Abstract
This paper proposes a novel unsupervised classification method to solve the problem of mechanical fault diagnosis under the condition of unknown number of fault types. The proposed method combining the three data processing stage. First, the deep encoding neural networks is used to complete the abstract signal feature representation under the unsupervised conditions. Second, the feature dimensionality reduction technique based on manifold learning is used to complete the low-dimensional mapping of the feature space. Third, the spatial clustering based on density criterion is introduced to classify the different fault samples. This paper uses two fault signals dataset to complete the performance verification experiment. The experimental results show that the DMDUC method respectively achieves the classification accuracy of 99.7 % and 100 % on the two datasets.
Similar content being viewed by others
Abbreviations
- SR :
-
Sparse regularization metric
- KLD :
-
Kullback-Leiber divergence
- DENN :
-
Deep encoding neural networks
- ML :
-
Manifold learning
- MFES :
-
Multi-functional fault experiment stand
- HF :
-
Handcrafted features
References
X. Chen, X. Qi and Z. Wang, Fault diagnosis of rolling bearing using marine predators algorithm-based support vector machine and topology learning and out-of-sample embedding, Measurement, 176 (2021) 109116, https://doi.org/10.1016/J.MEASUREMENT.2021.109116.
G. Mao, Y. Li and S. Jia, Interactive dual adversarial neural network framework: An open-set domain adaptation intelligent fault diagnosis method of rotating machinery, Measurement, 195 (2022) 111125, https://doi.org/10.1016/J.MEASUREMENT.2022.111125.
X. Li, W. Zhang and H. Ma, Deep learning-based adversarial multi-classifier optimization for cross-domain machinery fault diagnostics, Journal of Manufacturing Systems, 55 (2020) 334–347, https://doi.org/10.1016/j.jmsy.2020.04.017.
M. Li, D. Zou and S. Luo, A new generative adversarial network based imbalance fault diagnosis method, Measurement, 194 (2022) 111045, https://doi.org/10.1016/J.MEASUREMENT.2022.111045.
F. Jia, Y. Lei and N. Lu, Deep normalized convolutional neural network for imbalance fault classification of machinery and its understanding via visualization, Mechanical Systems and Signal Processing, 110 (2018) 349–367, https://doi.org/10.1016/j.ymssp.2018.03.025.
S. Xiong, H. Zhou and S. He, A novel end-to-end fault diagnosis approach for rolling bearings by integrating wavelet packet transform into convolutional neural network structures, Sensors, 20(17) (2020) 4965, https://doi.org/10.3390/s20174965.
Y. Liu, H. Jiang and Y. Wang, A conditional variational autoen-coding generative adversarial networks with self-modulation for rolling bearing fault diagnosis, Measurement, 192 (2002) 110888, https://doi.org/10.1016/J.MEASUREMENT.2022.110888.
K. Liu, Y. Feng and X. Xue, Fault diagnosis of hydraulic retraction system based on multi-source signals feature fusion and health assessment for the actuator, Journal of Intelligent and Fuzzy Systems, 34(6) (2018) 3635–3649, https://doi.org/10.3233/JIFS-169539.
J. R. Huerta-Rosales et al., Short-circuited turn fault diagnosis in transformers by using vibration signals, statistical time features, and support vector machines on FPGA, Sensors, 21(11) (2021) 3598, https://doi.org/10.3390/S21113598.
A. Ding, Y. Zhang and L. Zhu, Recognition method research on rough handling of express parcels based on acceleration features and CNN, Measurement, 163 (2020) 107942, https://doi.org/10.1016/j.measurement.2020.107942.
Y. Li, J. Wang and L. Duan, Association rule-based feature mining for automated fault diagnosis of rolling bearing, Shock and Vibration, 2019 (2019) 1518246, https://doi.org/10.1155/2019/1518246.
Y. Chang, J. Chen and C. Qu, Intelligent fault diagnosis of wind turbines via a deep learning network using parallel convolution layers with multi-scale kernels, Renewable Energy, 153 (2020) 205–213, https://doi.org/10.1016/jrenene.2020.02.004.
