2.624

2020影响因子

(CJCR)

  • 中文核心
  • EI
  • 中国科技核心
  • Scopus
  • CSCD
  • 英国科学文摘

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于关系网络的轴承剩余使用寿命预测方法

赵志宏 张然 孙诗胜

赵志宏, 张然, 孙诗胜. 基于关系网络的轴承剩余使用寿命预测方法. 自动化学报, 2022, 45(x): 1−9 doi: 10.16383/j.aas.c211195
引用本文: 赵志宏, 张然, 孙诗胜. 基于关系网络的轴承剩余使用寿命预测方法. 自动化学报, 2022, 45(x): 1−9 doi: 10.16383/j.aas.c211195
Zhao Zhi-Hong, Zhang Ran, Sun Shi-Sheng. Bearing remaining useful life prediction based on relation network. Acta Automatica Sinica, 2022, 45(x): 1−9 doi: 10.16383/j.aas.c211195
Citation: Zhao Zhi-Hong, Zhang Ran, Sun Shi-Sheng. Bearing remaining useful life prediction based on relation network. Acta Automatica Sinica, 2022, 45(x): 1−9 doi: 10.16383/j.aas.c211195

基于关系网络的轴承剩余使用寿命预测方法

doi: 10.16383/j.aas.c211195
基金项目: 国家自然科学基金 (11972236,11790282), 石家庄铁道大学研究生创新资助项目(YC2022056)资助
详细信息
    作者简介:

    赵志宏:石家庄铁道大学教授. 2012年获得北京交通大学博士学位. 主要研究方向为机械故障诊断, 机器学习, 信号处理和动力学分析.E-mail: hb_zhaozhihong@126.com

    张然:石家庄铁道大学信息科学与技术学院硕士研究生. 主要研究方向为故障诊断, 状态评估与预测和大数据分析. 本文通信作者. E-mail:sjz_zhangran@126.com

    孙诗胜:石家庄铁道大学信息科学与技术学院硕士研究生. 主要研究方向为故障诊断, 状态评估与预测和大数据分析. E-mail:lxr_sunshisheng@126.com

Bearing remaining useful life prediction based on relation network

Funds: Supported by National Natural Science Foundation of China(11972236、11790282), Graduate Innovation Funding Project of Shijiazhuang Tiedao University(YC2022056)
More Information
    Author Bio:

    ZHAO Zhi-Hong Professor at Shijiazhuang Tiedao University. He received his ph. D. degree from Beijing Jiaotong University in 2012. His research interest diagnosis of mechanical equipment,machine learning, signal processing, and dynamic analysis

    ZHANG Ran Master student at the College of Information Science and Technology, Shijiazhuang Tiedao University. Her research interest covers fault diagnosis, state assessment and prediction, and big data analysis.Corresponding author of this paper

    SUN Shi-Sheng Master student at the College of Information Science and Technology, Shijiazhuang Tiedao University. His research interest covers fault diagnosis, state assessment and prediction, and big data analysis

  • 摘要: 针对轴承全寿命周期数据获取困难、训练样本少的问题, 提出一种基于关系网络的轴承剩余使用寿命(Remaining useful life,RUL)预测方法. 关系网络是一种基于度量的元学习方法, 在少量训练样本下, 具有快速学习新任务的优点. 设计了一种基于关系网络的轴承健康评估模型, 利用关系网络的嵌入模块提取轴承状态特征, 利用关系模块度量轴承状态特征之间的相似性, 基于相似性构建轴承健康指标;对健康指标进行Savitzky-golay滤波平滑处理, 降低振荡对预测结果的影响;最后利用线性函数对健康指标进行拟合, 得到轴承RUL预测值. 为验证所提方法的有效性, 在PHM2012轴承实测数据集上进行实验. 结果表明所得健康指标能够反映轴承的退化趋势, 所得RUL预测结果与ConvLSTM、Transformer、RNN、LSTM、Attention mechanism方法相比, 误差百分比分别减少了1.68%、3.41%、9.03%、13.72%、30.49%. 方法在少量训练样本的基础上可以取得较好的预测结果, 具有一定的应用价值.
  • 图  1  关系网络结构

