; Zhang, T.C.; Liu, S.; He, S.L.; Lv, H.X. Generative adversarial network in mechanical fault diagnosis under small sample: A systematic review on applications and future perspectives. ISA Trans. 2021, 128, 1–10. [Google Scholar] [CrossRef] [PubMed]
- Pang, X.Y.; Wei, Z.H.; Tong, Y. Fault Diagnosis Method of Gear Based on SCGAN Network. J. Vib. Meas. Diagn. 2022, 42, 358–364. [Google Scholar]
- Xing, X.S.; Guo, W. Intelligent diagnosis method for bearing with few labelled samples based on an improved semi-supervised learning-based generative adversarial network. J. Vib. Shock 2022, 41, 184–192. [Google Scholar]
- Yang, Q.; Zhang, J.Y.; Wu, D.S.; Liu, Y.P. Fault Diagnosis for Rolling Bearings Based on Two-Dimensional Image and Switchable Normalization SGAN Network. Bearing 2021, 8, 39–46. [Google Scholar]
- Lu, J.L.; Zhang, X.G.; Zhang, W.; Guo, L.Y.; Wen, R.T. Fault Diagnosis of Main Bearing of Wind Turbine Based on Improved Auxiliary Classifier Generative Adversarial Network. Autom. Electr. Power Syst. 2021, 45, 148–154. [Google Scholar]
- Li, D.D.; Liu, Y.H.; Zhao, Y.; Zhao, Y. Fault Diagnosis Method of Wind Turbine Planetary Gearbox Based on Improved Generative Adversarial Network. Proc. CSEE 2021, 41, 7496–7507. [Google Scholar]
- Li, W.; Zhong, X.; Shao, H.D.; Cai, B.P.; Yang, X.K. Multi-mode data augmentation and fault diagnosis of rotating machinery using modified ACGAN designed with new framework. Adv. Eng. Inform. 2022, 52, 101552–101567. [Google Scholar] [CrossRef]
- He, W.P.; Chen, J.; Zhou, Y.; Liu, X.; Chen, B.Q.; Guo, B.L. An Intelligent Machinery Fault Diagnosis Method Based on GAN and Transfer Learning under Variable Working Conditions. Sensors 2022, 22, 9175. [Google Scholar] [CrossRef]
- He, Y.; Tang, H.S.; Ren, Y.; Kumar, A. A semi-supervised fault diagnosis method for axial piston pump bearings based on DCGAN. Meas. Sci. Technol. 2021, 32, 125104–125122. [Google Scholar] [CrossRef]
- Meng, Z.; Li, Q.; Sun, D.Y.; Cao, W.; Fan, F.J. An Intelligent Fault Diagnosis Method of Small Sample Bearing Based on Improved Auxiliary Classification Generative Adversarial Network. IEEE Sens. J. 2022, 22, 19543–19555. [Google Scholar] [CrossRef]
- Gao, Y.D.; Piltan, F.; Kim, J.M. A Novel Image-Based Diagnosis Method Using Improved DCGAN for Rotating Machinery. Sensors 2022, 22, 7534. [Google Scholar] [CrossRef] [PubMed]
- Li, H.; Zhang, Q.; Qin, X.R.; Sun, Y.T. Fault diagnosis method for rolling bearings based on short-time Fourier transform and convolution neural network. J. Vib. Shock 2018, 37, 124–131. [Google Scholar]
- Tao, H.F.; Wang, P.; Chen, Y.Y.; Stojanovic, V.; Yang, H.Z. An unsupervised fault diagnosis method for rolling bearing using STFT and generative neural networks. J. Frankl. Inst. 2020, 357, 7286–7307. [Google Scholar] [CrossRef]
- Kingma, D.P.; Ba, J. Adam: A Method for Stochastic Optimization. arXiv 2014, arXiv:1412.6980. [Google Scholar]
