In practical industrial scenarios, the fault diagnosis for rolling bearings faces the challenges of scarce fault data and strong interference. The adversarial transfer learning can address those issues, but it often relies on global feature alignment, overlooking local and class-level features, leading to poor performance under strong interference. To overcome these difficulties, this paper proposes a fault diagnosis method based on multi-adversarial domain transfer network with adaptive threshold generation pseudolabels (MADTN-ATGP). A two-stage adversarial transfer learning strategy with dual feature extractors is proposed to address the difficulty of extracting transferable features under strong interference. The high-quality target domain pseudolabels are generated by an improved adaptive threshold strategy. Finally, global and local alignments are achieved by adversarial training of global/local domain discriminators, and the local maximum mean discrepancy (LMMD) is introduced to construct a joint loss function to achieve class-level alignment. The transfer tasks from interference-free data to interference data collected from an external interference simulation experimental platform are verified to demonstrate the superiority of the proposed method, offering a reliable solution for fault diagnosis under strong interference.
FeiCHanYWenR, et al. (2024) Deep learning-based modeling method for probabilistic LCF life prediction of turbine blisk. Propulsion and Power Research13(1): 12–25.
2.
FuLWangSChenR, et al. (2023) Intelligent fault diagnosis of rolling bearings based on an improved empirical wavelet transform and ResNet under variable conditions. IEEE Sensors Journal23(23): 29097–29108.
3.
FuGWangXLiuY, et al. (2025) A robust bearing fault diagnosis method based on ensemble learning with adaptive weight selection. Expert Systems with Applications269: 126420.
4.
GaninYUstinovaEAjakanH, et al. (2016) Domain-adversarial training of neural networks. Journal of Machine Learning Research17(1): 2096.
5.
GaoTYangJTangQ (2024) A multi-source domain information fusion network for rotating machinery fault diagnosis under variable operating conditions. Information Fusion106: 102278.
6.
GrettonABorgwardtKMRaschMJ, et al. (2012) A kernel two-sample test. Journal of Machine Learning Research13: 723–773.
7.
GuoLLeiYXingS, et al. (2019) Deep convolutional transfer learning network: a new method for intelligent fault diagnosis of machines with unlabeled data. IEEE Transactions on Industrial Electronics66(9): 7316–7325.
8.
GuoYZhangJSunB, et al. (2023) Adversarial deep transfer learning in fault diagnosis: progress, challenges, and future prospects. Sensors23(16): 7263.
9.
HanBZhangXWangJ, et al. (2021) Hybrid distance-guided adversarial network for intelligent fault diagnosis under different working conditions. Measurement176: 109197.
10.
KouwWMLoogM (2021) A review of domain adaptation without target labels. IEEE Transactions on Pattern Analysis and Machine Intelligence43(3): 766–785.
11.
LiTZhaoZSunC, et al. (2021) Domain adversarial graph convolutional network for fault diagnosis under variable working conditions. IEEE Transactions on Instrumentation and Measurement70(1): 1–10.
12.
LiFWangLWangD, et al. (2023) An adaptive multiscale fully convolutional network for bearing fault diagnosis under noisy environments. Measurement216: 112993.
13.
LongMCaoYCaoZ, et al. (2019) Transferable representation learning with deep adaptation networks. IEEE Transactions on Pattern Analysis and Machine Intelligence41(12): 3071–3085.
14.
LuXXuWJiangQ, et al. (2023) Category-aware dual adversarial domain adaptation model for rolling bearings fault diagnosis under variable conditions. Measurement Science and Technology34(9): 095104.
15.
PeiXZhengXWuJ (2021) Rotating machinery fault diagnosis through a transformer convolution network subjected to transfer learning. IEEE Transactions on Instrumentation and Measurement70: 1–11.
16.
RagabMChenZWuM, et al. (2020) Adversarial multiple-target domain adaptation for fault classification. IEEE Transactions on Instrumentation and MeasurementPP(99): 1.
17.
