Citation: | TU X Y, DENG Q, ZHANG Z X, et al. Small sample fault diagnosis method of gearbox based on frequency band attention network[J]. Chinese Journal of Ship Research, 2025, 20(X): 1–10 (in Chinese). DOI: 10.19693/j.issn.1673-3185.04384 |
Deep learning-based fault diagnosis methods generally require a large number of fault samples. To achieve accurate gearbox fault diagnosis under small-sample scenarios, a novel diagnosis method based on frequency band attention network is proposed.
Initially, through the reconstruction-encoding layer, the vibration signals are transformed into sub-band encoded signals that are easy to classify; Subsequently, the intrinsic band attention layer is constructed to fully mine the representative time-frequency features from sub-band encoded signals; Finally, the multiple feature fusion module is used to integrate the time-frequency feature information for fault recognition under small-sample conditions.
The validation results on a gearbox fault simulation test bench show that the proposed method achieves a fault diagnosis accuracy of 99.85% under small-sample conditions, outperforming the comparison models.
The research results can provide a reference for the fault diagnosis of gearbox under small sample conditions.
[1] |
尚前明, 蒋婉莹, 周毅, 等. 基于MPDCNN的强噪声环境下船舶电力推进器齿轮箱故障诊断方法[J]. 中国舰船研究, 2025, 20(2): 30–38. doi: 10.19693/j.issn.1673-3185.03814
SHANG Q M, JIANG W Y, ZHOU Y, et al. Fault diagnosis of marine electric thruster gearbox based on MPDCNN under strong noisy environments[J]. Chinese Journal of Ship Research, 2025, 20(2): 30–38 (in Chinese). doi: 10.19693/j.issn.1673-3185.03814
|
[2] |
顾钦平. 深度神经网络在船用齿轮箱故障诊断中的应用[J]. 舰船科学技术, 2023, 45(9): 176–179. doi: 10.3404/j.issn.1672-7649.2023.09.039
GU Q P. Application of deep neural networks in fault diagnosis of marine gearboxes[J]. Ship Science and Technology, 2023, 45(9): 176–179 (in Chinese). doi: 10.3404/j.issn.1672-7649.2023.09.039
|
[3] |
余凯伟, 李子睿, 陈冲, 等. 基于深度残差收缩自适应网络的推进轴系云边端协同故障诊断方法[J]. 中国舰船研究, 2025, 20(2): 1–9. doi: 10.19693/j.issn.1673-3185.03779
YU K W, LI Z R, CHEN C, et al. Deep residual shrinkage adaptive network-based cloud-edge-end collaborative fault diagnosis method for propulsion shafting system[J]. Chinese Journal of Ship Research, 2025, 20(2): 1–9 (in Chinese). doi: 10.19693/j.issn.1673-3185.03779
|
[4] |
ZHANG C, WANG F, ZHOU D, et al. A CNN-based fault diagnosis method of multi-function integrated RF system using frequency domain scanning with Lasso regression[J]. Knowledge-Based Systems, 2025, 309: 112836. doi: 10.1016/j.knosys.2024.112836
|
[5] |
JIAO J Y, ZHAO M, LIN J, et al. A comprehensive review on convolutional neural network in machine fault diagnosis[J]. Neurocomputing, 2020, 417: 36–63. doi: 10.1016/j.neucom.2020.07.088
|
[6] |
邓琪, 汪承杰, 万海波, 等. 基于相关熵和GMSVM的推进轴系故障诊断方法[J]. 华中科技大学学报(自然科学版), 2025, 53(4): 1–7. doi: 10.13245/j.hust.250151
DENG Q, WANG C J, WAN H B, et al. Correntropy and GMSVM-based fault diagnosis method for propulsion shaft system[J]. Journal of Huazhong University of Science and Technology (Nature Science Edition), 2025, 53(4): 1–7 (in Chinese). doi: 10.13245/j.hust.250151
