Fault diagnosis using signal processing and deep learning-based image pattern recognition
-
Zhenxing Ren
and
Jianfeng Guo
Abstract
The vibration signal is a typical non-stationary signal, making it challenging to use traditional time-frequency analysis techniques for fault diagnosis. Therefore, this work investigates the processing of vibration signals and proposes a deep learning method based on processed signals for the fault diagnosis of ball bearings. In this work, the fault diagnosis is formulated as an image classification problem and solved with deep learning networks. The intrinsic mode functions (IMFs), converted from the vibration signals in the time domain, are then transformed into symmetrized dot pattern (SDP) images. In order to increase classification accuracy, the SDP parameters in this study are chosen by optimizing image similarity. The feasibility and accuracy of the proposed approach are examined experimentally.
Zusammenfassung
Ein Schwingungssignal ist ein typisches nichtstationäres Signal, was es schwierig macht, herkömmliche Zeit-Frequenz-Analysetechniken für die Fehlerdiagnose zu verwenden. Daher untersucht diese Arbeit die Verarbeitung von Schwingungssignalen und schlägt eine auf verarbeiteten Signalen basierende Deep-Learning-Methode für die Fehlerdiagnose von Kugellagern vor. In dieser Arbeit wird die Fehlerdiagnose als Bildklassifikationsproblem formuliert und mit Deep-Learning-Netzwerken gelöst. Die Intrinsic Mode Functions (IMFs), die aus den Schwingungssignalen im Zeitbereich umgewandelt werden, werden dann in symmetrische Punktmuster (SDP) umgewandelt. Um die Klassifikationsgenauigkeit zu erhöhen, werden die SDP-Parameter in dieser Studie durch Optimierung der Bildähnlichkeit ausgewählt. Die Machbarkeit und Genauigkeit des vorgeschlagenen Ansatzes werden experimentell untersucht.
Award Identifier / Grant number: 202103021223107
Funding source: National Natural Science Foundation of China
Award Identifier / Grant number: 12204345
Funding source: Natural Science Foundation of Shanxi Province
Award Identifier / Grant number: 20210302123188
About the authors
Zhenxing Ren was born in Taiyuan, China, on 21.02.1982. He received the B.S. in mechanical engineering from University Duisburg-Essen in Germany, M.S. degree in control and information systems from University Duisburg-Essen in Germany and Dr. Degree in control engineering from University Kassel in Germany. He is now a research fellow at the Taiyuan University of Technology in China. His research interests include machine learning, artificial intelligence, system identification, automatic control, etc.
Jianfeng Guo is a research fellow at the Taiyuan University of Technology in China. His research interests include acoustics, machine learning, active noise control, etc.
Acknowledgments
The authors would like to thank anonymous reviewers for carefully reading the paper.
-
Research ethics: Not applicable.
-
Author contributions: Zhenxing Ren: Project administration, Conceptualization, Methodology, Funding acquisition, Writing – original draft. Jianfeng Guo: Investigation, Data curation, Writing – review & editing.
-
Competing interests: The authors state no conflict of interest.
-
Research funding: This research is supported by the National Natural Science Foundation of China (Grant No. 12204345), Natural Science Foundation of Shanxi Province, China (Grant No. 20210302123188), and Natural Science Foundation for Young Scientists of Shanxi Province, China (Grant No. 202103021223107).
-
Data availability: The raw data can be obtained on request from the corresponding author.
