Deep imbalanced domain adaptation for transfer learning fault diagnosis of bearings under multiple working conditions

The tremendous success of deep learning and transfer learning broadened the scope of fault diagnosis, especially the latter further improved the diagnosis accuracy under multiple working conditions. However, most existing attempts assume that label distribution is domain-invariant despite taking int...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Reliability engineering & system safety Ročník 230; s. 108890
Hlavní autoři: Ding, Yifei, Jia, Minping, Zhuang, Jichao, Cao, Yudong, Zhao, Xiaoli, Lee, Chi-Guhn
Médium: Journal Article
Jazyk:angličtina
Vydáno: Barking Elsevier Ltd 01.02.2023
Elsevier BV
Témata:
Adaptation
Bearings
Deep learning
Domain shift
Domains
Fault diagnosis
Imbalanced domain adaptation
Label shift
Machine learning
Regularization
Reliability engineering
Transfer learning
Working conditions
Fault diagnosis
Domain shift
Bearings
Label shift
Imbalanced domain adaptation
ISSN:0951-8320, 1879-0836
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:The tremendous success of deep learning and transfer learning broadened the scope of fault diagnosis, especially the latter further improved the diagnosis accuracy under multiple working conditions. However, most existing attempts assume that label distribution is domain-invariant despite taking into account the different feature distributions. This does not accommodate the diversity of fault mode distributions under different operating conditions and weakens the generalization to imbalanced domain adaptation (IDA) scenarios. Therefore, this work proposed a novel deep imbalanced domain adaptation (DIDA) framework for fault diagnosis of bearings, aiming at the challenging scenario where feature shift and label shift exist simultaneously under different working conditions. Specifically, DIDA overcomes the class-imbalanced label shift and achieves a fine-grained latent space matching by cost-sensitive learning and categorical alignment. Besides, margin loss regularization is introduced to further optimize classification boundaries and improve cross-domain generalization for IDA fault diagnosis tasks. Finally, we simulated the IDA protocols on experimental datasets and conducted case studies under multiple working conditions, thus validating the effectiveness and superiority of the proposed framework. •A novel deep imbalanced domain adaptation (DIDA) is proposed.•DIDA narrows both feature shift and label shift.•DIDA broadens fault diagnosis to IDA scenarios.•Experimental case studies verified validity and superiority.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0951-8320
1879-0836
DOI:10.1016/j.ress.2022.108890