Deep Transfer Learning for Machine Diagnosis: From Sound and Music Recognition to Bearing Fault Detection
Source abstract
Abstract
Today’s deep learning strategies require ever-increasing computational efforts and demand for very large amounts of labelled data. Providing such expensive resources for machine diagnosis is highly challenging. Transfer learning recently emerged as a valuable approach to address these issues. Thus, the knowledge learned by deep architectures in different scenarios can be reused for the purpose of machine diagnosis, minimizing data collecting efforts. Existing research provides evidence that networks pre-trained for image recognition can classify machine vibrations in the time-frequency domain by means of transfer learning. So far, however, there has been little discussion about the potentials included in networks pre-trained for sound recognition, which are inherently suited for time-frequency tasks. This work argues that deep architectures trained for music recognition and sound detection can perform machine diagnosis. The YAMNet convolutional network was designed to serve extremely efficient mobile applications for sound detection, and it was originally trained on millions of data extracted from YouTube clips. That framework is employed to detect bearing faults for the CWRU dataset. It is shown that transferring knowledge from sound and music recognition to bearing fault detection is successful. The maximum accuracy is achieved using a few hundred data for fine-tuning the fault diagnosis model.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Machine Fault Diagnosis Techniques · Anomaly Detection Techniques and Applications · Water Systems and Optimization
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Dipartimento di Ingegneria Meccanica e Aerospaziale (DIMEAS), Politecnico di Torino, Corso Duca Degli Abruzzi 24, 10129 Torino, Italy
References (62)
Show source-provided references
1. Randall, R.B. (2011). Vibration-Based Condition Monitoring: Industrial, Aerospace and Automotive Applications, John Wiley & Sons.10.1002/9780470977668 ↗
2. Failure prediction by condition monitoring (part 1)
3. Mohanty, A.R. (2014). Machinery Condition Monitoring: Principles and Practices, CRC Press.10.1201/9781351228626 ↗
4. Artificial intelligence for fault diagnosis of rotating machinery: A review10.1016/j.ymssp.2018.02.016 ↗
5. Lei, Y., Yang, B., Jiang, X., Jia, F., Li, N., and Nandi, A.K. (2020). Applications of machine learning to machine fault diagnosis: A review and roadmap. Mech. Syst. Signal Process., 138.10.1016/j.ymssp.2019.106587 ↗
6. Rolling element bearing diagnostics-A tutorial10.1016/j.ymssp.2010.07.017 ↗
7. Model for the vibration produced by a single point defect in a rolling element bearing10.1016/0022-460X(84)90595-9 ↗
8. Vibration monitoring of rolling element bearings by the high-frequency resonance technique—A review10.1016/0301-679X(84)90076-8 ↗
9. The spectral kurtosis: A useful tool for characterising non-stationary signals10.1016/j.ymssp.2004.09.001 ↗
10. Antoni, J. (2004, January 6–10). The Spectral Kurtosis of nonstationary signals: Formalisation, some properties, and application. Proceedings of the 12th European Signal Processing Conference, Vienna, Austria.
