Contrastive learning (CL) is a learning strategy that trains a model by learning to distinguish between positive and negative sample pairs, enabling it to effectively learn key features in data. However, data augmentation methods in classical CL often generate views that are semantically independent, making it difficult to distinguish between false negative samples and hard negative samples under strong noise interference. In view of this, a variational graph CL based on dynamic negative sample optimization (VGCL-DNSO) is proposed in this paper. In VGCL-DNSO, a soft attention value is generated through the similarity matrix to assign different attention to negative samples, effectively improving the discriminative ability of key features under strong noise interference. Meanwhile, to maximize the retention of the feature information of the data, the potential probability distribution of the data in the hidden space is learned, from which a more representative feature view is obtained. Through two different types of rolling bearing fault diagnosis experiments, the analysis results show that VGCL-DNSO demonstrates significant advantages in generalization and robustness.
PanHWuZChengJ, et al. Characteristic energy ratio Ramanujan-Gram: a novel optimal multi-bands demodulation method. IEEE Trans Instrum Meas2025; 74: 1–9.
2.
ShaoHMingYLiuY, et al. Small sample gearbox fault diagnosis based on improved deep forest in noisy environments. Nondestruct Test Eval2025; 40(8): 3935–3956.
3.
XiaoYShaoHLiuB. Evaluating calibration of deep fault diagnostic models under distribution shift. Comput Ind2025; 171: 104334.
4.
ZhuZLeiYQiG, et al. A review of the application of deep learning in intelligent fault diagnosis of rotating machinery. Measurement2023; 206: 112346.
5.
ZhangXHeCLuY, et al. Fault diagnosis for small samples based on attention mechanism. Measurement2022; 187: 110242.
6.
TongJLiuCBaoJ, et al. A novel ensemble learning-based multisensor information fusion method for rolling bearing fault diagnosis. IEEE Trans Instrum Meas2022; 72: 1–12.
7.
ZhangSLiuZChenY, et al. Selective kernel convolution deep residual network based on channel-spatial attention mechanism and feature fusion for mechanical fault diagnosis. ISA Trans2023; 133: 369–383.
8.
DingSChenRLiuH, et al. ASG-HOMGAT: a high-order multi-head graph attention network with adaptive small graph structure for rolling bearing fault diagnosis. Meas Sci Technol2024; 35(6): 065018.
9.
ChuangCRobinsonJWLinY, et al. Debiased contrastive learning. Adv Neural Inform Process Syst2020; 33: 8765–8775.
10.
AlbelwiS. Survey on self-supervised learning: auxiliary pretext tasks and contrastive learning methods in imaging. Entropy2022; 24 (4): 551–551.
11.
GuptaEGuptaV. Margin-aware optimized contrastive learning for enhanced self-supervised histopathological image classification. Health Inf Sci Syst2024; 13(1): 2.
12.
PhamTNiuAZhangK, et al. Self-supervised visual representation learning via residual momentum. IEEE Access2023; 11: 116706–116720.
13.
XieYXuZZhangJ, et al. Self-supervised learning of graph neural networks: a unified review. IEEE Trans Patt Anal Mach Intell2022; 45(2): 2412–2429.
14.
LiuJChenYRenQ, et al. Joint data augmentations for automated graph contrastive learning and forecasting. Complex Intell Syst2024; 10(5): 6481–6490.
15.
CaronMMisraIMairalJ, et al. Unsupervised learning of visual features by contrasting cluster assignments. Adv Neural Inform Process Syst2020; 33: 9912–9924.
16.
LinSLiuCZhouP, et al. Prototypical graph contrastive learning. IEEE Trans Neural Netw Learn Syst2022; 35(2): 2747–2758.
17.
KhoslaPTeterwakPWangC, et al. Supervised contrastive learning. Adv Neural Inform Process Syst2020; 33: 18661–18673.
18.
WangWLiCLiA, et al. One-stage self-supervised momentum contrastive learning network for open-set cross-domain fault diagnosis. Knowl Based Syst2023; 275: 110692.
19.
DingYZhuangJDingP, et al. Self-supervised pretraining via contrast learning for intelligent incipient fault detection of bearings. Reliab Eng Syst Saf2022; 218: 108126.
20.
YangCLiuJZhouK, et al. Semisupervised machine fault diagnosis fusing unsupervised graph contrastive learning. IEEE Trans Ind Inform2022; 19(8): 8644–8653.
21.
CuiLTianXWeiQ, et al. A self-attention based contrastive learning method for bearing fault diagnosis. Expert Syst Appl2024; 238: 121645.
22.
DaiYMeiZLiJ, et al. Clustering-based contrastive learning for fault diagnosis with few labeled samples. IEEE Trans Instrum Meas2023; 73: 1–13.
23.
JiaMShenBLiuY. Graph contrastive learning with adaptive distribution transformation for fault detection and isolation using limited labeled faults. IEEE Trans Instrum Meas2025; 74: 1–13.
24.
PanXChenHWangW, et al. Adversarial domain adaptation based on contrastive learning for bearings fault diagnosis. Simul Model Pract Theory2025; 139: 103058.
25.
CuiWBaiLYangX, et al. A new contrastive learning framework for reducing the effect of hard negatives. Knowledge-Based Systems2023; 260: 110121.
26.
WuZPanSChenF, et al. A comprehensive survey on graph neural networks. IEEE Trans Neural Netw Learn Syst2020; 32(1): 4–24.
27.
XuJMoSJiangZ, et al. A novel positive–negative graph convolutional network-based fault diagnosis method with application to complex systems. Neurocomputing2024; 600: 128145.
28.
ZhangLChengWXingJ, et al. Self-supervised variational graph autoencoder for system-level anomaly detection. IEEE Trans Instrum Meas2023; 72: 1–11.
29.
WeiRGarciaCEl-SayedA, et al. Variations in variational autoencoders-a comparative evaluation. IEEE Access2020; 8: 153651–153670.
30.
SureshSLiPHaoC, et al. Adversarial graph augmentation to improve graph contrastive learning. Adv Neural Inform Process Syst2021; 34: 15920–15933.
31.
ConnorRDearleAClaydonB, et al. Correlations of cross-entropy loss in machine learning. Entropy2024; 26(6): 491.
32.
DingAQinYWangB, et al. Evolvable graph neural network for system-level incremental fault diagnosis of train transmission systems. Mech Syst Signal Process2024; 210: 111175.
33.
RainioOTeuhoJKlénR. Evaluation metrics and statistical tests for machine learning. Sci Rep2024; 14(1): 6086.
34.
LiuHPanHZhengJ, et al. Broad distributed game learning for intelligent classification in rolling bearing fault diagnosis. Appl Soft Comput2024; 167: 112470.
