Cross-Operating-Condition Fault Diagnosis of Gearboxes Based on Multi-Source Subdomain Adaptation and ClassifierAlignment
Abstract
gearboxes based on multi-source subdomain adaptation and classifier alignment. After data augmentation and normalization, raw signals
from multiple source domains and the target domain are fed into a shared feature extractor. Private feature extractors map each source-target
domain pair to a specific feature space. Local maximum mean differences are used to align distributions across domains while minimizing discrepancies among all classifiers. The final prediction is obtained by averaging outputs from multiple classifiers.
Keywords
Full Text:
PDFReferences
[1] Zhang L, Fan Q, Lin J, et al. A nearly end-to-end deep learning approach to fault diagnosis of wind turbine gearboxes under nonstationary conditions[J]. Engineering applications of artificial intelligence, 2023, 119: 105735.
[2] Xia J, Huang R, Chen Z, et al. A novel digital twin-driven approach based on physical-virtual data fusion for gearbox fault diagnosis[J].
Reliability Engineering & System Safety, 2023, 240: 109542.
[3] Qian Q, Wang Y, Zhang T, et al. Maximum mean square discrepancy: a new discrepancy representation metric for mechanical fault
transfer diagnosis[J]. Knowledge-Based Systems, 2023, 276: 110748.
[4] Xu G, Huang C, da Silva D S, et al. A compressed unsupervised deep domain adaptation model for efficient cross-domain fault
diagnosis[J]. IEEE Transactions on Industrial Informatics, 2022, 19(5): 6741-6749.
[5] Wang Z, Zhang J, Ma K, et al. Towards Unsupervised Domain Adaptation Fault Diagnosis: A Multi-Source-Multi-Target Method[J].
IEEE Sensors Journal, 2024.
[6] Zhu Y, Zhuang F, Wang J, et al. Deep subdomain adaptation network for image classification[J]. IEEE transactions on neural networks
and learning systems, 2020, 32(4): 1713-1722.
[7] PHMSociety.(Aug.2019). PHM09 Data Challenge. [Online]. Available: https://www.phmsociety.org/competition/PHM/09/apparatus
[8] Liu D, Cui L, Cheng W. A review on deep learning in planetary gearbox health state recognition: methods, applications, and dataset
publication[J]. Measurement Science and Technology, 2023, 35(1): 012002.
[9] Zhang Y, Ren Z, Zhou S, et al. Adversarial domain adaptation with classifier alignment for cross-domain intelligent fault diagnosis of
multiple source domains[J]. Measurement Science and Technology, 2020, 32(3): 035102.
[10] Long M, Cao Y, Wang J, et al. Learning transferable features with deep adaptation networks[C]//International conference on machine
learning. PMLR, 2015: 97-105.
DOI: http://dx.doi.org/10.70711/frim.v4i3.8748
Refbacks
- There are currently no refbacks.