B. L. Costa, S. G. Antonio and B. J. Nei, Fault detection of bearing: An unsupervised machine learning approach exploiting feature extraction and dimensionality reduction, Informatics, 8(4) (2021) 85, https://doi.org/10.3390/INFORMATICS8040085.
C. Cheng, W. Wang and H. Liu, Intelligent fault diagnosis using an unsupervised sparse feature learning method, Measurement Science and Technology, 31(9) (2020) 095903, https://doi.org/10.1088/1361-6501/ab8c0e.
M. Sun, H. Wang and P. Liu, A sparse stacked denoising autoencoder with optimized transfer learning applied to the fault diagnosis of rolling bearings, Measurement, 146 (2019) 305–314, https://doi.org/10.1016/j.measurement.2019.06.029.
X. Zhao and M. Jia, A novel unsupervised deep learning network for intelligent fault diagnosis of rotating machinery, Structural Health Monitoring, 19(6) (2020) 1745–1763, https://doi.org/10.1177/1475921719897317.
X. Li, X. Li and H. Ma, Deep representation clustering-based fault diagnosis method with unsupervised data applied to rotating machinery, Mechanical Systems and Signal Processing, 143 (2020) 106825, https://doi.org/10.1016/j.ymssp.2020.106825.
H. Aljemely Anas, X. Jianping and K. J. Jawad Farqad, A novel unsupervised learning method for intelligent fault diagnosis of rolling element bearings based on deep functional auto-encoder, Journal of Mechanical Science and Technology, 34 (2020) 4367–4381, https://doi.org/10.1007/s12206-020-1002-x.
Z. Zhang, S. Li and J. Lu, A novel unsupervised learning method based on cross-normalization for machinery fault diagnosis, IEEE Access, 8 (2020) 92407–92417, https://doi.org/10.1109/access.2020.2992201.
Z. Zhang, S. Li and J. Wang, General normalized sparse filtering: A novel unsupervised learning method for rotating machinery fault diagnosis, Mechanical Systems and Signal Processing, 124 (2019) 596–612, https://doi.org/10.1016/j.ymssp.2019.02.006.
Y. Chang, J. Chen and C. Qu, Intelligent fault diagnosis of wind turbines via a deep learning network using parallel convolution layers with multi-scale kernels, Renewable Energy, 153 (2020) 205–213, https://doi.org/10.1016/j.renene.2020.02.004.
Y. Zhang, L. He and G. Cheng, A vibrational signal fault diagnosis rule extraction method based on DST-ACI discriminant criterion, Shock and Vibration, 2021 (2021) 8085421, https://doi.org/10.1155/2021/8085421.
A. Ding, Y. Zhang and L. Zhu, Recognition method research on rough handling of express parcels based on acceleration features and CNN, Measurement, 163 (2020) 107942, https://doi.org/10.1016/j.measurement.
Y. Li, J. Wang and L. Duan, Association rule-based feature mining for automated fault diagnosis of rolling bearing, Shock and Vibration, 2019 (2019) 1518246, https://doi.org/10.1155/2019/1518246.
Z. Zhang, H. Chen and S. Li, Unsupervised domain adaptation via enhanced transfer joint matching for bearing fault diagnosis, Measurement, 165 (2020) 108071, https://doi.org/10.1016/j.measurement.2020.108071.
D. Xiao, C. Qin and H. Yu, Unsupervised machine fault diagnosis for noisy domain adaptation using marginal denoising autoencoder based on acoustic signals, Measurement, 176 (2021) 109186, https://doi.org/10.1016/J.MEASUREMENT.2021.109186.
N. Lu, H. Xiao and Y. Sun, A new method for intelligent fault diagnosis of machines based on unsupervised domain adaptation, Neurocomputing, 427 (2020) 96–109, https://doi.org/10.1016/j.neucom.2020.10.039.