- Radford, A.; Metz, L.; Chintala, S. Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks. arXiv 2015, arXiv:1511.06434. [Google Scholar]
- Smith, W.; Randall, R. Rolling element bearing diagnostics using the Case Western Reserve University data: A benchmark study. Mech. Syst. Signal Process. 2015, 64–65, 100–131. [Google Scholar] [CrossRef]
- Setiadi, D. PSNR vs. SSIM: Imperceptibility quality assessment for image steganography. Multimed. Tools Appl. 2021, 80, 8423–8444. [Google Scholar] [CrossRef]
- Pezzotti, N.; Lelieveldt, B.; Maaten, L.; Höllt, T.; Eisemann, E.; Vilanova, A. Approximated and User Steerable tSNE for Progressive Visual Analytics. IEEE Trans. Vis. Comput. Graph. 2017, 23, 1739–1752. [Google Scholar] [CrossRef]
- Wang, B.; Lei, Y.G.; Li, N.P.; Li, N.B. A Hybrid Prognostics Approach for Estimating Remaining Useful Life of Rolling Element Bearings. IEEE Trans. Reliab. 2020, 69, 401–412. [Google Scholar] [CrossRef]
Layer Type | Kernel Size | Kernel Num | Strides | Output Size |
---|---|---|---|---|
Embedding | / | / | / | 200 × 1 × 1 |
Deconv1 | 200 | 3 × 3 | 2 × 2 | 200 × 3 × 3 |
Deconv2 | 64 | 3 × 3 | 2 × 2 | 64 × 7 × 7 |
Deconv3 | 32 | 3 × 3 | 2 × 2 | 32 × 15 × 15 |
Deconv4 | 16 | 3 × 3 | 2 × 2 | 16 × 32 × 32 |
Deconv5 | 1 | 4 × 4 | 2 × 2 | 1 × 64 × 64 |
Layer Type | Kernel Size | Kernel Num | Strides | Padding | Output Size |
---|---|---|---|---|---|
Conv1 | 32 | 5 × 5 | 4 × 4 | 2 × 2 | 32 × 16 × 16 |
Conv2 | 64 | 5 × 5 | 2 × 2 | 2 × 2 | 64 × 8 × 8 |
Conv3 | 128 | 5 × 5 | 2 × 2 | 2 × 2 | 128 × 4 × 4 |
FC | / | / | / | / | 1 |
/ | / | / | / | K 1 |
Fault Type | Fault Diameter | Class Labels |
---|---|---|
Normal | - | Normal |
Rolling ball | 0.007 inch | BA007 |
Rolling ball | 0.014 inch | BA014 |
Rolling ball | 0.021 inch | BA021 |
Inner race | 0.007 inch | IR007 |
Inner race | 0.014 inch | IR014 |
Inner race | 0.021 inch | IR021 |
Outer race | 0.007 inch | OR007 |
Outer race | 0.014 inch | OR014 |
Outer race | 0.021 inch | OR021 |
Fault Type | Class Labels |
---|---|
Outer ring | OR1 |
Outer ring | OR2 |
Outer ring | OR3 |
Cage | Cage |
Inner ring and outer ring | IR&OR |
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
Wang, H.; Zhu, H.; Li, H. Multi-Mode Data Generation and Fault Diagnosis of Bearings Based on STFT-SACGAN. Electronics 2023, 12, 1910. https://doi.org/10.3390/electronics12081910
Wang H, Zhu H, Li H. Multi-Mode Data Generation and Fault Diagnosis of Bearings Based on STFT-SACGAN. Electronics. 2023; 12(8):1910. https://doi.org/10.3390/electronics12081910
Chicago/Turabian StyleWang, Hongxing, Hua Zhu, and Huafeng Li. 2023. "Multi-Mode Data Generation and Fault Diagnosis of Bearings Based on STFT-SACGAN" Electronics 12, no. 8: 1910. https://doi.org/10.3390/electronics12081910
APA StyleWang, H., Zhu, H., & Li, H. (2023). Multi-Mode Data Generation and Fault Diagnosis of Bearings Based on STFT-SACGAN. Electronics, 12(8), 1910. https://doi.org/10.3390/electronics12081910