11. Brusa, E., Bruzzone, F., Delprete, C., Di Maggio, L.G., and Rosso, C. (2020). Health indicators construction for damage level assessment in bearing diagnostics: A proposal of an energetic approach based on envelope analysis. Appl. Sci., 10.10.3390/app10228131 ↗
12. An enhanced Kurtogram method for fault diagnosis of rolling element bearings10.1016/j.ymssp.2012.10.003 ↗
13. Hebda-Sobkowicz, J., Zimroz, R., and Wyłomanska, A. (2020). Selection of the informative frequency band in a bearing fault diagnosis in the presence of non-gaussian noise-Comparison of recently developed methods. Appl. Sci., 10.10.3390/app10082657 ↗
14. A comparative experimental study on the use of acoustic emission and vibration analysis for bearing defect identification and estimation of defect size10.1016/j.ymssp.2004.10.013 ↗
15. The use of acoustic emission for estimation of bearing defect size10.1007/s11668-008-9119-8 ↗
16. Observations of changes in acoustic emission waveform for varying seeded defect sizes in a rolling element bearing10.1016/j.apacoust.2008.01.005 ↗
17. Support-vector networks10.1007/BF00994018 ↗
18. Support vector machine in machine condition monitoring and fault diagnosis10.1016/j.ymssp.2006.12.007 ↗
19. A fault diagnosis approach for roller bearing based on IMF envelope spectrum and SVM10.1016/j.measurement.2006.10.010 ↗
20. Rolling element bearings multi-fault classification based on the wavelet denoising and support vector machine10.1016/j.ymssp.2007.02.003 ↗
21. Deep learning and its applications to machine health monitoring10.1016/j.ymssp.2018.05.050 ↗
22. Imagenet classification with deep convolutional neural networks
23. Guo, X., Shen, C., and Chen, L. (2017). Deep fault recognizer: An integrated model to denoise and extract features for fault diagnosis in rotating machinery. Appl. Sci., 7.10.3390/app7010041 ↗
24. Fault diagnosis of rotary machinery components using a stacked denoising autoencoder-based health state identification10.1016/j.sigpro.2016.07.028 ↗
25. Guo, S., Yang, T., Gao, W., and Zhang, C. (2018). A novel fault diagnosis method for rotating machinery based on a convolutional neural network. Sensors, 18.10.3390/s18051429 ↗
26. Automated bearing fault diagnosis scheme using 2D representation of wavelet packet transform and deep convolutional neural network10.1016/j.compind.2019.01.008 ↗
27. A New Convolutional Neural Network-Based Data-Driven Fault Diagnosis Method10.1109/TIE.2017.2774777 ↗
28. Zhuang, Z., Lv, H., Xu, J., Huang, Z., and Qin, W. (2019). A deep learning method for bearing fault diagnosis through stacked residual dilated convolutions. Appl. Sci., 9.10.3390/app9091823 ↗
29. Brito, L.C., Susto, G.A., Brito, J.N., and Duarte, M.A.V. (2021). An Explainable Artificial Intelligence Approach for Unsupervised Fault Detection and Diagnosis in Rotating Machinery. arXiv.10.1016/j.ymssp.2021.108105 ↗
30. Rolling element bearing diagnostics using the Case Western Reserve University data: A benchmark study10.1016/j.ymssp.2015.04.021 ↗
31. Nectoux, P., Gouriveau, R., Medjaher, K., Ramasso, E., Chebel-morello, B., Zerhouni, N., and Varnier, C. (2012, January 18–21). PRONOSTIA: An experimental platform for bearings accelerated degradation tests. Proceedings of the IEEE International Conference on Prognostics and Health Management, Denver, CO, USA.
32. The Politecnico di Torino rolling bearing test rig: Description and analysis of open access data10.1016/j.ymssp.2018.10.010 ↗
33. Bearing data set. IMS, Univ. Cincinnati, NASA Ames Progn
34. A Survey on Transfer Learning10.1109/TKDE.2009.191 ↗
35. Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., and Li, F.-F. (2009, January 20–25). ImageNet: A large-scale hierarchical image database. Proceedings of the 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Miami, FL, USA.10.1109/CVPR.2009.5206848 ↗
36. Highly Accurate Machine Fault Diagnosis Using Deep Transfer Learning10.1109/TII.2018.2864759 ↗
37. Preprocessing-Free Gear Fault Diagnosis Using Small Datasets with Deep Convolutional Neural Network-Based Transfer Learning10.1109/ACCESS.2018.2837621 ↗
38. Transfer Learning with Neural Networks for Bearing Fault Diagnosis in Changing Working Conditions10.1109/ACCESS.2017.2720965 ↗
