Abstract
Rotating machinery failure analysis requires signal preprocessing to extract fault-related information. However, to promote accurate condition monitoring of bearing following two conditions must be achieved (1) intelligent instance annotation and (2) automatic feature extraction and selection. Therefore, an attempt has been made in this paper using Gaussian Mixture Models (GMM) and t-distributed stochastic neighbor embedding (t-SNE) techniques to realize intelligent instance annotation. In addition to this, a one-dimensional convolutional neural network (1DCNN) is utilized for automatic feature extraction and selection. To validate the proposed method, experimentation is conducted on a high-speed rotor-supported bearing test rig to acquire run-to-failure roller bearing lifetime data. Bearing lifetime responses are segregated into different operating conditions, namely normal, slight degradation (SLD), severe degradation (SVD), and failure using t-SNE and GMM techniques. The comparisons are conducted with the existing advanced classifiers based on assessment metrics ROC (receiver operating characteristic) curve, AUC values, precision, recall, and F1-score. The classification accuracy value is obtained as 98.23% for the proposed method, which is highest compared to SVM, KNN, DNN, DBN, LSTM, and BiLSTM classifiers. The comparison results revealed the superior performance of the proposed method compared to existing advanced classifiers. Therefore, the proposed methodology precisely classifies the bearing lifetime data into different operating conditions using vibration acceleration responses.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig5_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig6_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig7_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig8_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig9_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig10_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig11_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig12_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig13_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig14_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig15_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig16_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11668-022-01469-8/MediaObjects/11668_2022_1469_Fig17_HTML.png)
Similar content being viewed by others
References
M. S. Rathore, & S. P. Harsha. Rolling bearing prognostic analysis for domain adaptation under different operating conditions. Eng. Failure Anal., 106414 (2022)
A. Kumar, R. Kumar, Enhancing weak defect features using undecimated and adaptive wavelet transform for estimation of roller defect size in a bearing. Tribol. Trans. 60(5), 794–806 (2017)
F.H. Cakir, A. Sert, O.N. Celik, N. Dereoğlu, Maintenance error detection procedure and a case study of failure analysis locomotive diesel engine bearings. J. Fail. Anal. Prev. 18(2), 356–363 (2018)
D. Petersen, C. Howard, Z. Prime, Varying stiffness and load distributions in defective ball bearings: analytical formulation and application to defect size estimation. J. Sound Vib. 337, 284–300 (2015)
P.K. Kankar, S.C. Sharma, S.P. Harsha, Fault diagnosis of ball bearings using machine learning methods. Expert Syst. Appl. 38(3), 1876–1886 (2011)
K. Kaplan, Y. Kaya, M. Kuncan, M.R. Minaz, H.M. Ertunç, An improved feature extraction method using texture analysis with LBP for bearing fault diagnosis. Appl. Soft Comput. 87, 106019 (2020)
D. Zhu, Y. Pan, W. Gao, Fault feature extraction of rolling element bearing under complex transmission path based on multiband signals cross-correlation spectrum. J. Failure Anal. Prevent. 22, 1–16 (2022)
M.S. Safizadeh, S.K. Latifi, Using multi-sensor data fusion for vibration fault diagnosis of rolling element bearings by accelerometer and load cell. Inform. Fusion. 18, 1–8 (2014)
J. Tian, C. Morillo, M.H. Azarian, M. Pecht, Motor bearing fault detection using spectral kurtosis-based feature extraction coupled with K-nearest neighbor distance analysis. IEEE Trans. Industr. Electron. 63(3), 1793–1803 (2015)
X. Yan, M. Jia, A novel optimized SVM classification algorithm with multi-domain feature and its application to fault diagnosis of rolling bearing. Neurocomputing. 313, 47–64 (2018)
I. Attoui, N. Fergani, N. Boutasseta, B. Oudjani, A. Deliou, A new time–frequency method for identification and classification of ball bearing faults. J. Sound Vib. 397, 241–265 (2017)
Y. Chang, H. Fang, A hybrid prognostic method for system degradation based on particle filter and relevance vector machine. Reliab. Eng. Syst. Saf. 186, 51–63 (2019)
