一种改进原型网络的小样本轴承故障诊断方法

摘要
图/表
参考文献(22)
相关文章 (15)

全文: PDF (1584 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要原型网络小样本故障诊断中，由于故障样本少并且易受异常值的影响，导致所得类原型准确性不是很好。为了提高故障原型表示的准确性，本文提出一种改进原型网络的小样本故障诊断方法，通过引入辅助分类任务提取更具鲁棒性的特征，提高提取特征的区分能力。另外，利用查询集样本进一步优化类原型，提高类原型对故障轴承的表示能力。为验证所提方法的有效性，设置K取不同值，在滚动轴承数据集上进行C-way K-shot故障诊断实验。实验结果表明，改进原型网络所得类原型具有更好的区分性与准确性。在10-way 5-shot实验中，本文所提方法相较传统原型网络准确率提高了5.1%。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	赵志宏1
	2
	张然2
	刘克俭2
	杨绍普1

关键词 ：故障诊断, 小样本学习, 原型网络, 辅助任务

Abstract：In the fault diagnosis of prototype network with small samples, the accuracy of the prototype is not very good because of the small number of fault samples and the influence of outliers. In order to improve the accuracy of fault prototype representation, a small sample fault diagnosis method based on improved prototype network is proposed in this paper. By introducing an auxiliary classification task to extract more robust features, the distinguishing ability of extracted features is improved. In addition, the query set sample is used to further optimize the class prototype to improve the representation ability of the class prototype to the fault bearing. To verify the effectiveness of the proposed method, set K to different values, and conduct C-way K-shot fault diagnosis experiments on the rolling bearing data set. The experimental results show that the prototype obtained by the improved prototype network has better discrimination and accuracy. In the 10 way 5-shot experiment, the accuracy of the proposed method is 5.1% higher than that of the traditional prototype network.

Key words： Fault diagnosis Few-shot learning Prototype network Auxiliary task

收稿日期: 2022-06-15 出版日期: 2023-10-28

引用本文:

赵志宏1,2，张然2，刘克俭2，杨绍普1. 一种改进原型网络的小样本轴承故障诊断方法[J]. 振动与冲击, 2023, 42(20): 214-221.
ZHAO Zhihong1,2, ZHANG Ran2, LIU Kejian2, YANG Shaopu1. An improved prototype network method for small sample bearing fault diagnosis. JOURNAL OF VIBRATION AND SHOCK, 2023, 42(20): 214-221.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2023/V42/I20/214

