
浏览全部资源
扫码关注微信
重庆邮电大学 自动化学院,重庆 400065
胡向东,男,1971年生,四川广安人,博士,教授;主要研究方向为智能感知、网络化测量、工业互联网安全等;huxd@cqupt.edu.cn。
收稿:2025-10-11,
修回:2025-11-14,
网络首发:2026-10-29,
移动端阅览
胡向东,陈佳辉,白 银.基于多尺度元度量学习的小样本滚动轴承故障诊断方法[J].机械传动,XXXX,XX(XX):1-11.
HU Xiangdong,CHEN Jiahui,BAI Yin.A small⁃sample rolling bearing fault diagnosis method based on multi⁃scale meta⁃metric learning[J].Journal of Mechanical Transmission,XXXX,XX(XX):1-11.
目的
2
针对传统深度学习故障诊断方法依赖大量训练样本、难以适应工业现场轴承故障样本稀缺的问题,提出一种基于多尺度元度量学习的小样本滚动轴承故障诊断方法。
方法
2
首先,以信息熵为振动信号复杂度的量化指标,对采集信号进行定量评估与特征优化,以加速模型收敛并缓解由样本不平衡引发的过拟合问题;其次,引入多尺度时间卷积网络,捕捉不同尺度上更丰富的故障特征,克服局部与全局特征融合困难,增强特征表征能力;最后,设计基于改进皮尔逊相似度与三元组损失函数加权的度量学习模块,确保相似样本在特征空间的高度一致性,提高小样本特征的判别能力。
结果
2
基于公开数据集与实测数据集的验证结果表明,所提方法在两类数据集上的平均准确率分别达到99.03%、98.53%,优于其他对比算法。所提方法有效克服了小样本条件下的特征学习瓶颈,显著提升了滚动轴承故障的辨识能力与模型的泛化稳定性。
Objective
2
Aiming at the problem that traditional deep learning fault diagnosis methods rely on sufficient training samples and are difficult to adapt to industrial scenarios where bearing fault samples are scarce
a small-sample rolling bearing fault diagnosis method based on multi-scale meta-metric learning is proposed.
Methods
2
Firstly
information entropy was used as a quantitative index for vibration signal complexity to evaluate the collected signal segments and optimize signal features
which accelerated model convergence and alleviated the overfitting problem caused by sample imbalance. Secondly
a multi-scale temporal convolutional network was introduced to capture richer fault features at different scales
overcoming the fusion difficulties between local and global features and strengthening the feature characterization capability. Finally
a metric learning module weighted by improved Pearson similarity and a triplet loss function was designed to ensure high consistency of similar samples in the feature space and enhance the feature discrimination ability for small-sample learning.
Results
2
Validation results based on public and measured datasets show that the average accuracies of the proposed method are 99.03% and 98.53%
respectively
outperforming other comparative fault diagnosis methods. The proposed method effectively improves the fault identification capability and generalization stability of the model for rolling bearings under small-sample conditions.
Hu X S , Zhang K , Liu K L , et al . Advanced fault diagnosis for lithium-ion battery systems:a review of fault mechanisms,fault features,and diagnosis procedures [J]. IEEE Industrial Electronics Magazine , 2020 , 14 ( 3 ): 65 - 91 .
王波 , 宁毅 , 张亚虎 . 多核多分类相关向量机的滚动轴承多特征融合智能故障诊断方法 [J]. 机械科学与技术 , 2022 , 41 ( 6 ): 869 - 876 .
Wang Bo , Ning Yi , Zhang Yahu . Multi-kernel multi-class relevance vector machine and its application to fault diagnosis of rolling bearing with multi-feature fusion [J]. Mechanical Science and Technology for Aerospace Engineering , 2022 , 41 ( 6 ): 869 - 876 .
Dong Z L , Zhao D Z , Cui L L . An intelligent bearing fault diagnosis framework:one-dimensional improved self-attention-enhanced CNN and empirical wavelet transform [J]. Nonlinear Dynamics , 2024 , 112 ( 8 ): 6439 - 6459 .
Rafiee J , Arvani F , Harifi A , et al . Intelligent condition monitoring of a gearbox using artificial neural network [J]. Mechanical Systems and Signal Processing , 2007 , 21 ( 4 ): 1746 - 1754 .