References
[1] T. Schneider, S. Klein, and A. Schütze, “Machine learning in industrial measurement technology for detection of known and unknown faults of equipment and sensors,” TM – Tech. Mess., vol. 86, no. 11, pp. 706–718, 2019. https://doi.org/10.1515/teme-2019-0086.Search in Google Scholar
[2] H. Li, W. Wang, P. Huang, and Q. Li, “Fault diagnosis of rolling bearing using symmetrized dot pattern and density-based clustering,” Measurement, vol. 152, p. 107293, 2020, https://doi.org/10.1016/j.measurement.2019.107293.Search in Google Scholar
[3] M. Grzeszkowski, et al.., “Classification of gear pitting damage using vibration measurements,” TM – Tech. Mess., vol. 88, no. 5, pp. 282–293, 2021. https://doi.org/10.1515/teme-2021-0010.Search in Google Scholar
[4] T. Zhongda, L. Shujiang, W. Yanhong, and S. Yi, “A prediction method based on wavelet transform and multiple models fusion for chaotic time series,” Chaos Solit. Fractals, vol. 98, pp. 158–172, 2017, https://doi.org/10.1016/j.chaos.2017.03.018.Search in Google Scholar
[5] A. Dogra, B. Goyal, and S. Agrawal, “Bone vessel image fusion via generalized reisz wavelet transform using averaging fusion rule,” J. Comput. Sci., vol. 21, pp. 371–378, 2017, https://doi.org/10.1016/j.jocs.2016.10.009.Search in Google Scholar
[6] A. Dibaj, M. M. Ettefagh, R. Hassannejad, and M. B. Ehghaghi, “A hybrid fine-tuned VMD and CNN scheme for untrained compound fault diagnosis of rotating machinery with unequal-severity faults,” Expert Syst. Appl., vol. 167, p. 114094, 2021. https://doi.org/10.1016/j.eswa.2020.114094.Search in Google Scholar
[7] S. Lu, J. Wang, and Y. Xue, “Study on multi-fractal fault diagnosis based on EMD fusion in hydraulic engineering,” Appl. Therm. Eng., vol. 103, pp. 798–806, 2016, https://doi.org/10.1016/j.applthermaleng.2016.04.036.Search in Google Scholar
[8] D. Yang, H. Li, Y. Hu, J. Zhao, H. Xiao, and Y.-S. Lan, “Vibration condition monitoring system for wind turbine bearings based on noise suppression with multi-point data fusion,” Renewable Energy, vol. 92, pp. 104–116, 2016, https://doi.org/10.1016/j.renene.2016.01.099.Search in Google Scholar
[9] X. Xu, H. Liu, H. J. Zhu, and S. Wang, “Fan fault diagnosis based on symmetrized dot pattern analysis and image matching,” J. Sound Vib., vol. 374, pp. 297–311, 2016, https://doi.org/10.1016/j.jsv.2016.03.030.Search in Google Scholar
[10] K. Shibata, A. Takahashi, and T. Shirai, “Fault diagnosis of rotating machinery through visualisation of sound signals,” Mech. Syst. Signal Process., vol. 14, no. 2, pp. 229–241, 2000. https://doi.org/10.1006/mssp.1999.1255.Search in Google Scholar
[11] C. Yang and T. Feng, “Abnormal noise diagnosis of internal combustion engine using wavelet spatial correlation filter and symmetrized dot pattern,” Appl. Mech. Mater., vol. 141, pp. 168–173, 2011, https://doi.org/10.4028/www.scientific.net/amm.141.168.Search in Google Scholar
[12] X. Zhu, et al.., “Rotor fault diagnosis using a convolutional neural network with symmetrized dot pattern images,” Measurement, vol. 138, pp. 526–535, 2019, https://doi.org/10.1016/j.measurement.2019.02.022.Search in Google Scholar
[13] Y. K. Gu, L. Zeng, and G. Q. Qiu, “Bearing fault diagnosis with varying conditions using angular domain resampling technology, SDP and DCNN,” Measurement, vol. 156, 2020, Art. no. 107616, https://doi.org/10.1016/j.measurement.2020.107616.Search in Google Scholar
[14] R. Bassler, T. Bassler, and M. Kley, “Classification of load and rotational speed at wire-race bearings using Convolutional Neural Networks with vibration spectrograms,” TM – Tech. Mess., vol. 89, no. 5, pp. 352–362, 2022, https://doi.org/10.1515/teme-2021-0143.Search in Google Scholar
[15] Z. Ren, “An optimized excitation signal design for identification of PWA model and application to automotive throttles,” Meas. Control, vol. 56, no. 3–4, pp. 844–856, 2023. https://doi.org/10.1177/00202940221135903.Search in Google Scholar
[16] Z. Ren and X. Ji, “On prediction of air pollutants with Takagi-Sugeno models based on a hierarchical clustering identification method,” Atmos. Pollut. Res., vol. 14, no. 4, 2023, Art. no. 101731.10.1016/j.apr.2023.101731Search in Google Scholar
[17] T. Bäßler, R. Bäßler, and M. Kley, “Augmented mel-spectrogram VGG-16 model for axial and radial load classification at wire-race bearings,” TM – Tech. Mess., vol. 89, no. 9, pp. 573–579, 2022. https://doi.org/10.1515/teme-2022-0039.Search in Google Scholar
