TF-MESN: a multi-sensor collaborative small-sample fault identification method for offshore wind turbine gearbox
Measurement Science and Technology · 10.04.2026 · 37 (15) · 155008
DOI: 10.1088/1361-6501/ae59a2 ↗Kurz gefasst
Fachartikel in der Zeitschrift Measurement Science and Technology, online erschienen am 16.04.2026. Verfasst von Qing Xu und 5 weiteren Autor:innen. Die Arbeit nennt 29 Literaturangaben. Gefördert von: Key Laboratory of Marine Environmental Survey Technology and Application Ministry of Natural Resources P.R. China, Guangdong Basic and Applied Basic Research Foundation, Key Field Special Project of Department of Education of Guangdong Province.
Abstract (Originalsprache)
Offshore wind turbines face harsh conditions and high maintenance costs, necessitating precise fault classification based on deep learning. However, gearbox variability, scarce fault signals, and poor interpretability challenge this task. Therefore, this paper proposes a multimodal time–frequency enhancement synergistic network (TF-MESN). First, a time–frequency graphic block compression enhancement method is introduced, to highlight local feature structures and macroscopic time–frequency domain features of signals. Second, a cascaded heterogeneous convolution module is proposed, which uses heterogeneous convolution kernels through cascaded residual structures of standard convolution and separable convolution to build a convolutional neural network, capable of capturing signal features comprehensively while controlling the number of model parameters effectively. Third, an improved multiscale statistics and statistical dynamic spatial channel reconstruction module is proposed, which can reasonably segment effective signals from redundant ones, making the model more suitable for complex scenarios where steady-state and transient faults coexist. Finally, multi-channel feature fusion information is input into a multi-scale convolutional weighted classifier, which utilizes stride-based multi-scale down sampling convolution and multiple statistics for fault classification adapted to complex information representation. Experiments on two gearbox datasets, validated via visualization, show that TFMESN outperforms four representative methods, with superior generalization and small-sample fault identification capability, meeting offshore wind gearbox fault diagnosis needs.
Förderung
- Key Laboratory of Marine Environmental Survey Technology and Application Ministry of Natural Resources P.R. China
- Guangdong Basic and Applied Basic Research Foundation
- Key Field Special Project of Department of Education of Guangdong Province
- National Natural Science Foundation of China
- Natural Science Foundation of Guangdong Province
- Special Fund for Scientific and Technological Innovation Strategy of Guangdong Province
- Key Realm R&D Program of Guangdong Province
- Chinese Academy of Sciences Science and Technology Service Network Program Huangpu
Zitieren
APA
Xu, Q., Cui, D., Peng, Z., Li, Q., He, J., & Xiong, J. (2026). TF-MESN: a multi-sensor collaborative small-sample fault identification method for offshore wind turbine gearbox. Measurement Science and Technology, 37(15), 155008. https://doi.org/10.1088/1361-6501/ae59a2
Vancouver
Xu Q, Cui D, Peng Z, Li Q, He J, Xiong J. TF-MESN: a multi-sensor collaborative small-sample fault identification method for offshore wind turbine gearbox. Measurement Science and Technology. 2026;37(15):155008. doi:10.1088/1361-6501/ae59a2
BibTeX
@article{Xu2026,
author = {Xu, Qing and Cui, Delong and Peng, Zhiping and Li, Qirui and He, Jieguang and Xiong, Jianbin},
title = {TF-MESN: a multi-sensor collaborative small-sample fault identification method for offshore wind turbine gearbox},
journal = {Measurement Science and Technology},
year = {2026},
volume = {37},
number = {15},
pages = {155008},
doi = {10.1088/1361-6501/ae59a2}
}
EUVIMED