一种基于CNN-BiGRU孪生网络的轴承故障诊断方法

摘要
图/表
参考文献(27)
相关文章 (15)

全文: PDF (1212 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要针对轴承故障样本稀缺，传统深度神经网络模型在小样本情况下容易出现过拟合现象，泛化性能不好的问题，本文提出一种基于CNN-BiGRU孪生网络的轴承故障诊断方法。孪生网络采用两个结构相同、权值共享的卷积神经网络和双向门控循环单元组成，构造相同类别和不同类别的轴承样本对输入孪生网络，通过计算轴承样本对之间的L1距离进行相似性度量，实现轴承故障诊断。与传统深度神经网络相比，孪生网络采用样本对训练的方法，在相同样本数量情况下，增加对网络模型的有效训练次数，从而提高轴承故障诊断性能。设计卷积神经网络和双向门控循环单元共同组成孪生网络结构，可以从振动信号中同时提取空间特征与时序特征，提高特征提取的准确性。利用实测轴承故障信号进行故障诊断实验，并与其他深度神经网络模型进行对比，实验结果表明，CNN-BiGRU孪生网络方法在少量训练样本情况下，取得了较优的故障诊断性能，有一定的工程应用价值。

	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
	吴冬冬1
	窦广鉴1
	杨绍普2

关键词 ：孪生网络, 小样本, 故障诊断, 特征提取

Abstract：Aiming at the problems that the fault sample was scare and over-fitting in traditional deep neural network model in small samples and poor generalization performance, a fault diagnosis method based on CNN-BiGRU Siamese network was proposed. Siamese networks were composed of two convolution neural networks and bidirectional gated recurrent unit that had the same structure and shared weights, the bearing sample pairs of the same category and different categories were constructed to input the Siamese network and the similarity was compared based on the L1 distance to achieve fault classification. Compared with the traditional deep neural network, the Siamese network adopted the method of sample pair training, which increased the effective training times of the network model under the same number of samples, so as to improve the performance of bearing fault diagnosis. The convolution neural network and bidirectional gated recurrent unit were used to construct the Siamese network, which can extract spatial and temporal features from vibration signals at the same time, improved the accuracy of feature extraction. The fault diagnosis experiment was carried out by using the measured bearing fault signal, and compared with other deep neural network models. The experimental results show that the CNN-BiGRU Siamese network method still had superior diagnostic performance in the case of small samples and had a certain engineering application value.

Key words： Siamese network limited data bearing fault diagnosis feature extraction

收稿日期: 2021-11-22 出版日期: 2023-03-28

引用本文:

赵志宏1,2，吴冬冬1，窦广鉴1，杨绍普2. 一种基于CNN-BiGRU孪生网络的轴承故障诊断方法[J]. 振动与冲击, 2023, 42(6): 166-171.
ZHAO Zhihong1,2,WU Dongdong1,DOU Guangjian1,YANG Shaopu2. Bearing fault diagnosis method based on a CNN-BiGRU Siamese network. JOURNAL OF VIBRATION AND SHOCK, 2023, 42(6): 166-171.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2023/V42/I6/166

