A New Fault Diagnosis Method for Unbalanced Data Based on 1DCNN and L2-SVM
Source abstract
Abstract
In general, the measured health condition data from rolling bearings usually exhibit imbalanced distribution. However, traditional intelligent fault diagnosis methods usually assume that the data categories are balanced. To improve the diagnosis accuracy of unbalanced datasets, a new fault diagnosis method for unbalanced data based on 1DCNN and L2-SVM is proposed in this paper. Firstly, to prevent the minority class samples from being heavily suppressed by the rectified linear unit (ReLU) activation function in the traditional convolutional neural network (CNN), ReLU is improved by linear and scaled exponential linear units (SELUs). Secondly, to solve the problem where the cross-entropy loss treats all input samples equally, it is replaced by the L2-support vector machine (L2-SVM) loss. Furthermore, a dynamic adjustment parameter is introduced to assign less misclassification cost to the majority of class samples. Finally, we add a new modulation factor that reduces the weight of more distinguishable samples to generate more focus on training indiscernible samples. The proposed method is carried out on two kinds of bearing datasets. The experimental results illustrate a significant improvement in recognition accuracy and the higher diagnosis performance of the model when dealing with unbalanced data compared with other intelligent methods.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Machine Fault Diagnosis Techniques · Engineering Diagnostics and Reliability · Evaluation and Optimization Models
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Mechanical and Electrical Engineering, Lanzhou University of Technology, Lanzhou 730050, China
School of Engineering, Xi’an International University, Xi’an 710077, China
School of Automation, Northwestern Polytechnical University, Xi’an 710129, China
Funding
National Natural Science Foundation of China · National Key Research and Development Plan · Hongliu First-class Disciplines Development Program of Lanzhou University of Technology
References (40)
Show source-provided references
1. A method for rapidly evaluating reliability and predicting remaining useful life using two-dimensional convolutional neural network with signal conversion10.1007/s12206-019-0504-x ↗
2. A review on signal processing techniques utilized in the fault diagnosis of rolling element bearings10.1016/j.triboint.2015.12.037 ↗
3. Support vector machine in machine condition monitoring and fault diagnosis10.1016/j.ymssp.2006.12.007 ↗
4. A review on empirical mode decomposition in fault diagnosis of rotating machinery10.1016/j.ymssp.2012.09.015 ↗
5. Recent advances in time-frequency analysis methods for machinery fault diagnosis: A review with application examples10.1016/j.ymssp.2013.01.017 ↗
6. A Systematic Review of Fuzzy Formalisms for Bearing Fault Diagnosis10.1109/TFUZZ.2018.2878200 ↗
7. Xu, G., Liu, M., Jiang, Z., Söffker, D., and Shen, W. (2019). Bearing Fault Diagnosis Method Based on Deep Convolutional Neural Network and Random Forest Ensemble Learning. Sensors, 19.10.3390/s19051088 ↗
8. Permutation entropy-based 2D feature extraction for bearing fault diagnosis10.1007/s11071-020-06014-6 ↗
9. Reducing the Dimensionality of Data with Neural Networks10.1126/science.1127647 ↗
10. Deep learning10.1038/nature14539 ↗
11. Prediction of Surface Tension of Various Aqueous Amine Solutions Using the UNIFAC Model and Artificial Neural Networks10.1021/acs.iecr.1c01048 ↗
12. Monitoring the effect of surface functionalization on the CO2 capture by graphene oxide/methyl diethanolamine nanofluids10.1016/j.jece.2021.106202 ↗
13. A Survey of the Usages of Deep Learning for Natural Language Processing10.1109/TNNLS.2020.2979670 ↗
14. Computer-vision-based research on friction vibration and coupling of frictional and torsional vibrations in water-lubricated bearing-shaft system10.1016/j.triboint.2020.106336 ↗