M.S. Rathore, S.P. Harsha, Prognostic analysis of high-speed cylindrical roller bearing using weibull distribution and k-nearest neighbor. J. Nondestruct. Eval. Diagn. Progn. Eng. Syst. (2022). https://doi.org/10.1115/1.4051314
A. Sharma, M. Amarnath, P.K. Kankar, Life assessment and health monitoring of rolling element bearings: an experimental study. Life Cycle Reliab. Safety Eng. 7(2), 97–114 (2018)
M.S. Rathore, S.P. Harsha, Prognostics analysis of rolling bearing based on Bi-Directional LSTM and attention mechanism. J. Failure Anal. Prevent. (2022). https://doi.org/10.1007/s11668-022-01357-1
M. Tekkalmaz, Ü. Er, F.H. Çakir, F. Bozkurt, A new approach to monitor wear tracks propagation on-site with electromechanical impedance technique. J. Intell. Mater. Syst. Struct. 33(2), 342–351 (2022)
X. Li, W. Zhang, Q. Ding, Deep learning-based remaining useful life estimation of bearings using multi-scale feature extraction. Reliab. Eng. Syst. Saf. 182, 208–218 (2019)
Y. Lin, X. Li, Y. Hu, Deep diagnostics and prognostics: an integrated hierarchical learning framework in PHM applications. Appl. Soft Comput. 72, 555–564 (2018)
H. Li, W. Zhao, Y. Zhang, E. Zio, Remaining useful life prediction using multi-scale deep convolutional neural network. Appl. Soft Comput. 89, 106113 (2020)
T. Ince, S. Kiranyaz, L. Eren, M. Askar, M. Gabbouj, Real-time motor fault detection by 1-D convolutional neural networks. IEEE Trans. Industr. Electron. 63(11), 7067–7075 (2016)
S. Kiranyaz, T. Ince, M. Gabbouj, Real-time patient-specific ECG classification by 1-D convolutional neural networks. IEEE Trans. Biomed. Eng. 63(3), 664–675 (2015)
R. Zhang, Z. Peng, L. Wu, B. Yao, Y. Guan, Fault diagnosis from raw sensor data using deep neural networks considering temporal coherence. Sensors. 17(3), 549 (2017)
L. Eren, Bearing fault detection by one-dimensional convolutional neural networks. Math. Prob. Eng. 2017, 1–9 (2017)
S. Teng, G. Chen, Deep convolution neural network-based crack feature extraction, detection and quantification. J. Failure Anal. Prevent. 25, 1–14 (2022)
A. L. Maas, A. Y Hannun, A. Y. Ng,. Rectifier nonlinearities improve neural network acoustic models. In Proceedings Icml. Vol. 30, No. 1, p. 3 (2013)
S. Santurkar, D. Tsipras, A. Ilyas, A. Madry. How does batch normalization help optimization?. In Advances in Neural Information Processing Systems (pp. 2483–2493) (2018)
L.V.D. Maaten, G. Hinton, Visualizing data using t-SNE. J. Mach. Learn. Res. 9, 2579–2605 (2008)
M. Alabsi, Y. Liao, A.A. Nabulsi, Bearing fault diagnosis using deep learning techniques coupled with handcrafted feature extraction: a comparative study. J. Vib. Control. 27(3–4), 404–414 (2021)
M. Abbas, A. El-Zoghabi, A. Shoukry, DenMune: Density peak based clustering using mutual nearest neighbors. Pattern Recogn. 109, 107589 (2020)
E. Roman-Rangel, S. Marchand-Maillet, Inductive t-SNE via deep learning to visualize multi-label images. Eng. Appl. Artif. Intell. 81, 336–345 (2019)
B.M. Devassy, S. George, Dimensionality reduction and visualization of hyperspectral ink data Using t-SNE. Forensic Sci Int. 311, 110194 (2020)
C.C. Hsu, W.H. Huang, Integrated dimensionality reduction technique for mixed-type data involving categorical values. Appl. Soft Comput. 43, 199–209 (2016)
Z. Ju, H. Liu, Fuzzy gaussian mixture models. Pattern Recogn. 45(3), 1146–1158 (2012)
M. Karami, L. Wang, Fault detection and diagnosis for nonlinear systems: a new adaptive Gaussian mixture modeling approach. Energy Build. 166, 477–488 (2018)
M.S. Yang, C.Y. Lai, C.Y. Lin, A robust EM clustering algorithm for Gaussian mixture models. Pattern Recogn. 45(11), 3950–3961 (2012)
T.A. Harris, M.N. Kotzalas, Fatigue life: basic theory and rating standards. Rolling Bear. Anal. 2007, 195–252 (2007)
E. V. Zaretsky. Rolling bearing life prediction, theory, and application (No. NASA/TP-2013–215305) (2013)
P. Nectoux, R. Gouriveau, K. Medjaher, E. Ramasso, B. Chebel-Morello, N. Zerhouni, C. Varnier. PRONOSTIA: An experimental platform for bearings accelerated degradation tests. In IEEE International Conference on Prognostics and Health Management, PHM'12. (pp. 1–8). IEEE Catalog Number: CPF12PHM-CDR. (2012)
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
We declare that we do not have any potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendices
Appendix I
See Table 6.
Appendix II
See Table 7.
Rights and permissions
About this article
Cite this article
Rathore, M.S., Harsha, S.P. Roller Bearing Failure Analysis using Gaussian Mixture Models and Convolutional Neural Networks. J Fail. Anal. and Preven. 22, 1853–1871 (2022). https://doi.org/10.1007/s11668-022-01469-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11668-022-01469-8