RESEARCH PAPER
Research on a Lightweight Multi-Scale Feature Fusion and its Fault Diagnosis Method for Rolling Bearing with Limited Labeled Samples
1
School of Mechanical Engineering and Automation, Northeastern University, China
2
Key Laboratory of Vibration and Control of Aero-Propulsion System, Ministry of Education, Northeastern University, Shenyang 110819, China
3
Liaoning Province Key Laboratory of Multidisciplinary Design Optimization of Complex Equipment, Northeastern University, Shenyang 110819,
China
4
College of Mechanical and Electrical Engineering, Northeast Forestry University, China
Submission date: 2024-05-12
Final revision date: 2024-07-08
Acceptance date: 2024-08-11
Online publication date: 2024-08-29
Publication date: 2024-08-29
Corresponding author
Xiangwei Kong
School of Mechanical Engineering and Automation, Northeastern University, NO. 3-11, Wenhua Road, Heping District, Shenyang, 110819, Shenyang, China
School of Mechanical Engineering and Automation, Northeastern University, NO. 3-11, Wenhua Road, Heping District, Shenyang, 110819, Shenyang, China
Eksploatacja i Niezawodność – Maintenance and Reliability 2025;27(1):192235
HIGHLIGHTS
- The proposed method aims to realize fault diagnosis on limited labeled samples.
- A multi-scale depth-separable convolutional neural network is proposed.
- An improved feature soft threshold denoising module is introduced.
- The framework is simpler and clearer, with high robustness and generalization ability.
- The proposed method is more suitable for complex practical engineering scenarios.
KEYWORDS
Rolling bearingMulti-scale feature fusionLimited labeled samplesDepth-separable convolution networkSoft threshold
TOPICS
ABSTRACT
Convolutional neural networks(CNNs) show significant potential for bearing fault diagnosis. However, tra-ditional CNNs face challenges such as poor noise resistance, high computational complexity, reliance on extensive samples, and limited generalizability. As a result, this paper proposes WDSC-Net, a lightweight, multiscale feature fusion method, focusing on limited labeled fault samples. Initially, a wide kernel convo-lutional is employed, aiming to reduce parameters and computational complexity. Next, features are fed into a 1×1 convolutional layer reduces feature dimensionality. Subsequently, leveraging the benefits of depth-separable convolution (DSC) allows the separation of spatial and channel features, constructing four convolutional layers of varying scales to amplify the nonlinear fault representation. Finally, an improved feature soft-threshold denoising module is introduced for global feature denoising. Validation on CWRU and MCDS datasets shows that the WDSC-Net method exhibits superior generalizability and noise resistance compared to typical deep-learning fault methods
ACKNOWLEDGEMENTS
This work was partly supported by the National Key Research and Development Program of China under Grant 2019YFB1704500, partly by the State Ministry of Science and Technology Innovation Fund of China under Grant 2018IM030200, the National Natural Foundation of China under Grant U1708255, and in part by the National Science and Technology Major Project under Grant J2019-V-0009-0103.
REFERENCES (49)
1.
Zhou H, Huang X, Wen G et al. Construction of health indicators for condition monitoring of rotating machinery: A review of the research. Expert systems with applications 2022; 203 117297, https://doi.org/10.1016/j.eswa....
2.
Li Z, Chen J, Zi Y et al. Independence-oriented VMD to identify fault feature for wheel set bearing fault diagnosis of high speed locomotive. Mechanical systems and signal processing 2017; 85 512-529, https://doi.org/10.1016/j.ymss....
3.
Zhang J, Kong X, Cheng L et al. Intelligent fault diagnosis of rolling bearings based on continuous wavelet transform-multiscale feature fusion and improved channel attention mechanism. Eksploatacja i Niezawodność, 2023, 25(1), http://doi.org/10.17531/ein.20....
4.
Kapoor R, Gupta R, Jha S et al. Boosting performance of power quality event identification with KL Divergence measure and standard deviation. Measurement 2018; 126 134-142, https://doi.org/10.1016/j.meas....
