Early fault detection in rolling bearings is essential for monitoring the condition of industrial equipment. However, detection accuracy is often limited by noise interference and weak fault signals, a semi-supervised method based on a Siamese Denoising Normalizing Flow Convolutional Variational Autoencoder (Siamese DNFCVAE) is introduced. Firstly, a novel Denoising Normalizing Flow (DNF) module is designed and embedded into the CVAE encoder, which dynamically adjusts denoising intensity at each transformation step through adaptive noise estimation networks and scaling factors, progressively filtering noise while enhancing posterior distribution flexibility. Secondly, a Siamese architecture is integrated with DNFCVAE to introduce contrastive learning mechanisms, explicitly modeling inter-sample similarities to amplify sensitivity to early weak fault signals. Finally, an adaptive loss weight optimization strategy is developed, which automatically balances reconstruction and contrastive objectives, and realizes more stable and efficient model training. Experimental results on two rolling bearing datasets demonstrate that the Siamese DNFCVAE model achieves superior fault detection performance under complex noise conditions, significantly surpassing existing methods and highlighting its effectiveness and practicality for early fault detection.
BerroucheYVashishthaGChauhanS, et al. (2024) Local damage detection in rolling element bearings based on a single ensemble empirical mode decomposition. Knowledge-Based Systems301: 112265.
2.
Campos-RomeroMCarranza-GarcíaMRiquelmeJ (2024) Advancing unsupervised anomaly detection with normalizing flow and multi-scale ensemble learning. Engineering Applications of Artificial Intelligence136: 109088.
3.
ChenZGrylliasKLiW (2021a) Mechanical fault diagnosis using convolutional neural networks and extreme learning machine. Mechanical Systems and Signal Processing133: 106272.
4.
ChenGYanCMengJ, et al. (2021b) Improved VMD-FRFT based on initial center frequency for early fault diagnosis of rolling element bearing. Measurement Science and Technology32: 115024.
5.
ChenBChenXChenF, et al. (2022) Integrated early fault diagnosis method based on direct fast iterative filtering decomposition and effective weighted sparseness kurtosis to rolling bearings. Mechanical Systems and Signal Processing171: 108897.
6.
FangXZhengJJiangB (2024) A rolling bearing fault diagnosis method based on vibro-acoustic data fusion and fast fourier transform (FFT). International Journal of Data Science and Analytics20: 1–10.
7.
HeDZhaoJJinZ, et al. (2025) DCAGGCN: a novel method for remaining useful life prediction of bearings. Reliability Engineering & System Safety260: 110978.
HouZWangHYueY, et al. (2024) A novel framework based on two-stage multi-view feature optimization and improved support vector data description for aeroengine bearing early fault detection. Reliability Engineering & System Safety245: 110027.
10.
HuMWangGMaK, et al. (2020) Bearing performance degradation assessment based on optimized EWT and CNN. Measurement172: 108868.
11.
KangYChenGWangH, et al. (2024) Early fault detection of rolling bearings based on self-supervised deep one classification. Journal of Aerospace Power7: 1–12.
12.
LeeJQiuHYuG, et al. (2007) IMS, University of Cincinnati. In: “Bearing Data Set”, NASA Prognostics Data Repository. Moffett Field, CA: NASA Ames Research Center.
13.
LiTWuXLuoZ, et al. (2024) A bearing fault diagnosis method under small sample conditions based on the fractional order Siamese deep residual shrinkage network. Fractal and Fractional8: 134.
14.
LiuBYanCLiuY, et al. (2023) Multiscale residual anti-noise network via interpretable dynamic recalibration mechanism for rolling bearing fault diagnosis with few samples. IEEE Sensors Journal23: 31425–31439.
15.
LiuGLiRZhouX, et al. (2023) Reconstruction and fast prediction of 3D heat and mass transfer based on a variational autoencoder. International Communications in Heat and Mass Transfer149: 107112.
16.
LiuJYangGLiX, et al. (2024) A deep generative model based on CNN-CVAE for wind turbine condition monitoring. Measurement Science and Technology34: 035902.