[18] O. Abdeljaber, S. Sassi, O. Avcı, S. Kiranyaz, A. A. Ibrahim, and M. Gabbouj, “Fault detection and severity identification of ball bearings by online condition monitoring,” IEEE Trans. Ind. Electron., vol. 66, no. 10, pp. 8136–8147, 2019. https://doi.org/10.1109/tie.2018.2886789.Search in Google Scholar
[19] Z. Xie, et al.., “Ball screw fault diagnosis based on continuous wavelet transform and two-dimensional convolution neural network,” Meas. Control, vol. 56, no. 3–4, pp. 518–528, 2023. https://doi.org/10.1177/00202940221107620.Search in Google Scholar
[20] S. J. Hao, F. X. Ge, Y. M. Li, and J. Y. Jiang, “Multisensor bearing fault diagnosis based on one-dimensional convolutional long short-term memory networks,” Measurement, vol. 159, p. 107802, 2020. https://doi.org/10.1016/j.measurement.2020.107802.Search in Google Scholar
[21] CWRU Bearing Data Center Website. Available at: http://csegroups.case.edu/bearingdatacenter/home.Search in Google Scholar
[22] Y. J. Sun, S. H. Li, Y. L. Wang, and X. H. Wang, “Fault diagnosis of rolling bearing based on empirical mode decomposition and improved manhattan distance in symmetrized dot pattern image,” Mech. Syst. Signal Process., vol. 159, p. 107817, 2021.10.1016/j.ymssp.2021.107817Search in Google Scholar
[23] N. E. Huang, “New method for nonlinear and nonstationary time series analysis: empirical mode decomposition and Hilbert spectral analysis,” in SPIE Defense + Commercial Sensing, 2000.10.1117/12.381681Search in Google Scholar
[24] G. Rilling, P. Flandrin, and P. Gonçalves, “On empirical mode decomposition and its algorithms. IEEE-EURASIP workshop on nonlinear signal and image processing,” Citeseer, vol. 3, no. 3, pp. 8–11, 2003.Search in Google Scholar
[25] Q. Fu, B. Jing, P. He, S. Si, and Y. Wang, “Fault feature selection and diagnosis of rolling bearings based on EEMD and optimized Elman–AdaBoost algorithm,” IEEE Sens. J., vol. 18, no. 12, pp. 5024–5034, 2018. https://doi.org/10.1109/jsen.2018.2830109.Search in Google Scholar
[26] C. Zhang, W. Wang, and H. Li, “Tool wear prediction method based on symmetrized dot pattern and multi-covariance Gaussian process regression,” Measurement, vol. 189, p. 110466, 2021.10.1016/j.measurement.2021.110466Search in Google Scholar
[27] Y. Sun, S.-Y. Li, and X. Wang, “Bearing fault diagnosis based on EMD and improved Chebyshev distance in SDP image,” Measurement, vol. 176, p. 109100, 2021, https://doi.org/10.1016/j.measurement.2021.109100.Search in Google Scholar
[28] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality assessment: from error visibility to structural similarity,” IEEE Trans. Image Process., vol. 13, no. 4, pp. 600–612, 2004. https://doi.org/10.1109/tip.2003.819861.Search in Google Scholar PubMed
[29] B. Wang, Y. Lei, N. Li, and N. Li, “A hybrid prognostics approach for estimating remaining useful life of rolling element bearings,” IEEE Trans. Reliab., vol. 69, no. 1, pp. 401–412, 2020. https://doi.org/10.1109/tr.2018.2882682.Search in Google Scholar
Supplementary Material
This article contains supplementary material (https://doi.org/10.1515/teme-2023-0089).
© 2024 Walter de Gruyter GmbH, Berlin/Boston
Articles in the same Issue
- Frontmatter
- Research Articles
- Messsystem zur Dehnungsmessung von Faserwerkstoffen basierend auf subjektiven Laser-Speckle-Mustern
- In-field calibration of triaxial accelerometer based on PE-ANGO
- Charakterisierung eines parallelkinematisch aktuierten In-situ-Referenzmesssystems für 5D-Nanomess- und Fabrikationsanwendungen
- Number of samples to use in estimating sinewave amplitude in the presence of noise
- Real-time imaging-ellipsometry on cylindrical substrates with a polarization camera
- Fault diagnosis using signal processing and deep learning-based image pattern recognition
- Erratum
- Erratum to: Potentials and challenges of deep-learning-assisted porosity prediction based on thermographic in-situ monitoring in laser powder bed fusion
Articles in the same Issue
- Frontmatter
- Research Articles
- Messsystem zur Dehnungsmessung von Faserwerkstoffen basierend auf subjektiven Laser-Speckle-Mustern
- In-field calibration of triaxial accelerometer based on PE-ANGO
- Charakterisierung eines parallelkinematisch aktuierten In-situ-Referenzmesssystems für 5D-Nanomess- und Fabrikationsanwendungen
- Number of samples to use in estimating sinewave amplitude in the presence of noise
- Real-time imaging-ellipsometry on cylindrical substrates with a polarization camera
- Fault diagnosis using signal processing and deep learning-based image pattern recognition
- Erratum
- Erratum to: Potentials and challenges of deep-learning-assisted porosity prediction based on thermographic in-situ monitoring in laser powder bed fusion