5.
Zhang Y, Xing K, Bai R et al. An enhanced convolutional neural network for bearing fault diagnosis based on time–frequency image. Measurement 2020; 157 107667, https://doi.org/10.1016/j.meas....
6.
Zhao R, Yan R, Chen Z et al. Deep learning and its applications to machine health monitoring. Mechanical Systems and Signal Processing, 2019, 115: 213-237, https://doi.org/10.1016/j.ymss....
7.
Gong W, Wang Y, Zhang M et al. A fast anomaly diagnosis approach based on modified CNN and multisensor data fusion. IEEE Transactions on Industrial Electronics 2021; 69(12) 13636-13646, https://doi.org/10.1109/TIE.20....
8.
Xu G, Liu M, Jiang Z et al. Bearing fault diagnosis method based on deep convolutional neural network and random forest ensemble learning. Sensors 2019; 19(5) 1088, https://doi.org/10.3390/s19051....
9.
Fuan W, Hongkai J, Haidong S et al. An adaptive deep convolutional neural network for rolling bearing fault diagnosis. Measurement Science and Technology 2017; 28(9) 095005, https://doi.org/10.1088/1361-6....
10.
Yang J, Zhang Y, Zhu Y. Intelligent fault diagnosis of rolling element bearing based on SVMs and fractal dimension. Mechanical Systems and Signal Processing 2007; 21(5) 2012-2024, https://doi.org/10.1016/j.ymss....
11.
Ye Z, Yu J. Deep morphological convolutional network for feature learning of vibration signals and its applications to gearbox fault diagnosis. Mechanical Systems and Signal Processing 2021; 161 107984, https://doi.org/10.1016/j.ymss....
12.
Ke H, Chen D, Li X et al. Towards brain big data classification: Epileptic EEG identification with a lightweight VGGNet on global MIC. IEEE Access 2018; 6 14722-14733, https://doi.org/10.1109/ACCESS....
13.
Lu S, Lu Z, Zhang Y D. Pathological brain detection based on AlexNet and transfer learning. Journal of computational science 2019; 30 41-47, https://doi.org/10.1016/j.jocs....
14.
Wu Z, Shen C, Van Den Hengel A. Wider or deeper: Revisiting the resnet model for visual recognition. Pattern recognition 2019; 90 119-133, https://doi.org/10.1016/j.patc....
15.
Fang H, Deng J, Zhao B et al. LEFE-net: A lightweight efficient feature extraction network with strong robustness for bearing fault diagnosis. IEEE Transactions on Instrumentation and Measurement 2021; 70: 1–11, https://doi.org/10.1109/TIM.20....
16.
Deng J, Jiang W, Zhang Y et al. HS-KDNet: A lightweight network based on hierarchical-split block and knowledge distillation for fault diagnosis with extremely imbalanced data. IEEE Transactions on Instrumentation and Measurement 2021; 70: 1–9, https://doi.org/10.1109/TIM.20....
17.
Lu R, Liu S, Gong Z et al. Lightweight Knowledge Distillation-Based Transfer Learning Framework for Rolling Bearing Fault Diagnosis. Sensors 2024; 24(6): 1758, https://doi.org/10.3390/s24061....
18.
Xiong S, Zhou H, He S et al. Fault diagnosis of a rolling bearing based on the wavelet packet transform and a deep residual network with lightweight multi-branch structure. Measurement Science and Technology 2021; 32: 085106, https://doi.org/10.1088/1361-6....
19.
Liu W, Guo P, Ye L. A low-delay lightweight recurrent neural network (LLRNN) for rotating machinery fault diagnosis. Sensors 2019; 19: 3109, https://doi.org/10.3390/s19143....
20.
Cui J, Zhong Q, Zheng S et al. A lightweight model for bearing fault diagnosis based on gramian angular field and coordinate attention. Machines 2022; 10: 282, https://doi.org/10.3390/machin....