SheDYangZDuanY, et al. (2024) A meta transfer learning method for gearbox fault diagnosis with limited data. Measurement Science and Technology35(8): 086114.
18.
SonkulNADhageGSVyasNS (2021) Single and multi-label fault classification in rotors from unprocessed multi-sensor data through deep and parallel CNN architectures. Expert Systems with Applications185: 115565.
19.
SunBSaenkoK (2016) Deep CORAL: correlation alignment for deep domain adaptation. In: GuangHHervéJ (eds) Computer Vision – ECCV 2016 Workshops. Springer International Publishing, 443–450.
20.
SunYTaoHStojanovicV (2025a) End-to-end multi-scale residual network with parallel attention mechanism for fault diagnosis under noise and small samples. ISA Transactions157(000): 419–433.
21.
SunYTaoHStojanovicV (2025b) Pseudo-label guided dual classifier domain adversarial network for unsupervised cross-domain fault diagnosis with small samples. Advanced Engineering Informatics64: 102986.
22.
WanSZhangX (2020) Bearing fault diagnosis based on teager energy entropy and mean-shift fuzzy C-means. Structural Health Monitoring19(6): 1976–1988.
23.
WangSXiangJZhongY, et al. (2018) Convolutional neural network-based hidden markov models for rolling element bearing fault identification. Knowledge-Based Systems144: 65–76.
24.
WangXShenCXiaM, et al. (2020) Multi-scale deep intra-class transfer learning for bearing fault diagnosis. Reliability Engineering & System Safety202: 107050.
25.
WangYGeLXueC, et al. (2023) Multiple local domains transfer network for equipment fault intelligent identification. Engineering Applications of Artificial Intelligence120: 105791.
26.
WeiYPengHRongM, et al. (2025) You can be more trustworthy: a feature fusion reinforcement network for credible anti-noise fault diagnosis. Advanced Engineering Informatics64: 103056.
27.
WenHGuoWLiX (2023) A novel deep clustering network using multi-representation autoencoder and adversarial learning for large cross-domain fault diagnosis of rolling bearings. Expert Systems with Applications225: 120066.
28.
YangJFZhangNHeYL, et al. (2024) Novel dual-network autoencoder based adversarial domain adaptation with wasserstein divergence for fault diagnosis of unlabeled data. Expert Systems with Applications238: 122393.
29.
YuXWangYLiangZ, et al. (2023a) An adaptive domain adaptation method for rolling bearings’ fault diagnosis fusing deep convolution and self-attention networks. IEEE Transactions on Instrumentation and Measurement72: 1–14.
30.
YuZZhangCDengC (2023b) An improved GNN using dynamic graph embedding mechanism: a novel end-to-end framework for rolling bearing fault diagnosis under variable working conditions. Mechanical Systems and Signal Processing200: 110534.
31.
ZhangWPengGLiC, et al. (2017) A new deep learning model for fault diagnosis with good anti-noise and domain adaptation ability on raw vibration signals. Sensors17(2): 425.
32.
ZhangRZhaoZTaoX, et al. (2024) Fault diagnosis method for bearings under variable working conditions based on transfer relation network. Measurement Science and Technology35(12): 126202.
33.
ZhaoZAlzubaidiLZhangJ, et al. (2024) A comparison review of transfer learning and self-supervised learning: definitions, applications, advantages and limitations. Expert Systems with Applications242: 122807.
34.
ZhouTYaoDYangJ, et al. (2024) DRSwin-ST: an intelligent fault diagnosis framework based on dynamic threshold noise reduction and sparse transformer with shifted windows. Reliability Engineering & System Safety250: 110327.
35.
ZhuYZhuangFWangJ, et al. (2021) Deep subdomain adaptation network for image classification. IEEE Transactions on Neural Networks and Learning Systems32(4): 1713–1722.
36.
ZhuJWangYHuangC, et al. (2024) A new incremental learning for bearing fault diagnosis under noisy conditions using classification and feature-level information. IEEE Transactions on Instrumentation and Measurement73(1): 1–14.