[1] 雷亚国,贾峰,孔德同,等.大数据下机械智能故障诊断的机遇与挑战[J].机械工程学报,2018,54(05): 94-104.
LEI Ya-guo, JIA Feng, KONG De-tong, et al. A deep learning-based method for machinery health monitoring with big data[J]. Journal of Mechanical Engineering, 2018,54(05): 94-104.
[2] Goyal D, Choudhary A, Pabla B S, et al. Support vector machines based non-contact fault diagnosis system for bearings[J]. Journal of Intelligent Manufacturing, 2020,31: 1275-1289.
[3] Rafifiee J, Arvani F, Harififi A, et al. Intelligent condition monitoring of a gearbox using artifificial neural network[J]. Mechanical Systems and Signal Processing, 2007,21(4): 1746-1754.
[4] Shevchik S A, Saeidi F, Meylan B, et al. Prediction of failure in lubricated surfaces using acoustic time frequency features and random forest algorithm[J]. IEEE Transactions on Industrial Informatics, 2017,13(4): 1541-1553.
[5] Chen X, Zhang B, Gao D. Bearing fault diagnosis base on multi-scale CNN and LSTM model[J]. Journal of Intelligent Manufacturing, 2020,32: 971–987.
[6] 赵志宏,赵敬娇,魏子洋.基于BiLSTM的滚动轴承故障诊断研究[J].振动与冲击,2021,40(01):95-101.
ZHAO Zhi-hong, ZHAO Jing-jiao, WEI Zi-yang. Rolling bearing fault diagnosis based on BiLSTM network[J]. Journal of Vibration and Shock, 2021,40(01):95-101.
[7] 张鑫,郭顺生,李益兵,等.基于拉普拉斯特征映射和深度置信网络的半监督故障识别[J].机械工程学报,2020,56(01):69-81.
ZHANG Xin, GUO Shun-sheng, LI Yi-bing, et al. Semi-supervised Fault Identification Based on Laplacian Eigenmap and Deep Belief Networks[J].Journal of Mechanical Engineering,2020,56(01):69-81.
[8] 赵志宏,李乐豪,杨绍普,等. 一种频域特征提取自编码器及其在故障诊断中的应用研究[J]. 中国机械工程,2021,32(20):2468-2474.
ZHAO Zhi-hong, LI Le-hao, YANG Shao-pu, et al. A Frequency Domain Feature Extraction Auto-encoder and Its Application on Fault Diagnosis[J].China Mechanical Engineering,2021,32(20):2468-2474.
[9] 赵凯琳,靳小龙,王元卓. 小样本学习研究综述[J].软件学报,2021,32(02):349-369.DOI:10.13328/j.cnki.jos.006138.
ZHAO Kai-lin, JIN Xiao-long, WANG Yuan-zhuo. Survey on few-shot Learning[J].Journal of Software, 2021,32(02): 349-369. DOI:10.13328/j.cnki.jos.006138.
[10] 李凡长,刘洋,吴鹏翔,等. 元学习研究综述[J]. 计算机学报,2021,44(02):422-446.
Li Fan-chang, Liu Yang, Wu Peng-xiang, et al. A Survey on Recent Advances in Meta-Learning[J]. Chinese Journal of Computers, 2021,44(02):422-446.
[11] Snee J, Swersky K, Zemel R. Prototypical network for few-shot learing[J]. In:Proceedings of the Advances in Neural Information Processing System.Long Beach, USA:IEEE, 2017: 4077-4087.
[12] Yoon S W, Kim D Y, Seo J, et al. XtarNet: Learning to Extract Task-Adaptive Representation for Incremental Few-Shot Learning[C]//Proceedings of the 37th International Conference on Machine Learning. Vienna：PMLR，2020.
[13] Ji Z, Chai X, Yu Y, et al. Improved Prototypical Networks for Few-Shot Learning[J]. Pattern Recognition Letters, 2020, 140.
[14] Fritzler A, Logacheva V, Kretov M. Few-shot classification in named entity recognition task [C]// Proceedings of the 34th ACM/SIGAPP Symposium on Applied Computing. 2019: 993-1000.
[15] Jiang C, Chen H, Xu Q, et al. Few-shot fault diagnosis of rotating machinery with two-branch prototypical networks[J]. Journal of Intelligent Manufacturing, 2022: 1-15.
[16] Wang Y, Lei C, Yang L, et al. Wavelet-Prototypical Network Based on Fusion of Time and Frequency Domain for Fault Diagnosis[J]. Sensors, 2021, 21(4) : 1483-1483.
[17] 余浩帅，汤宝平，张楷，谭骞，魏静．小样本下混合自注意力原型网络的风电齿轮箱故障诊断方法[J]．中国机械工程,2021,32(20):2475-2481.
YU Hao-shuai, TANG Bao-ping, ZHANG Kai, TAN Qian, WEI Jing. Fault Diagnosis methtd of wind gearboxmixed with attention prototype network under small sample[J]． China Mechanical Engineering,2021,32(20): 2475-2481.
[18] Liu F, Li Q, Li M. Boosting Transferable Meta-learning via Unsupervised Visual Representation［EB/OL]. https://bcmi.sjtu.edu.cn/home/niuli/teaching/2020_final_5_classification.pdf.2020.
[19] 于俊杰,程华,房一泉. 少样本文本分类的多任务原型网络[J/OL]. 计算机应用研究:1-7[2022-01-19]. DOI:10.19734/j.issn.1001-3695.2021.11.0463.
YU Jun-jie, CHENG Hua, FANG Yi-quan. Multiple-task prototypical network for few-shot text classification[J/OL]. Application Research of Computers:1-7[2022-01-19]. DOI:10.19734/j.issn.1001-3695.2021.11.0463.
[20] Chen W, Chen J Q, Xian Y, T. Prototype Network for Text Entity Relationship Recognition in Metallurgical Field Based on Integrated Multi-class Loss Functions[J]. Journal of Intelligent & Fuzzy Systems, 2021, 40(6) : 12061-12073.
[21] Case Western Reserve University Bearing Data Center［EB/OL］．https://cse groups．case．edu/bearing data center/pages/download-data-file．2018.
[22]Huang H, Baddour N, Bearing vibration data collected under timevarying rotational speed conditions[J]. Data in Brief, 2018, 21: 1745-1749.