    Fig.  1  Relation network structure

    图  2  关系网络模型结构

    Fig.  2  Structure of relational network model

    图  3  基于关系网络的轴承RUL预测流程

    Fig.  3  Bearing RUL prediction process based on relation network

    图  4  PRONOSTIA实验台

    Fig.  4  PRONOSTIA test bench

    图  5  轴承时域图

    Fig.  5  Bearing time domain diagram

    图  6  训练集轴承健康指标

    Fig.  6  Training set bearing health indicators

    图  7  轴承健康指标

    Fig.  7  Bearing health indicators

    图  8  轴承RUL预测结果

    Fig.  8  Bearing RUL prediction results

    表  1  PHM2012轴承数据集

    Table  1  PHM2012bearing dataset

    工况 工况一 工况二 工况三
    训练集 轴承1_1 轴承2_1 轴承3_1
    轴承1_2 轴承2_2 轴承3_2
    轴承1_5 轴承2_4
    轴承1_6 轴承2_5
    轴承1_7 轴承2_7
    测试集 轴承1_3 轴承2_3轴承3_3
    轴承1_4 轴承2_6
    下载: 导出CSV

    表  2  不同模型参数量对比

    Table  2  Comparison of different model parameters

    方法 本文方法 ConvLSTM Transformer CNN+LSTM
    参数量 78.609k 220.497k 6.310M 1.110M
    下载: 导出CSV

    表  3  轴承RUL预测结果

    Table  3  Bearing RUL prediction results

    轴承 当前时刻(10s)真实寿命(10s)预测寿命(10s)本文方法(10s)文献[13](%) 文献[28](%) 文献[11](%) 文献[29](%) 文献[12](%)
    轴承1_3 1801 573 937 −63.52 33.68 74.17 74.17 54.73 7.62
    轴承1_4 1138 290 338 −16.55 47.24 −0.69 −0.69 38.69 −157.71
    轴承2_3 1201 753 1005 −33.46 −32.80 61.36 61.36 75.53 81.24
    轴承2_6 571 129 131 −1.55 8.52 0.78 0.78 17.87 24.92
    轴承3_3 351 82 77 6.09 7.32 1.22 1.22 2.93 2.09
    平均误差 24.23 25.91 27.64 33.26 37.95 54.72
    下载: 导出CSV
  • [1] 雷亚国, 贾峰, 孔德同, 林京, 邢赛博. 大数据下机械智能故障诊断的机遇与挑战. 机械工程学报, 2018, 54(5): 94-104 doi: 10.3901/JME.2018.05.094

    Lei Ya-Guo, Jia Feng, Kong De-Tong, Lin Jing, Xing Sai-Bo. A deep learning-based method for machinery health monitoring with big data. Journal of Mechanical Engineering, 2018, 54(5): 94-104 doi: 10.3901/JME.2018.05.094
    [2] 赵志宏, 李乐豪, 杨绍普, 李晴. 一种无监督的轴承健康指标及早期故障检测方法. 中国机械工程: 1-11[2022-03-15]. http://kns.cnki.net/kcms/detail/42.1294.TH.20210915.132-8.002.html

    Zhao Zhi-Hong, Li Le-Hao, Yang Shao-Pu, Li Qing. An unsupervised bearing health indicator and early fault detection method. China Mechanical Engineering: 1-11[2022-03-15]. http://kns.cnki.net/kcms/detail/42.1294.TH.20210915.132-8.002.html
    [3] 刘建昌, 权贺, 于霞, 何侃, 李镇华. 基于参数优化VMD和样本熵的滚动轴承故障诊断. 自动化学报: 1-12[2022-03-15]. DOI: 10.16383/j.aas.c190345

    Liu Jian-Chang, Quan He, Yu Xia, He Kan, Li Zhen-Hua. Rolling bearing fault diagnosis based on parameter optimization VMD and sample entropy. Acta Automatica Sinica: 1-12[2022-03-15]. DOI: 10.16383/j.aas.c190345
    [4] 康守强, 邢颖怡, 王玉静, 王庆岩, 谢金宝, V.I.MIKULOVICH. 基于无监督深度模型迁移的滚动轴承寿命预测方法. 自动化学报: 1-11[2022-03-15] DOI: 10.16383/j.aas.c200890

