不同转速下基于深度注意力迁移学习的滚动轴承故障诊断方法

摘要
图/表
参考文献(16)
相关文章 (15)

全文: PDF (2482 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要针对不同转速下滚动轴承故障诊断中特征提取困难，数据分布差异巨大影响诊断结果，提出一种不同转速下基于深度注意力迁移学习（Deep Attention Transfer Learning，DATL）的滚动轴承故障诊断方法。首先，利用小波变换获取振动信号的时频图以展现时频特征；其次，搭建融合空间和通道的注意力卷积神经网络（Attention Convolutional Neural Network，ACNN）提取滚动轴承的关键特征以防止特征丢失；然后，在ACNN中添加领域适配层，完成不同转速下滚动轴承深度特征的迁移适配；最后，通过softmax分类层对目标数据进行故障状态识别。通过不同转速下不同故障程度的滚动轴承故障诊断实验证明了所提方法的可行性和有效性。

	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
	陈仁祥1
	唐林林1
	胡小林2
	杨黎霞3
	赵玲1

关键词 ：滚动轴承, 不同转速, 注意力机制, 迁移学习, 故障诊断

Abstract：Considering that it is difficult to extract features of rolling bearings at different speed, and the data distribution is very different, which affects the diagnosis results, a rolling bearing fault diagnosis method based on deep attention transfer learning at different rotating speeds was proposed. Firstly, the time-frequency diagram of vibration signal is obtained by wavelet transform to show the time-frequency characteristics. Secondly, an attention convolutional neural network (ACNN) based on space and channel is built as a feature extractor to extract the key features of the rolling bearing to prevent feature loss. Then, the domain adaptation layer is added to the ACNN to complete the distribution adaptation of rolling bearing depth features at different speeds. Finally, the health status of the target data was identified through the softmax classification layer. The feasibility and effectiveness of this method are proved by analyzing the experimental data of rolling bearing simulations with different degrees of failure at different speeds.

Key words： Rolling bearing Different speed Attention mechanism Transfer learning；Fault diagnosis

收稿日期: 2021-01-25 出版日期: 2022-06-28

引用本文:

陈仁祥1，唐林林1，胡小林2，杨黎霞3，赵玲1. 不同转速下基于深度注意力迁移学习的滚动轴承故障诊断方法[J]. 振动与冲击, 2022, 41(12): 95-101.
CHEN Renxiang1，TANG Linlin1，HU Xiaolin2，YANG Lixia3，ZHAO Ling1. A rolling bearing fault diagnosis method based on deep attention transfer learning at different rotations. JOURNAL OF VIBRATION AND SHOCK, 2022, 41(12): 95-101.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2022/V41/I12/95

[1] 陈仁祥, 陈思杨, 杨黎霞,等. 改进TrAdaBoost多分类算法的滚动轴承故障诊断[J]. 振动与冲击, 2019(15):36-41+48.
Chen Renxiang, Chen Siyang, Yang Lixia, et al. Fault diagnosis of rolling bearings based on improved TrAdaBoost multi-classification algorithm [J]. Journal of vibration and impact, 2019(15):36-41+48.
[2] YU J. A selective deep stacked denoising autoencoders ensemble with negative correlation learning for gearbox fault diagnosis[J]. Computers in Industry, 2019, 108:62-72.
[3] 沈长青, 汤盛浩, 江星星,等. 独立自适应学习率优化深度信念网络在轴承故障诊断中的应用研究[J]. 机械工程学报, 2019, 55(7):81-88.
SHEN Changqing, TANG Shenghao, JIANG Xingxing, et al. Bearings fault diagnosis based on improved deep belief network by self-individual adaptive learning rate[J]. Journal of Mechanical Engineering, 2019, 55(7): 81-88
[4] 曲建岭, 余路, 袁涛,等. 基于一维卷积神经网络的滚动轴承自适应故障诊断算法[J].仪器仪表学报,2018,39(07):134-143.
Qu Jianling, Yu Lu, Yuan Tao, et al. Adaptive fault diagnosis algorithm of rolling bearing based on one-dimensional convolutional neural network[J]. Chinese Journal of Scientific Instrument, 2018,39(07):134-143.
[5] Lei Y, Yang B, Jiang X, et al. Applications of machine learning to machine fault diagnosis: A review and roadmap[J]. Mechanical Systems and Signal Processing, 2020, 138: 106587.
[6] 王震, 黄如意, 李霁蒲,等. 一种用于故障分类与预测的多任务特征共享神经网络[J]. 仪器仪表学报, 2019,40(07):169-177.
Wang Zhen, Huang Ruyi, Li Jipu, et al. A multi-task feature sharing neural network for fault classification and prediction[J]. Chinese Journal of Scientific Instrument, 2019,40(07):169-177.
[7] 荆云建, 胡晓依, 宋志坤,等. 基于CNN-SVM的深度卷积神经网络轴承故障识别研究[J]. 振动与冲击, 2019, 38(18):173-178.
Jin Yunjian, Hu Xiaoyi, Song Zhikun, et al. Research on Bearing Fault Recognition of Deep Convolutional Neural Network Based on CNN-SVM[J]. Journal of vibration and impact, 2019, 38(18):173-178.
[8] 陈仁祥, 黄鑫, 杨黎霞,等. 基于卷积神经网络和离散小波变换的滚动轴承故障诊断[J]. 振动工程学报, 2018, 31(05):161-169.
Chen Renxiang, Huang Xing, Yang Lixia, et al. Fault diagnosis of rolling bearing based on convolutional neural network and discrete wavelet transform[J]. Journal of Vibration Engineering, 2018, 31(05):161-169.
[9] Voronin N A, Stroganova T A. A study of lateralized modules of visual spatial attention[J]. Voprosy Psikhologii, 2008(6):119-129+189.
[10] He Z, Chen X. Object tracking based on channel attention[J]. IEEE Access, 2020, (99):1-1.
[11] Pan S J, Yang Q. A Survey on Transfer Learning[J]. IEEE Transactions on Knowledge & Data Engineering, 2010, 22(10):1345-1359.
[12] 康守强, 胡明武, 王玉静,等. 基于特征迁移学习的变工况下滚动轴承故障诊断方法[J].中国电机工程学报,2019,39(03):764-772+955.
Kang Shouqiang, Hu Mingwu, Wang Yujing, et al. Fault Diagnosis Method of Rolling Bearing under Variable Conditions Based on Feature Transfer Learning[J]. China Journal of Electrical Engineering,2019,39(03):764-772+955.
[13] 郭亮, 董勋, 高宏力,等. 无标签数据下基于特征知识迁移的机械设备智能故障诊断[J]. 仪器仪表学报, 2019,40(08):58-64.
Guo Liang, Dong Xun, Gao Hongli, et al. Intelligent fault diagnosis of mechanical equipment based on feature knowledge transfer without label data[J]. Chinese Journal of Scientific Instrument,, 2019,40(08):58-64.
[14] LONG M, CAO Y, WANG J, et al. Learning transferable features with deep adaptation networks[J]. Proceedings of the 32nd International Conference on International Conference on Machine Learning, 2015, 37(7): 97–105.
[15] Laurens V D M, Hinton G. Visualizing Data using t-SNE[J]. Journal of Machine Learning Research, 2008, 9(2605):2579-2605.
[16] Hoult D I, Richards R E. The Signal-to-Noise Ratio of the Nuclear Magnetic Resonance Experiment[J]. Journal of Magnetic Resonance, 2011, 213(2):329-343.