Y. Shi, A. Deng and X. Ding, Multisource domain factorization network for cross-domain fault diagnosis of rotating machinery: An unsupervised multisource domain adaptation method, Mechanical Systems and Signal Processing, 164 (2022) 108219, https://doi.org/10.1016/J.YMSSP.2021.108219.
P. Dai, J. Luo and K. Zhao, Stacked denoising autoencoder for missing traffic data reconstruction via mobile edge computing, Neural Computing and Applications, 35 (2023) 14259–14274, https://doi.org/10.1088/2632-2153/ACE81A.
Q. Wang, K. Qin and B. Lu, Time-feature attention-based convolutional auto-encoder for flight feature extraction, Scientific Reports, 13 (2023) 14175, https://doi.org/10.1038/S41598-023-41295-Y.
Y. Tang, F. Xu and L. Xu, Remain useful life forecasting for roller bearings using sparse auto-encoder, Management System Engineering, 2 (2023) 8, https://doi.org/10.1007/S44176-023-00019-2.
J. Rydzewski, M. Chen and O. Valsson, Manifold learning in atomistic simulations: a conceptual review, Machine Learning: Science and Technology, 4 (2023) 031001, https://doi.org/10.1007/S00521-023-08475-3.
J. Shi and H. Zhao, FS-MGKC: Feature selection based on structural manifold learning with multi-granularity knowledge coordination, Information Sciences, 648 (2023) 119555, https://doi.org/10.1016/J.INS.2023.119555.
Z. Masoumeh, A. Najmeh and K. Ali, Supervised feature selection on gene expression microarray datasets using manifold learning, Chemometrics and Intelligent Laboratory Systems, 237 (2023) 104828, https://doi.org/10.1016/J.CHEMOLAB.2023.104828.
Y. Sun, S. Chun and Y. Lee, Learned semantic index structure using knowledge graph embedding and density-based spatial clustering techniques, Applied Sciences, 12(13) (2022) 6713, https://doi.org/10.3390/APP12136713.
J. Zhao and Y. Li, Binary multi-view sparse subspace clustering, Neural Computing and Applications, 35 (2023) 21751–21770, https://doi.org/10.1007/S00521-023-08915-0.
M. Fang, W. Gao and Z. Feng, Deep robust multi-channel learning subspace clustering networks, Image and Vision Computing, 137 (2023) 104769, https://doi.org/10.1016/J.IMAVIS.2023.104769.
Acknowledgments
This work is supported by The National Natural Science Foundation of China (Project No. 11904407), China.
Author information
Authors and Affiliations
Corresponding author
Additional information
Yalun Zhang is a lecturer of Combat Command Department, Naval Command College, Nanjing, China. He received his Ph.D. in Naval University of Engineering. His research interests include fault diagnosis, signal analysis, machine learning.
Rongwu Xu is a Professor of Laboratory of Vibration and Noise, Naval University of Engineering, Wuhan, China. He received his Ph.D. in Naval University of Engineering. His research interests include vibration & noise control, acoustic theory, signal analysis.
Guo Cheng is a Professor of Laboratory of Vibration and Noise, Naval University of Engineering, Wuhan, China. He received his Ph.D. in Naval University of Engineering. His research interests include vibration & noise control, acoustic theory, signal analysis.
Xiufeng Huang is an ABD of Laboratory of Vibration and Noise, Naval Uni-versity of Engineering, Wuhan, China. He received his Ph.D. in Naval University of Engineering. Her research interests include vibration & noise control, acoustic theory.
Wenjing Yu is a Doctor of Laboratory of Vibration and Noise, Naval University of Engineering, Wuhan, China. She received his Ph.D. in Naval University of Engineering. Her research interests include vibration & noise control, signal analysis, fault diagnosis.
Rights and permissions
About this article
Cite this article
Zhang, Y., Xu, R., Cheng, G. et al. An unsupervised mechanical fault classification method under the condition of unknown number of fault types. J Mech Sci Technol 38, 605–622 (2024). https://doi.org/10.1007/s12206-024-0109-x
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12206-024-0109-x