Gearboxes are critical mechanical components widely deployed in industrial applications, where their reliable operation directly impacts system safety and efficiency. However, conventional fault diagnostic approaches face significant challenges when operating under extraneous transient noise conditions, particularly with limited labeled fault samples. These challenges manifest as performance degradation with extremely sparse labeled datasets, vulnerability in pseudo-label generation mechanisms under intense transient noise, and inconsistent feature scale representations due to noise-induced interference. Furthermore, existing methods struggle to maintain diagnostic accuracy when confronted with both data scarcity and transient disturbances, often resulting in compromised model generalization and unreliable fault classification. To address these limitations, this research proposes the Semi-Supervised Transfer Graph Representation Learning with Few-Shot Adaptation (SSTGRL-FSA) framework, featuring three innovative components: a novel pseudo-label reliability enhancement mechanism leveraging systematic knowledge transfer from established source domains, an advanced label transmission and matching strategy exploiting homologous signal patterns across operational domains, and an integrated first-order Markov state probability transition matrix with amplitude-constrained scaling. SSTGRL-FSA significantly advances the field by effectively handling both labeled data scarcity and transient noise interference while enhancing model robustness through sophisticated temporal dependency modeling and stable feature scale maintenance, ultimately providing a more reliable and practical solution for industrial gearbox fault diagnosis under challenging operational conditions.
BacharLMataniaOCohenR, et al. Anomaly detection of gear wear: introducing a novel health indicator through experimentation and physical investigation. Struc Health Monitor. Epub ahead of print 13January2025. DOI: 10.1177/14759217241307837.
2.
ChenPMaJHeC, et al. Semi-supervised consistency models for automated defect detection in carbon fiber composite structures with limited data. Meas Sci Technol2025; 36(4): 046109.
3.
XuCChenPGaoJ, et al. Semi-supervised transfer learning preserving spatial homogeneity for gearbox diagnostics in extraneous transient noise. Nondestruct Test Eval. Epub ahead of print 28August2025. DOI: 10.1080/10589759.2025.2544893.
4.
ChenPXuCMaZ, et al. A mixed samples-driven methodology based on denoising diffusion probabilistic model for identifying damage in carbon fiber composite structures. IEEE Trans Instrum Meas2023; 72: 3513411.
5.
WangYChenPWeiQ, et al. Multi-channel fusion scale transformed signals with magnetic leakage for damage detection in steel wire ropes. Nondestr Test Eval. Epub ahead of print 9August2025. DOI: 10.1080/10589759.2025.2543513.
6.
AhmadHChengWXingJ, et al. Deep learning-based fault diagnosis of planetary gearbox: a systematic review. J Manuf Syst2024; 77: 730–745.
7.
ChenPZhangRFanS, et al. Step-wise contrastive representation learning for diagnosing unknown defective categories in planetary gearboxes. Knowl Based Syst2024; 309: 112863.
8.
Ghanbari KohyaniNRiyaziYShiraziFA, et al. Noise-robust gearbox fault detection: a deep learning approach. J Theor Appl Vib Acoust2024; 10(1): 54–66.
9.
LiDLiCYangJ, et al. Bayesian optimization-attention-feedforward neural network based train traction motor-gearbox coupled noise prediction. Measurement2024; 238: 115323.
10.
ChenPZhangRHeC, et al. Progressive contrastive representation learning for defect diagnosis in aluminum disk substrates with a bio-inspired vision sensor. Expert Syst Appl2025; 289: 128305.
11.
ZhangQSongCYuanY.Fault diagnosis of vehicle gearboxes based on adaptive wavelet threshold and LT-PCA-NGO-SVM. Appl Sci2024; 14(3): 1212.
12.
WangZGaoYYuJ, et al. Data-driven fault diagnosis of PEMFC water management with segmented cell and deep learning technologies. Int J Hydrogen Energy2024; 67: 715–727.
13.