17.
LuJWuWHuangX, et al. (2024) A modified active learning intelligent fault diagnosis method for rolling bearings with unbalanced samples. Advanced Engineering Informatics60: 102397.
18.
LunusovaEGonzalezMSzipkaK, et al. (2024) Early fault diagnosis in rolling element bearings: comparative analysis of a knowledge-based and a data-driven approach. Journal of Intelligent Manufacturing35: 2327–2347.
19.
NiGChenJWangH (2020) Degradation assessment of rolling bearing towards safety based on random matrix single ring machine learning. Safety Science127: 104719.
20.
PanYChengDWeiT, et al. (2022) Rolling bearing performance degradation assessment based on deep belief network and improved support vector data description. Mechanical Systems and Signal Processing181: 109458.
21.
QinYLiuHMaoY (2024) Faulty rolling bearing digital twin model and its application in fault diagnosis with imbalanced samples. Advanced Engineering Informatics61: 102513.
22.
TianJJiangYZhangJ, et al. (2024) A novel data augmentation approach to fault diagnosis with class-imbalance problem. Reliability Engineering & System Safety243: 109832.
23.
WangBLeiYLiN, et al. (2018) A hybrid prognostics approach for estimating remaining useful life of rolling element bearings. IEEE Transactions on Reliability69: 1–12.
24.
WangBGuoYBZhangZ, et al. (2023) Developing and applying OEGOA-VMD algorithm for feature extraction for early fault detection in cryogenic rolling bearing. Measurement216: 112908.
25.
WangLLiuJLiuY, et al. (2024) A lightweight tomato leaf disease identification method based on shared-twin neural networks. IET Image Processing18: 2291–2303.
26.
WangJZhangQZhuG, et al. (2024) Utilizing VQ-VAE for end-to-end health indicator generation in predicting rolling bearing RUL. Measurement Science and Technology35: 066106.
27.
WangSVidalYPozoF (2025) An unsupervised approach to early fault detection and performance degradation assessment in bearings. Advanced Engineering Informatics68: 103620.
28.
XiaJZhangJWangY, et al. (2022) WC-KNNG-PG: watershed clustering based on k-nearest-neighbor graph and Pauta criterion. IET Image Processing121: 108177.
29.
XuYZhenDGuX, et al. (2021) Autocorrelated envelopes for early fault detection of rolling bearings. Mechanical Systems and Signal Processing146: 106990.
30.
XuXWangHLiangX, et al. (2025) Rolling bearing early fault detection method based on feature clustering fusion degradation index. Chinese Journal of Mechanical Engineering38: 106.
31.
YangCMaJWangX, et al. (2022) A novel performance degradation indicator RUL prediction model and its application in rolling bearing. ISA Transactions121: 349–364.
32.
YaoYLuXJiangQ, et al. (2024) SSPENet: semi-supervised prototype enhancement network for rolling bearing fault diagnosis under limited labeled samples. Advanced Engineering Informatics61: 102560.
33.
ZhaoQZhangX (2025) Prediction of remaining useful life for rolling bearing based on ISOMAP and multi-head self-attention with gated recurrent unit. Journal of Vibration and Control31: 3187–3205.
34.
ZhaoXHuYLiuJ, et al. (2024) A novel intelligent multicross domain fault diagnosis of servo motor-bearing system based on domain generalized graph convolution autoencoder. Structural Health Monitoring24: 1485–1499.
35.
ZhaoJHeDJinZ, et al. (2026) A new method for bearing remaining useful life prediction based on dynamic wavelet and physical information constraints. Expert Systems with Applications296: 129023.
36.
ZhuHJiangHYaoR, et al. (2023) Rolling bearing incipient fault feature extraction using impulse-enhanced sparse time-frequency representation. Measurement Science and Technology34: 105124.
37.
ZhuWMiaoQCaoY, et al. (2025) An early fault online detection model of rolling bearing based on deep attention convolutional autoencoder and multi-decision fusion under variable operation conditions. Measurement253: 117752.