|
[7] |
刘颉, 杨超颖, 周凯波. 基于图数据深度挖掘的旋转机械故障诊断[J]. 华中科技大学学报(自然科学版), 2021, 49(9): 1–5. doi: 10.13245/j.hust.210901
LIU J, YANG C Y, ZHOU K B. Fault diagnosis of rotating machinery based on graph data deep mining[J]. Journal of Huazhong University of Science and Technology (Nature Science Edition), 2021, 49(9): 1–5 (in Chinese). doi: 10.13245/j.hust.210901
|
[8] |
KUMAR S, SINGH O P, KUMAR V, et al. ML-based bevel gearbox fault diagnosis: an extensive time domain feature extraction approach with limited data[J]. International Journal of Non-Linear Mechanics, 2025, 170: 105003. doi: 10.1016/j.ijnonlinmec.2024.105003
|
[9] |
MU M Z, JIANG H K, WANG X, et al. Adaptive model-agnostic meta-learning network for cross-machine fault diagnosis with limited samples[J]. Engineering Applications of Artificial Intelligence, 2025, 141: 109748. doi: 10.1016/j.engappai.2024.109748
|
[10] |
郑直, 赵文博, 李克, 等. 基于改进多图卷积网络的液压泵小样本故障诊断[J]. 振动与冲击, 2024, 43(24): 59–67,83. doi: 10.13465/j.cnki.jvs.2024.24.006
ZHENG Z, ZHAO W B, LI K, et al. Hydraulic pump fault diagnosis of small samples based on an improved multi-graph convolutional network[J]. Journal of Vibration and Shock, 2024, 43(24): 59–67,83 (in Chinese). doi: 10.13465/j.cnki.jvs.2024.24.006
|
[11] |
HOU B C, XIE M, PENG Z K, et al. Enspectrumix: novel adaptive methodology for fault component extraction from vibration mixtures[J]. Mechanical Systems and Signal Processing, 2024, 219: 111611. doi: 10.1016/j.ymssp.2024.111611
|
[12] |
朱仁杰, 宋恩哲, 姚崇, 等. 强噪声背景下基于CEEMDAN与BRECAN的船舶电机故障诊断[J]. 中国舰船研究, 2025, 20(2): 20–29. doi: 10.19693/j.issn.1673-3185.04059
ZHU R J, SONG E Z, YAO C, et al. Marine motor fault diagnosis based on CEEMDAN and BRECAN under strong noise conditions[J]. Chinese Journal of Ship Research, 2025, 20(2): 20–29 (in Chinese). doi: 10.19693/j.issn.1673-3185.04059
|
[13] |
DAI X, YI K, WANG F L, et al. Bearing fault diagnosis based on POA-VMD with GADF-Swin transformer transfer learning network[J]. Measurement, 2024, 238: 115328. doi: 10.1016/j.measurement.2024.115328
|
[14] |
SINGH P, JOSHI S D, PATNEY R K, et al. The Fourier decomposition method for nonlinear and non-stationary time series analysis[J]. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2017, 473(2199): 20160871. doi: 10.1098/rspa.2016.0871
|
[15] |
HE C, SHI H M, SI J, et al. Physics-informed interpretable wavelet weight initialization and balanced dynamic adaptive threshold for intelligent fault diagnosis of rolling bearings[J]. Journal of Manufacturing Systems, 2023, 70: 579–592. doi: 10.1016/j.jmsy.2023.08.014
|
[16] |
ZHANG X, HE C, LU Y P, et al. Fault diagnosis for small samples based on attention mechanism[J]. Measurement, 2022, 187: 110242. doi: 10.1016/j.measurement.2021.110242
|
[17] |
YAN S, SHAO H D, WANG J, et al. LiConvFormer: a lightweight fault diagnosis framework using separable multiscale convolution and broadcast self-attention[J]. Expert Systems with Applications, 2024, 237: 121338. doi: 10.1016/j.eswa.2023.121338
|