15. A Survey on the New Generation of Deep Learning in Image Processing10.1109/ACCESS.2019.2956508 ↗
16. Application of GMDH neural network technique to improve measuring precision of a simplified photon attenuation based two-phase flowmeter10.1016/j.flowmeasinst.2020.101804 ↗
17. Speech Emotion Recognition Using Deep Learning Techniques: A Review10.1109/ACCESS.2019.2936124 ↗
18. Hydrocarbon reservoirs characterization by co-interpretation of pressure and flow rate data of the multi-rate well testing10.1016/j.petrol.2015.08.016 ↗
19. A novel unsupervised learning method for intelligent fault diagnosis of rolling element bearings based on deep functional auto-encoder10.1007/s12206-020-1002-x ↗
20. Domain adaptation-based deep feature learning method with a mixture of distance measures for bearing fault diagnosis10.1088/1361-6501/abeddd ↗
21. A method for mechanical fault recognition with unseen classes via unsupervised convolutional adversarial auto-encoder10.1088/1361-6501/abb38c ↗
22. An architecture of deep learning network based on ensemble empirical mode decomposition in precise identification of bearing vibration signal10.1007/s12206-018-1205-6 ↗
23. An enhancement deep feature fusion method for rotating machinery fault diagnosis10.1016/j.knosys.2016.12.012 ↗
24. A convolutional neural network based feature learning and fault diagnosis method for the condition monitoring of gearbox10.1016/j.measurement.2017.07.017 ↗
25. A New Convolutional Neural Network-Based Data-Driven Fault Diagnosis Method10.1109/TIE.2017.2774777 ↗
26. Mukhopadhyay, R., Panigrahy, P.S., Misra, G., and Chattopadhyay, P. (2018, January 23–25). Quasi 1D CNN-based Fault Diagnosis of Induction Motor Drives. Proceedings of the 2018 5th International Conference on Electric Power and Energy Conversion Systems (EPECS), Kitakyushu, Japan.10.1109/EPECS.2018.8443552 ↗
27. Intelligent fault diagnosis method of planetary gearboxes based on convolution neural network and discrete wavelet transform10.1016/j.compind.2018.11.003 ↗
28. Rolling element bearing fault diagnosis using convolutional neural network and vibration image10.1016/j.cogsys.2018.03.002 ↗
29. Lessmeier, C., Kimotho, J., Zimmer, D., and Sextro, W. (2016, January 5–8). Condition Monitoring of Bearing Damage in Electromechanical Drive Systems by Using Motor Current Signals of Electric Motors: A Benchmark Data Set for Data-Driven Classification. Proceedings of the European Conference of the PHM Society 2016, Bilbao, Spain.10.36001/phme.2016.v3i1.1577 ↗
30. Real-Time Motor Fault Detection by 1-D Convolutional Neural Networks10.1109/TIE.2016.2582729 ↗
31. Bearing Fault Detection by One-Dimensional Convolutional Neural Networks10.1155/2017/8617315 ↗
32. Huang, S., Tang, J., Dai, J., Askar, M., and Gabbouj, M. (2019). Signal Status Recognition Based on 1DCNN and Its Feature Extraction Mechanism Analysis. Sensors, 19.10.3390/s19092018 ↗
33. Li, H., Huang, J., and Ji, S. (2019). Bearing Fault Diagnosis with a Feature Fusion Method Based on an Ensemble Convolutional Neural Network and Deep Neural Network. Sensors, 19.10.3390/s19092034 ↗
34. An Antinoise Fault Diagnosis Method Based on Multiscale 1DCNN
35. Multitask Learning Based on Lightweight 1DCNN for Fault Diagnosis of Wheelset Bearings10.1109/TIM.2021.3118090 ↗
36. 1DCNN Fault Diagnosis Based on Cubic Spline Interpolation Pooling
37. Deep normalized convolutional neural network for imbalanced fault classification of machinery and its understanding via visualization10.1016/j.ymssp.2018.03.025 ↗
38. Tang, Y. (2015). Deep Learning using Linear Support Vector Machines. arXiv.
39. Zhang, W., Peng, G., Li, C., Chen, Y., and Zhang, Z. (2017). A New Deep Learning Model for Fault Diagnosis with Good Anti-Noise and Domain Adaptation Ability on Raw Vibration Signals. Sensors, 17.10.20944/preprints201701.0132.v1 ↗
40. CWRU (2006). Case Western Reserve University Bearing Date Center Website, CWRU. Available online: https://engineering.case.edu/bearingdatacenter/download-data-file.