[1] ZHANG S, ZHANG S B, WANG B N, et al. Deep learning algorithms for bearing fault diagnosticsx a comprehensivereview. IEEE Access, 2020, 8: 29857–29881.
[2] Shenfield A, Howarth M. A Novel Deep Learning Model for the Detection and Identification of Rolling Element-Bearing Faults[J]. Sensors, 2020, 20(18)：450-460.
[3] Arcos B P, Bakker C, Flipsen B, et al. Practices of fault diagnosis in household appliances: Insights for design - ScienceDirect[J]. Journal of Cleaner Production, 2020,265:1-11.
[4] Jiang S, Lian M, Lu C, et al. SVM-DS fusion based soft fault detection and diagnosis in solar water heaters. Energy Exploration & Exploitation, 37(3):1125–1146.
[5] Rao K R，Kim D N，Hwang JJ. Fast fourier transform： algorithm and application[M]. China Machine Press,2013:41-45.
[6] 向玲,李媛媛.经验小波变换在旋转机械故障诊断中的应用[J].动力工程学报,2015,35(12):975-981.
Xiang Ling, Li Yuanyuan. Application of Empirical Wavelet Transform in Fault Diagnosis of Rotating Machinery[J]. Journal of Power Engineering, 2015, 35(12): 975-981.
[7] 孔德同,刘庆超,雷亚国,等.一种改进的EEMD方法及其应用研究[J].振动工程学报,2015,28(06):1015-1021.
Kong Detong, Liu Qingchao, Lei Yaguo, et al. improved EEMD method and its application research[J]. Journal of Vibration Engineering,2015,28(06):1015-1021.
[8] Hinton G E, Osindero S, Teh Y W. A Fast Learning Algorithm for Deep Belief Nets[J]. Neural Computation, 2014, 18(7):1527-1554.
[9] 雷亚国,贾峰,周昕,等.基于深度学习理论的机械装备大数据健康监测方法[J].机械工程学报,2015,51(21):49-56.
LEI Yaoguo, JIA feng, ZHOU Xin, et al. A Deep Learning-based Method for Machinery Health Monitoring with Big Data [J]. Journal of Mechanical Engineering, 2015, 51(21): 49-56.
[10] 周奇才,刘星辰,赵炯,等.旋转机械一维深度卷积神经网络故障诊断研究[J].振动与冲击,2018,37(23):31-37.
Zhou Qicai, Liu Xingchen, Zhao Jiong, et al. Research on the Fault Diagnosis of Rotating Machinery One-dimensional Deep Convolutional Neural Network[J].Journal of Vibration and Shock, 2018, 37 (23):31-37.
[11] 陈保家,刘浩涛,徐超,等.深度置信网络在齿轮故障诊断中的应用[J].中国机械程,2019,30 (02):205-211.
CHEN Baojia，LIU Haotao，XU Chao，et al．Gear fault diagnosis based on DBNS [J]．China Mechanical Engineering，2019，30( 2) : 205－211．
[12] 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 Zhihong, Zhao Jingjiao, Wei Ziyang. Research on Fault Diagnosis of Rolling Bearing Based on BiLSTM[J].Vibration and Shock,2021,40(01):95-101.
[13] 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.
Zhao Kailin, Jin Xiaolong, Wang Yuanzhuo.Review of Small Sample Learning Research[J].Journal of Software, 2021, 32(02):349-369.
[14] Wu. H, Xu. Z, Zhang. J, et al, Face recognition based on convolution siamese networks, 2017 10th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), 2017: 1-5.
[15] Bertinetto L, Valmadre J, Henriques J F, et al. Fully-Convolutional Siamese Networks for Object Tracking[J]. Springer, Cham, 2016,9914:850-865.
[16] Reimers N, Gurevych I. Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks[J]. 2019:3982 –3992.
[17] Wei P, Li H, Hu P. Inverse discriminative networks for handwritten signature verification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2019:5757-5765.
[18] Yuan J, Han G, Jin Z, et al.One-shot learning for fine-grained relation extraction via convolutional siamese neural network, 2017 IEEE International Conference on Big Data (Big Data), 2017:2194-2199
[19] 朱瑞金,郝东光,胡石峰.小样本条件下基于卷积孪生网络的变压器故障诊断[J].电力系统及其自动化学报,2021,33(01):64-69+84.
Zhu Ruijin, Hao Dongguang, Hu Shifeng.Transformer Fault Diagnosis Based on Convolutional Siamese Network under Small Sample Conditions[J].Journal of Electric Power Systems and Automation,2021,33(01):64-69+84.
[20] Zhang. A, Li. S, Cui. Y, et al. Limited Data Rolling Bearing Fault Diagnosis with Few-Shot Learning. IEEE Access, 2019: 110895–110904.
[21] Yang. Y, Wang. H, Liu. Z, et al. Few-shot Learning for Rolling Bearing Fault Diagnosis Via Siamese Two-dimensional Convolutional Neural Network. International Conference on Prognostics and System Health Management (PHM-2020 Jinan).2020:373-378.
[22] ROOPAK, SHAH, EDUARD, et al. SIGNATURE VERIFICATION USING A "SIAMESE" TIME DELAY NEURAL NETWORK[J]. International Journal of Pattern Recognition and Artificial Intelligence, 1993, 07(4):669-669.
[23] Chopra. S, Hadsell. R, LeCun. Y, Learning a similarity metric discriminatively, with application to face verification, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05), 2005,01:539-546.
[24] Hsiao S C, Kao D Y, Liu Z Y, et al. Malware Image Classification Using One-Shot Learning with Siamese Networks[J]. Procedia Computer Science, 2019,159:1863-1871.
[25] Lou. X, Loparo. K. A, Bearing fault diagnosis based on wavelet transform and fuzzy inference Mechanical Systems and Signal Processing. 2004,18(05): 1077-1095.
[26] 高浩寒,潮群,徐孜等.小样本下基于孪生神经网络的柱塞泵故障诊断[J/OL].北京航空航天大学报,2022-03-02:1-13.
Gao Haohan, Chao Qun, Xu Zi, et al. Mingyang,Liu Chengliang. Fault Diagnosis of Piston Pump Based on Twin Neural Network in Small Samples[J/OL].Journal of Beihang University: 2022-03-02:1-13..
[27] 杨平,苏燕辰,张振.基于卷积胶囊网络的滚动轴承故障诊断研究[J].振动与冲击,2020,39(04):55-62+68.
Yang Ping, Su Yanchen, Zhang Zhen. Research on Fault Diagnosis of Rolling Bearing Based on Convolutional Capsule Network[J].Journal of Vibration and Shock, 2020,39(04):55-62+68