Unsupervised domain adaptation-based transfer learning (TL) has been widely used in rolling bearing fault diagnosis to overcome the problem of limited and non-identically distributed labeled data. Discrepancy-based alignment is a popular domain adaptation method in TL. However, due to the inability to completely eliminate domain drift, the classifier learned from the source domain may easily misclassify some target domain samples that are scattered near the decision edge. In this work, a multi-order moment matching-based domain adaptation is proposed to address the issue. Low- and high-order moment matching is simultaneously applied to describe the complex non-Gaussian distributions in more detail and realize coarse- and fine-grained hybrid domain alignment. Furthermore, a discriminative clustering approach is employed to extract domain-invariant features of inter-class discrimination and intra-class compactness, which effectively reduces the negative transfer caused by hard-aligned target samples. The application of the proposed model to the experimental dataset demonstrates that the model can significantly improve the diagnosis accuracy of rolling bearing faults in cross-working conditions. This study can be of assistance to engineers in promptly identifying and addressing rolling bearing faults, ultimately enhancing the reliability and safety of equipment.
YangZWangXYangR. Transfer learning based rolling bearing fault diagnosis. In: 2021 IEEE 10th data driven control and learning systems conference (DDCLS), Suzhou, China, 14–16 May 2021, pp. 354–359. IEEE.
2.
WanSLiuJLiX, et al. Transfer-learning-based bearing fault diagnosis between different machines: a multi-level adaptation network based on layered decoding and attention mechanism. Measurement2022; 203: 111996.
3.
TongJLiuCZhengJ, et al. 1D-DRSETL: a novel unsupervised transfer learning method for cross-condition fault diagnosis of rolling bearing. Meas Sci Technol2022; 33: 085110.
4.
XiaPHuangYTaoZ, et al. A digital twin-enhanced semi-supervised framework for motor fault diagnosis based on phase-contrastive current dot pattern. Reliab Eng Syst Saf2023; 235: 109256.
5.
ChenZQinWHeG, et al. Explainable deep ensemble model for bearing fault diagnosis under variable conditions. IEEE Sensors J2023; 23: 17737–17750.
6.
XuZLiCYangY.Fault diagnosis of rolling bearings using an improved multi-scale convolutional neural network with feature attention mechanism. ISA Trans2021; 110: 379–393.
7.
WangQXuYYangS, et al. A Cross-Domain Bearing Fault Diagnosis Method with Multi-Source Incomplete Data. SSRN J. Epub ahead of print 2022. DOI: 10.2139/ssrn.4066966.
8.
AzariMSFlamminiFSantiniS, et al. A systematic literature review on transfer learning for predictive maintenance in Industry 4.0. IEEE Access2023; 11: 12887–12910.
9.
ZhangWLiXMaH, et al. Universal domain adaptation in fault diagnostics with hybrid weighted deep adversarial learning. IEEE Trans Ind Inf2021; 17: 7957–7967.
10.
ZhangYLiuYSunP, et al. IFCNN: a general image fusion framework based on convolutional neural network. Inf Fusion2020, 54: 99–118.
11.
NiQJiJCHalkonBJ, et al. Physics-informed residual network (PIResNet) for rolling element bearing fault diagnostics. Mech Syst Signal Process2023; 200: 110544.
12.
PengFZhengLPengY, et al. Digital Twin for rolling bearings: a review of current simulation and PHM techniques. Measurement2022; 201: 111728.
13.
DengYDuSWangD, et al. A calibration-based hybrid transfer learning framework for RUL prediction of rolling bearing across different machines. IEEE Trans Instrum Meas2023; 72: 1–15.
14.
WangJYangSLiuY, et al. Deep subdomain transfer learning with spatial attention ConvLSTM network for fault diagnosis of wheelset bearing in high-speed trains. Machines2023; 11: 304.
15.
QianQQinYLuoJ, et al. Deep discriminative transfer learning network for cross-machine fault diagnosis. Mech Syst Signal Process2023; 186: 109884.
16.
ZhaoKJiaFShaoH.A novel conditional weighting transfer Wasserstein auto-encoder for rolling bearing fault diagnosis with multi-source domains. Knowl Based Syst2023; 262: 110203.
17.
ChenCFuZChenZ, et al. HoMM: higher-order moment matching for unsupervised domain adaptation. AAAI2020; 34(4): 3422–3429.
18.
WangRHuangWShiJ, et al. A higher-order moment matching based fine-grained adversarial domain adaptation method for intelligent bearing fault diagnosis. Meas Sci Technol2022; 33(5): 055022.
19.
HuQSiXQinA, et al. Balanced adaptation regularization based transfer learning for unsupervised cross-domain fault diagnosis. IEEE Sensors J2022; 22: 12139–12151.
20.
TianQXuHSunH, et al. Unsupervised domain adaptation through high-order tensor matching with discriminative manifold learning. Comput Electr Eng2022; 103: 108375.
21.
DanJJinTChiH, et al. HOMDA: high-order moment-based domain alignment for unsupervised domain adaptation. Knowl Based Syst2023; 261: 110205.
22.
LiRLiSXuK, et al. Adversarial domain adaptation of asymmetric mapping with CORAL alignment for intelligent fault diagnosis. Meas Sci Technol2022; 33: 055101.
23.
ZhaiJZhangSChenJ, et al. Autoencoder and its various variants. In 2018 IEEE international conference on systems, man, and cybernetics (SMC) Miyazaki, Japan, 07–10 October 2018, pp.415–419. IEEE.
24.
AnJChoS.Variational autoencoder based anomaly detection using reconstruction probability. Spl Lecture IE2015; 2(1): 1–18.
CemgilTGhaisasSDvijothamK, et al. The autoencoding variational autoencoder. Advances in Neural Information Processing Systems2020; 33: 15077–15087.
27.
GhorveiMKavianpourMBeheshtiMT, et al. Spatial graph convolutional neural network via structured subdomain adaptation and domain adversarial learning for bearing fault diagnosis, http://arxiv.org/abs/2112.06033 (2021, accessed 20 July 2023).
28.
CaoHShaoHLiuB, et al. Clustering-guided novel unsupervised domain adversarial network for partial transfer fault diagnosis of rotating machinery. IEEE Sensors J2022; 22: 14387–14396.
29.
HuoCJiangQShenY, et al. New transfer learning fault diagnosis method of rolling bearing based on ADC-CNN and LATL under variable conditions. Measurement2022; 188: 110587.
30.
TzengEHoffmanJZhangN, et al. Deep domain confusion: maximizing for domain invariance, http://arxiv.org/abs/1412.3474 (2014, accessed 29 November 2022).
31.
SunBSaenkoK.Deep CORAL: correlation alignment for deep domain adaptation, http://arxiv.org/abs/1607.01719 (2016, accessed 27 November 2022).