ChenPMaZXuC, et al. Self-supervised transfer learning for remote wear evaluation in machine tool elements with imaging transmission attenuation. IEEE Internet Things J2024; 11: 23045–23054.
14.
ChenPGaoJJinY, et al. Physics-aware digital twin and few-shot learning for self-supervised gearbox fault diagnosis. Meas Sci Technol2025; 36(9): 096106.
15.
El IdrissiADerouichAMahfoudS, et al. Bearing faults diagnosis by current envelope analysis under direct torque control based on neural networks and fuzzy logic—a comparative study. Electronics2024; 13(16): 3195.
16.
ChenPGaoJZhangR, et al. Metric-guided graph contrastive learning: an unsupervised approach for few-shot gearbox fault diagnosis. Meas Sci Technol2025; 36(7): 076110.
17.
ChenPWuYXuC, et al. Interference suppression of nonstationary signals for bearing diagnosis under transient noise measurements. IEEE Trans Reliab2025; 74: 4047–4061.
18.
ChenPWuYFanS, et al. Adaptive signal regime for identifying transient shifts: a novel approach toward fault diagnosis in wind turbine systems. Ocean Eng2025; 325: 120798.
19.
ChenPWuYXuC, et al. Markov modeling of signal condition transitions for bearing diagnostics under external interference conditions. IEEE Trans Instrum Meas2024; 73: 3518308.
20.
VashishthaGChauhanSZimrozR, et al. Optimization of spectral kurtosis-based filtering through flow direction algorithm for early fault detection. Measurement2025; 241: 115737.
21.
ChinCAbdullahSAriffinA, et al. A review of the wavelet transform for durability and structural health monitoring in automotive applications. Alexandria Eng J2024; 99: 204–216.
22.
YuzgecUDokurEBalciM.A novel hybrid model based on empirical mode decomposition and echo state network for wind power forecasting. Energy2024; 300: 131546.
23.
IndragandhiVKumarRSSaranyaR.Hilbert-Huang Transform and machine learning based electromechanical analysis of induction machine under power quality disturbances. Results Eng2024; 24: 103075.
24.
KeshunYPuzhouWPengH, et al. A sound-vibration physical-information fusion constraint-guided deep learning method for rolling bearing fault diagnosis. Reliab Eng Syst Saf2025; 253: 110556.
25.
BilalHObaidatMSAslamMS, et al. Online fault diagnosis of industrial robot using IoRT and hybrid deep learning techniques: an experimental approach. IEEE Internet Things J2024; 11(19): 31422–31437.
26.
AminAAMubarakAWaseemS.Application of physics-informed neural networks in fault diagnosis and fault-tolerant control design for electric vehicles: a review. Measurement2025; 246: 116728.
27.
LuLWangSYuX, et al. An advanced fault diagnosis approach for wind turbine planetary gearbox based on optimized multi-layer attention denoising autoencoders. Meas Sci Technol2025; 36(2): 026132.
28.
CuiDZhouRLiH, et al. Rotor fault diagnosis of centrifugal pumps in nuclear power plants based on CWGAN-GP-CNN for imbalanced dataset. Prog Nucl Energy2025; 178: 105500.
29.
JoungBGNathCLiZ, et al. Bearing anomaly detection in an air compressor using an LSTM and RNN-based machine learning model. Int J Adv Manuf Technol2024; 134(7): 3519–3530.
30.
ObaidoGMienyeIDEgbelowoOF, et al. Supervised machine learning in drug discovery and development: algorithms, applications, challenges, and prospects. Mach Learn Appl2024; 17: 100576.
31.
GuiJChenTZhangJ, et al. Tao. A survey on self-supervised learning: algorithms, applications, and future trends. IEEE Trans Pattern Anal Mach Intell2024; 46(12): 9052–9071.
32.
LiangGLiFPangX, et al. A gear fault diagnosis method based on reactive power and semi-supervised learning. Meas Sci Technol2024; 35(12): 126107.
33.
YeTYuanXYangX, et al. A semi-supervised intelligent fault diagnosis method for bearings under low labeled rates. IEEE Trans Instrum Meas2024; 73: 3531111.
