@article{TU2026, 
author = {Xuyuan TU and Qi DENG and Zuoxiu ZHANG and Zimuzhi WANG and Jun WU},
title = {Small sample gearbox fault diagnosis method based on a frequency band attention network},
year = {2026},
journal = {Chinese Journal of Ship Research},
volume = {21},
number = {3},
pages = {263-271},
keywords = {deep learning, gearboxes, rotating machinery, failure analysis, signal reconstruction, small-sample, frequency band attention network},
url = {https://www.sciopen.com/article/10.19693/j.issn.1673-3185.04384},
doi = {10.19693/j.issn.1673-3185.04384},
abstract = {ObjectiveDeep learning-based fault diagnosis methods typically require large amounts of fault data. To enable accurate gearbox fault diagnosis in small-sample scenarios, a novel diagnosis method based on a frequency band attention network is proposed. MethodFirst, a reconstruction-encoding layer is used to transform vibration signals into sub-band encoded signals that are more suitable for classification. Then, an intrinsic band attention layer is designed to effectively extract salient time-frequency features from the sub-band encoded signals. Finally, a multi-feature fusion module is used to integrate the extracted time-frequency features for fault recognition in small-sample conditions.ResultsExperimental results on a gearbox fault simulation platform show that the proposed method achieves a fault diagnosis accuracy of 99.85% in small-sample conditions, surpassing existing benchmark models. ConclusionThese findings can provide a valuable reference for gearbox fault diagnosis in small-sample conditions.}
}