21.
Qin Z, Li Z, Zhang Z, et al. ThunderNet: Towards real-time generic object detection on mobile devices. Proceedings of the IEEE/CVF international conference on computer vision. 2019; 6718-6727. https://doi.org/10.1109/ICCV.2....
22.
Iandola FN, Han S, Moskewicz MW et al. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size. arXiv 2016; https://doi.org/10.48550/arXiv....
23.
Li Z Q. Adaptive Anti-noise Fault Diagnosis Algorithm for Bearings based on An Improved Convolution Neural Network. Machinery Design & Manufacture 2024;1-7. https://doi.org/10.19356/j.cnk....
24.
Dong R, Xu Y W, Long Z H et al. Fault Diagnosis Model of Noise-resistant Bearings Using Large Kernel Attention Mechanism. Noise and Vibration Control, 2023; 43(02): 162-168, https://doi.org/10.3969/j.issn....
25.
Li R, Wu J, Li Y et al. Periodnet: Noise-robust fault diagnosis method under varying speed conditions. IEEE Transactions on Neural Networks and Learning Systems 2023; https://doi.org/ 10.1109/TNNLS.2023.3274290.
26.
Fan W, Chen Z, Li Y et al. A reinforced noise resistant correlation method for bearing condition monitoring. IEEE Transactions on Automation Science and Engineering 2022; 20(2): 995-1006, https://doi.org/10.1109/TASE.2....
27.
Dong J, Jiang H, Su D et al. Transfer learning rolling bearing fault diagnosis model based on deep feature decomposition and class-level alignment. Measurement Science and Technology, 2024; 35(4): 046006, https://doi.org/10.1088/1361-6....
28.
Yang G, Liu L, Xi C. Bearing fault diagnosis based on sa-acgan data generation model. China Mechanical Engineering 2022; 33(13) 1613, https://doi.org/10.3969/j.issn....
29.
Meng A, Chen S, Ou Z et al. A novel few-shot learning approach for wind power prediction applying secondary evolutionary generative adversarial network. Energy 2022; 261: 125276, https://doi.org/10.1016/j.ener....
30.
Lei Y, Yang B, Jiang X et al. Applications of machine learning to machine fault diagnosis: A review and roadmap. Mechanical Systems and Signal Processing 2020; 138 106587, https://doi.org/10.1016/j.ymss....
31.
Meng X, Hu T, Li J et al. A digital twin library of mechanical transmission system for the application of small sample fault diagnosis problem. Measurement Science and Technology 2024; 35(6): 066125, https://doi.org/10.1088/1361-6....
32.
Yang J, Liu J, Xie J et al. Conditional GAN and 2-D CNN for bearing fault diagnosis with small samples. IEEE Transactions on Instrumentation and Measurement 2021; 70: 1-12, https://doi.org/10.1109/TIM.20....
33.
Zhou X K, Yu J B. Gearbox Fault Diagnosis Based on One-dimension Residual Convolutional Auto-encoder. Journal of Mechanical Engineering 2020; 56(7), 96-108. (in Chinese). https://doi.org/10.3901/JME.20....
34.
Wang H, Liu Z, Peng D et al. Understanding and learning discriminant features based on multiattention 1DCNN for wheelset bearing fault diagnosis. IEEE Transactions on Industrial Informatics 2019; 16(9) 5735-5745, https://doi.org/10.1109/TII.20....
35.
He Z, Yang L, Angizi S et al. Sparse BD-Net: A multiplication-less DNN with sparse binarized depth-wise separable convolution. ACM Journal on Emerging Technologies in Computing Systems (JETC) 2020; 16(2) 1-24, https://doi.org/10.1145/336939....
36.
Zhang J, Kong X, Cheng L et al. Intelligent fault diagnosis of rolling bearings based on continuous wavelet transform-multiscale feature fusion and improved channel attention mechanism. Eksploatacja i Niezawodność 2023; 25(1), http://doi.org/10.17531/ein.20....