39. Hasan, M.J., and Kim, J.M. (2018). Bearing fault diagnosis under variable rotational speeds using Stockwell transform-based vibration imaging and transfer learning. Appl. Sci., 8.10.3390/app8122357 ↗
40. Li, Y., Liu, M., Drossos, K., and Virtanen, T. (2020, January 4–8). Sound event detection via dilated convolutional recurrent neural networks. Proceedings of the ICASSP 2020—2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Barcelona, Spain.10.1109/ICASSP40776.2020.9054433 ↗
41. Drossos, K., Mimilakis, S.I., Gharib, S., Li, Y., and Virtanen, T. (2020, January 19–24). Sound event detection with depthwise separable and dilated convolutions. Proceedings of the 2020 International Joint Conference on Neural Networks (IJCNN), Glasgow, UK.10.1109/IJCNN48605.2020.9207532 ↗
42. Hershey, S., Chaudhuri, S., Ellis, D.P.W., Gemmeke, J.F., Jansen, A., Moore, R.C., Plakal, M., Platt, D., Saurous, R.A., and Seybold, B. (2017, January 5–9). CNN architectures for large-scale audio classification. Proceedings of the 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), New Orleans, LA, USA.10.1109/ICASSP.2017.7952132 ↗
43. Sound Event Detection: A tutorial10.1109/MSP.2021.3090678 ↗
44. Anomalous Sound Detection Using Deep Audio Representation and a BLSTM Network for Audio Surveillance of Roads10.1109/ACCESS.2018.2872931 ↗
45. Two-source acoustic event detection and localization: Online implementation in a Smart-room
46. Ensemble of Convolutional Neural Networks for Weakly-Supervised Sound Event Detection Using Multiple Scale Input
47. Peng, Y.T., Lin, C.Y., Sun, M.T., and Tsai, K.C. (July, January 28). Healthcare audio event classification using hidden Markov models and hierarchical hidden Markov models. Proceedings of the 2009 IEEE International Conference on Multimedia and Expo, New York, NY, USA.
48. Gemmeke, J.F., Ellis, D.P.W., Freedman, D., Jansen, A., Lawrence, W., Moore, R.C., Plakal, M., and Ritter, M. (2017, January 5–9). Audio Set: An ontology and human-labeled dataset for audio events. Proceedings of the 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), New Orleans, LA, USA.10.1109/ICASSP.2017.7952261 ↗
49. Gururani, S., Summers, C., and Lerch, A. (2018, January 23–27). Instrument activity detection in polyphonic music using deep neural networks. Proceedings of the 19th International Society for Music Information Retrieval Conference, ISMIR 2018, Paris, France.
50. Handwritten digit recognition with a back-propagation network
51. Zhou, P., Zhou, G., Zhu, Z., Tang, C., He, Z., Li, W., and Jiang, F. (2018). Health monitoring for balancing tail ropes of a hoisting system using a convolutional neural network. Appl. Sci., 8.10.3390/app8081346 ↗
52. Yoo, Y., and Baek, J.G. (2018). A novel image feature for the remaining useful lifetime prediction of bearings based on continuous wavelet transform and convolutional neural network. Appl. Sci., 8.10.3390/app8071102 ↗
53. Kingma, D.P., and Ba, J.L. (2015, January 7–9). Adam: A method for stochastic optimization. Proceedings of the 3rd International Conference on Learning Representations, ICLR 2015—Conference Track Proceedings, San Diego, CA, USA.
54. Deep Convolutional Transfer Learning Network: A New Method for Intelligent Fault Diagnosis of Machines with Unlabeled Data10.1109/TIE.2018.2877090 ↗
55. (2021, May 07). YAMNet Mathworks. Available online: https://it.mathworks.com/help/audio/ref/yamnet.html.
56. (2021, May 07). YAMNet Tensorflow. Available online: https://www.tensorflow.org/hub/tutorials/yamnet.
57. (2021, May 07). YAMNet GitHub. Available online: https://github.com/tensorflow/models/tree/master/research/audioset/yamnet.
58. Howard, A.G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., and Adam, H. (2017). MobileNets: Efficient convolutional neural networks for mobile vision applications. arXiv.
59. Ioffe, S., and Christian, S. (2015, January 6–11). Batch normalization: Accelerating deep network training by reducing internal covariate shift. Proceedings of the International Conference on Machine Learning, Lille, France. PMLR.
60. A Scale for the Measurement of the Psychological Magnitude Pitch10.1121/1.1915893 ↗
61. Rabiner, L., and Schafer, R. (2010). Theory and Applications of Digital Speech Processing, Prentice Hall Press.
62. (2020, August 03). CWRU Bearing Data Center. Available online: https://engineering.case.edu/bearingdatacenter.