李赟 , 于洋 , 杨平 . 改进Teager能量包络和EWT的轴承故障声发射诊断方法 [J]. 机械传动 , 2025 , 49 ( 9 ): 143 - 150 .
Li Yun , Yu Yang , Yang Ping . Acoustic emission diagnosis method of bearing faults based on improved Teager energy envelope and EWT [J]. Journal of Mechanical Transmission , 2025 , 49 ( 9 ): 143 - 150 .
刘敏 , 杨清清 , 涂君焕 , 等 . 基于CEEMD-MOMEDA-FCM的圆锥滚子轴承故障诊断 [J]. 机械传动 , 2026 , 50 ( 9 ): 164 - 173 .
Liu Min , Yang Qingqing , Tu Junhuan , et al . Fault diagnosis of tapered roller bearing based on CEEMD-MOMEDA-FCM [J]. Journal of Mechanical Transmission , 2026 , 50 ( 9 ): 164 - 173 .
李凡长 , 刘洋 , 吴鹏翔 , 等 . 元学习研究综述 [J]. 计算机学报 , 2021 , 44 ( 2 ): 422 - 446 .
Li Fanzhang , Liu Yang , Wu Pengxiang , et al . A survey on recent advances in meta-learning [J]. Chinese Journal of Computers , 2021 , 44 ( 2 ): 422 - 446 .
Li C J , Li S B , Wang H , et al . Attention-based deep meta-transfer learning for few-shot fine-grained fault diagnosis [J]. Knowledge-Based Systems , 2023 , 264 : 110345 .
Hu Y D , Liu R N , Li X L , et al . Task-sequencing meta learning for intelligent few-shot fault diagnosis with limited data [J]. IEEE Transactions on Industrial Informatics , 2022 , 18 ( 6 ): 3894 - 3904 .
Finn C , Abbeel P , Levine S . Model-agnostic meta-learning for fast adaptation of deep networks [EB/OL]. 2017 : arXiv : 1703 . 03400 . https://arxiv.org/abs/1703.03400 https://arxiv.org/abs/1703.03400 .
方桂花 , 杜壮 , 高旭 . 香农熵改进的变分模态分解与故障特征提取 [J]. 机械科学与技术 , 2020 , 39 ( 7 ): 1022 - 1027 .
Fang Guihua , Du Zhuang , Gao Xu . Shannon entropy improved variational mode decomposition and fault features extraction [J]. Mechanical Science and Technology for Aerospace Engineering , 2020 , 39 ( 7 ): 1022 - 1027 .
Zou D C , Wang S Z , Li X F , et al . MultiSPANS: a multi-range spatial-temporal transformer network for traffic forecast via structural entropy optimization [C]// Proceedings of the 17th ACM International Conference on Web Search and Data Mining . New York : ACM , 2024 : 1032 - 1041 .
Yang L , Jin R . Distance metric learning: A comprehensive survey [R]. East Lansing,MI : Michigan State University , 2006 : 1 - 51 .
Mahmoud R A . Automatic power factor correction based on Pearson similarity for distribution networks [J]. The Journal of Engineering , 2022 ,2022( 8 ): 798 - 831 .
Schroff F , Kalenichenko D , Philbin J . FaceNet:a unified embedding for face recognition and clustering [C]// 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . New York : IEEE , 2015 : 815 - 823 .
Zhang D M , Zheng K , Bai Y , et al . Few-shot bearing fault diagnosis based on meta-learning with discriminant space optimization [J]. Measurement Science and Technology , 2022 , 33 ( 11 ): 115024 .
Li F D , Chen J L , Pan J , et al . Cross-domain learning in rotating machinery fault diagnosis under various operating conditions based on parameter transfer [J]. Measurement Science and Technology , 2020 , 31 ( 8 ): 085104 .
Su H , Xiang L , Hu A J , et al . A novel method based on meta-learning for bearing fault diagnosis with small sample learning under different working conditions [J]. Mechanical Systems and Signal Processing , 2022 , 169 : 108765 .
Snell J , Swersky K , Zemel R S . Prototypical networks for few-shot learning [C]// Advances in Neural Information Processing Systems 30 . NewYork : Curran Associates,Inc. , 2017 : 4077 - 4087 .
Jiang C X , Chen H , Xu Q F , et al . Few-shot fault diagnosis of rotating machinery with two-branch prototypical networks [J]. Journal of Intelligent Manufacturing , 2023 , 34 ( 4 ): 1667 - 1681 .
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621