Rolling bearing fault diagnosis under complex operating conditions forms the essential foundation for the predictive maintenance of rotating machinery. However, traditional methods are often overwhelmed by strong noise, and constrained by the empirical risk minimization (ERM) principle, leading to significant overfitting in small sample learning scenarios. To address the aforementioned limitations, a lightweight diagnostic model integrating S-Transform (ST), convolutional neural network (CNN), and support vector machine (SVM) is proposed in this paper. Time-frequency features are extracted by leveraging the multi-resolution characteristics of the ST, deep feature mapping is performed through a customized CNN, and SVM is introduced to construct the maximum-margin classification hyperplane based on the structural risk minimization (SRM) principle. The experimental results illustrate that the method exhibits exceptional diagnostic accuracy under intense noise and small sample sizes. Randomized subset cross-validation confirms that this architecture effectively eliminates the interference of sampling randomness. Consequently, the ST-CNN-SVM model demonstrates high statistical stability.
NiQJiJCFengK. Data-driven prognostic scheme for bearings based on a novel health indicator and gated recurrent unit network. IEEE Trans Ind Inf2023; 19: 1301–1311. https://doi.org/10.1109/tii.2022.3169465
2.
LiYWangSXieJ, et al.A lightweight dual-compression fault diagnosis framework for high-speed train bogie bearing. IEEE Trans Instrum Meas2024; 73: 1–14. https://doi.org/10.1109/tim.2024.3453318
3.
ChenXYangRXueY, et al.Deep transfer learning for bearing fault diagnosis: a systematic review since 2016. IEEE Trans Instrum Meas2023; 72: 1–21. https://doi.org/10.1109/tim.2023.3244237
LiHWuXLiuT, et al.Rotating machinery fault diagnosis based on typical resonance demodulation methods: a review. IEEE Sens J2023; 23: 6439–6459. https://doi.org/10.1109/jsen.2023.3235585
6.
SinghMShaikAG. Incipient fault detection in stator windings of an induction motor using stockwell transform and SVM. IEEE Trans Instrum Meas2020; 69: 9496–9504. https://doi.org/10.1109/tim.2020.3002444
7.
GaoSShiSZhangY. Rolling bearing compound fault diagnosis based on parameter optimization MCKD and convolutional neural network. IEEE Trans Instrum Meas2022; 71: 1–8. https://doi.org/10.1109/tim.2022.3158379
8.
AbidFBSallemMBrahamA. Optimized SWPT and decision tree for incipient bearing fault diagnosis. In: 2019 19th international conference on sciences and techniques of automatic control and computer engineering (STA), Sousse, Tunisia. New York: IEEE, 2019, pp. 231–236.
9.
XieTHuangXChoiSK. Intelligent mechanical fault diagnosis using multisensor fusion and convolution neural network. IEEE Trans Ind Inf2022; 18: 3213–3223. https://doi.org/10.1109/tii.2021.3102017
10.
XuYYanXSunB, et al.Deep coupled visual perceptual networks for motor fault diagnosis under nonstationary conditions. IEEE/ASME Trans Mechatron2022; 27: 4840–4850. https://doi.org/10.1109/tmech.2022.3166839
11.
LiYYangHWuK, et al.A Gramian angular field for constructing graph-based GNNs and its applications in rolling bearing defect detection. IEEE Sens J2024; 24: 35141–35155. https://doi.org/10.1109/jsen.2024.3458409
12.
WangGLiuDCuiL. Auto-embedding transformer for interpretable few-shot fault diagnosis of rolling bearings. IEEE Trans Reliab2024; 73: 1270–1279. https://doi.org/10.1109/tr.2023.3328597
13.
ZhaoJChenSLiuG, et al.Few-shot fault diagnosis based on model-agnostic meta-learning: a dual-channel approach combined with enhanced matrix modules. In: 2025 5th international conference on mechanical, electronics and electrical and automation control (METMS). IEEE, 2025, pp. 402–406.
14.
HuYLiuRLiX, et al.Task-sequencing Meta learning for intelligent few-shot fault diagnosis with limited data. IEEE Trans Ind Inf2022; 18: 3894–3904. https://doi.org/10.1109/tii.2021.3112504
15.
QiaoZNingSGaiY, et al.A digital twin guided physical-virtual denoising method for early fault detection of rolling element bearings. Mech Syst Signal Process2026; 249: 114108. https://doi.org/10.1016/j.ymssp.2026.114108
16.
FengKXuYWangY, et al.Digital twin enabled domain adversarial graph networks for bearing fault diagnosis. IEEE Trans Ind Cyber-Phys Syst2023; 1: 113–122. https://doi.org/10.1109/ticps.2023.3298879
17.
XuYHeDSunH, et al.Self-supervised learning for train bearing fault diagnosis based on time–frequency dual domain prediction. Struct Health Monit2026. https://doi.org/10.1177/14759217251405584, Epub ahead of print 12.
18.
HeDXuYSunH, et al.Self-supervised learning for vehicle bearing fault diagnosis based on time-frequency dual-domain contrast and fusion. Nonlinear Dyn2025; 113: 17385–17412. https://doi.org/10.1007/s11071-025-11101-7
19.
StockwellRGMansinhaLLoweRP. Localization of the complex spectrum: the S transform. IEEE Trans Signal Process1996; 44: 998–1001. https://doi.org/10.1109/78.492555
20.
XuYWangLYuG, et al.Generalized S-synchroextracting transform for fault diagnosis in rolling bearing. IEEE Trans Instrum Meas2022; 71: 1–14. https://doi.org/10.1109/tim.2021.3127305
21.
LiangPDengCWuJ, et al.Intelligent fault diagnosis of rolling element bearing based on convolutional neural network and frequency spectrograms. In: 2019 IEEE international conference on prognostics and health management (ICPHM). IEEE, 2019, pp. 1–5.
NeupaneDSeokJ. Bearing fault detection and diagnosis using case Western reserve university dataset with deep learning approaches: a review. IEEE Access2020; 8: 93155–93178. https://doi.org/10.1109/access.2020.2990528
25.
YangCQiaoZLiuL, et al.Positive-incentive noise in artificial intelligence-enabled machine fault diagnosis. Struct Health Monit2025. https://doi.org/10.1177/14759217251370358, Epub ahead of print 12.