34.
SuHYaoQXiangL, et al. Semi-supervised temporal meta-learning framework for wind turbine bearing fault diagnosis under limited annotation data. IEEE Trans Instrum Meas2024; 73: 3512309.
35.
ChenXChenZGuoL, et al. Pseudo-label assisted semi-supervised adversarial enhancement learning for fault diagnosis of gearbox degradation with limited data. Mech Syst Signal Process2025; 224: 112108.
36.
HuangQLiCHanY, et al. Semi-supervised prototype networks with similarity information selection for fault diagnosis of wind turbine gearboxes. IEEE Trans Instrum Meas2025; 74: 3507012.
37.
CaiWZhaoDWangT.Spatial-temporal graph attention contrastive learning for semi-supervised bearing fault diagnosis with limited labeled samples. Comput Ind Eng2025; 204: 111106.
38.
JiaMYaoYLiuY.Review on graph neural networks for process soft sensor development, fault diagnosis, and process monitoring. Ind Eng Chem Res2025; 64(17): 8543–8564.
39.
DangXLiYMaW, et al. Towards exploring the limitations of test selection techniques on graph neural networks: an empirical study. Empir Softw Eng2024; 29(5): 112.
40.
YueKWangLDingX, et al. Physics-informed dual guidance method using physical envelope harmonic distribution and transfer learning for few-shot gear fault classification. Eng Appl Artif Intell2025; 153: 110956.
41.
LuoRLiYGuoH, et al. Cross-operating-condition fault diagnosis of a small module reactor based on CNN-LSTM transfer learning with limited data. Energy2024; 313: 133901.
42.
DaiXYiKWangF, et al. Bearing fault diagnosis based on POA-VMD with GADF-Swin Transformer transfer learning network. Measurement2024; 238: 115328.
43.
DongXChengMWangH, et al. A potential solution to insufficient target-domain noise data: transfer learning and noise modeling. IEEE Trans Geosci Remote Sens2023; 61: 1–15.
44.
ZhanXLiuCZhangA, et al. Prediction of compressor aerodynamic noise based on transfer learning Random Forest. J Acoust Soc Am158(1): 336–351.
45.
BonetDMontserratDMGiró-i Nieto, et al. Hyperfast: instant classification for tabular data. Proc AAAI Conf Artif Intell2024; 38: 11114–11123.
46.
HooSBMüllerSSalinasD, et al. The tabular foundation model TabPFN outperforms specialized time series forecasting models based on simple features. In: NeurIPS workshop on time series in the age of large models, https://arxiv.org/pdf/2501.02945 (2024).
47.
YanSShaoHXiaoY, et al. Semi-supervised fault diagnosis of machinery using LPS-DGAT under speed fluctuation and extremely low labeled rates. Adv Eng Inform2022; 53: 101648.
48.
HuangKChenC.Subgraph generation applied in GraphSAGE deal with imbalanced node classification. Soft Comput2024; 28(17): 10727–10740.
49.
GuoJLiuCLiuS, et al. Graph contrastive learning for semi-supervised wind turbine fault diagnosis with few labeled SCADA data. Measurement2025; 245: 116531.
50.
LiuZPengZ.Few-shot bearing fault diagnosis by semi-supervised meta-learning with graph convolutional neural network under variable working conditions. Measurement2025; 240: 115402.
51.
LiJDingWMaoW, et al. Causal dilated Convolution-Based residual DenseNet with channel attention for RUL prediction of rolling bearings. Measurement2024; 235: 115012.
52.
QinGZhangKLaiX, et al. An adaptive symmetric loss in dynamic wide-kernel ResNet for rotating machinery fault diagnosis under noisy labels. IEEE Trans Instrum Meas2024; 73: 3517512.
53.
RaoM.Development of deep learning-based methods for rotating machinery fault diagnosis under varying speed conditions. PhD thesis, University of Alberta, 2023.