37.
Zhang J, Xiangwei K, Xueyi L I et al. Fault diagnosis of bearings based on deep separable convolutional neural network and spatial dropout. Chinese Journal of Aeronautics 2022; 35(10) 301-312, https://doi.org/10.1016/j.cja.....
38.
Xu X, Wang J, Zhong B et al. Deep learning-based tool wear prediction and its application for machining process using multi-scale feature fusion and channel attention mechanism. Measurement 2021; 177 109254, https://doi.org/10.1016/j.meas....
39.
Sun Y, Weng Y, Luo B et al. Gesture recognition algorithm based on multi-scale feature fusion in RGB-D images. IET Image Processing 2023; 17(4) 1280-1290, https://doi.org/10.1049/ipr2.1....
40.
Guo J, Si Z, Xiang J. A compound fault diagnosis method of rolling bearing based on wavelet scattering transform and improved soft threshold denoising algorithm. Measurement 2022; 196: 111276, https://doi.org/10.1016/j.meas....
41.
Zhao M, Zhong S, Fu X et al. Deep residual shrinkage networks for fault diagnosis. IEEE Transactions on Industrial Informatics 2019; 16(7) 4681-4690, https://doi.org/ 10.1109/TII.2019.2943898.
43.
Li X, Zhang W. Deep learning-based partial domain adaptation method on intelligent machinery fault diagnostics. IEEE Transactions on Industrial Electronics 2020; 68(5) 4351-4361; https://doi.org/10.1109/TIE.20....
44.
Sun L, Zhu X, Xiao J et al. A hybrid fault diagnosis method for rolling bearings based on GGRU-1DCNN with AdaBN algorithm under multiple load conditions. Measurement Science and Technology 2024; 35(7): 076201, https://doi.org/10.1088/1361-6....
45.
Zhang W, Peng G, Li C et al. A new deep learning model for fault diagnosis with good anti-noise and domain adaptation ability on raw vibration signals. Sensors 2017; 17(2) 425, https://doi.org/10.3390/s17020....
46.
Zhu Y, Li G, Wang R et al. Intelligent fault diagnosis of hydraulic piston pump combining improved LeNet-5 and PSO hyperparameter optimization. Applied Acoustics 2021; 183: 108336, https://doi.org/10.1016/j.apac....
47.
Xu H, Wamg Z. Condition Evaluation and Fault Diagnosis of Power Transformer Based on GAN-CNN. Journal of Electrotechnology, Electrical Engineering and Management 2023; 6(3) 8-16, https://doi.org/10.23977/jeeem....
48.
Hu Z, Han T, Bian J et al. A deep feature extraction approach for bearing fault diagnosis based on multi-scale convolutional autoencoder and generative adversarial networks. Measurement Science and Technology 2022; 33(6) 065013, https://doi.org/10.1088/1361-6....
49.
Wu X, Zhang Y, Cheng C et al. A hybrid classification autoencoder for semi-supervised fault diagnosis in rotating machinery. Mechanical Systems and Signal Processing 2021; 149 107327, https://doi.org/10.1016/j.ymss....
CITATIONS (1):
1.
Localization of Conveyor Belt Damages Using a Deep Neural Network and a Hybrid Method for 1D Sequential Data Augmentation
Iwona Komorska, Andrzej Puchalski, Damian Bzinkowski
Applied Sciences
Iwona Komorska, Andrzej Puchalski, Damian Bzinkowski
Applied Sciences
| eISSN: | 2956-3860 |
| ISSN: | 1507-2711 |
We process personal data collected when visiting the website. The function of obtaining information about users and their behavior is carried out by voluntarily entered information in forms and saving cookies in end devices. Data, including cookies, are used to provide services, improve the user experience and to analyze the traffic in accordance with the Privacy policy. Data are also collected and processed by Google Analytics tool (more).
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