    Kang Shou-Qiang, Xing Ying-Yi, Wang Yu-Jing, Wang Qing-Yan, Xie Jin-Bao, V.I.MIKULOVICH. Rolling bearing life prediction based on unsupervised deep model transfer. Acta Automatica Sinica: 1-11[2022-03-15]. DOI: 10.16383/j.aas.c200890
    [5] Jouin M, Gouriveau R, Hissel D, Péra M C, Zerhouni N. Particle filter based prognostics: Review, discussion and perspectives. Mechanical Systems and Signal Processing, 2016, 72-73:2-31 doi: 10.1016/j.ymssp.2015.11.008
    [6] Jouin M, Gouriveau R, Hissel D, Péra M C, Zerhouni N. Degradations analysis and aging modeling for health assessment and prognostics of PEMFC. Reliability Engineering & System Safety, 2016, 148:78-95
    [7] Ali J B, Chebel-Morello B, Saidi L, Malinowski S, Fnaiech F. Accurate bearing remaining useful life prediction based on weibull distribution and artificial neural network. Mechanical Systems and Signal Processing, 2015, 56-57:150-172 doi: 10.1016/j.ymssp.2014.10.014
    [8] Yang H B, Sun Z, Jiang G D, Fei Z, Mei X. Remaining useful life prediction for machinery by establishing scaled-corrected health indicators. Measurement, 2020, 163: Article No.108035
    [9] Hong S, Zhou Z, Zio E, Hong K. Condition assessment for the performance degradation of bearing based on a combinatorial feature extraction method. Digital Signal Processing, 2014, 27:159-166 doi: 10.1016/j.dsp.2013.12.010
    [10] Hinton G E, Osindero S, Teh Y W. A fast learning algorithm for deep belief nets. Neural Computation, 2006, 18(7): 1527-1554 doi: 10.1162/neco.2006.18.7.1527
    [11] Guo L, Li N P, Jia F, Lei Y G, Jing L. A recurrent neural network based health indicator for remaining useful life prediction of bearings. Neurocomputing, 2017, 240:98-109 doi: 10.1016/j.neucom.2017.02.045
    [12] Chen Y H, Peng G L, Zhu Z Y, Li S J. A novel deep learning method based on attention mechanism for bearing remaining useful life prediction. Applied Soft Computing, 2019, 86(C): Article No.105919
    [13] 王久健, 杨绍普, 刘永强, 文桂林. 一种基于空间卷积长短时记忆神经网络的轴承剩余寿命预测方法. 机械工程学报, 2021, 57(21):88-95 doi: 10.3901/JME.2021.21.088

    Wang Jiu-Jian, Yang Shao-Pu, Liu Yong-Qiang, Wen Gui-Lin. A method of bearing remaining useful life estimation based on convolutional long short-term memory neural network. Journal of Mechanical Engineering, 2021, 57(21):88-95 doi: 10.3901/JME.2021.21.088
    [14] 康守强, 周月, 王玉静, 谢金宝, V.I.MIKULOVICH. 基于改进SAE 和双向LSTM的滚动轴承RUL预测方法. 自动化学报: 1-11[2022-03-15]. DOI: 10.16383/j.aas.c190796

    Kang Shou-Qiang, Zhou Yue, Wang Yu-Jing, Xie Jin-Bao, V.I.MIKULOVICH. RUL Prediction method of a rolling bearing based on improved SAE and Bi-LSTM. Acta Automatica Sinica: 1-11[2022-03-15]. DOI: 10.16383/j.aas.c190796
    [15] 李凡长, 刘洋, 吴鹏翔, 董方, 蔡奇, 王哲. 元学习研究综述. 计算机学报, 2021, 44(2):422-446 doi: 10.11897/SP.J.1016.2021.00422

    Li Fan-Chang, Liu Yang, Wu Peng-Xiang, Dong Fang, Cai Qi, Wang Zhe. A survey on recent advances in meta-learning. Chinese Journal of Computers, 2021, 44(2):422-446 doi: 10.11897/SP.J.1016.2021.00422
    [16] Koch G, Zemel R, Salakhutdinov R. Siamese neural networks for one-shot image recognition.In: Proceedings of the 2015 ICML Deep Learning Workshop. Lille, France: IEEE, 2015.2
    [17] Vinyals O, Blundell C, Lillicrap T, Kavukcuoglu K, Wierstra D. Matching networks for one shot learning. In: Proceedings of the 2016 Advances in Neural Information Processing System. Barcelona, Spain: IEEE, 2016.3630−3638
    [18] Snee J, Swersky K, Zemel R. Prototypical network for few-shot learing. In: Proceedings of the 2017 Advances in Neural Information Processing System.Long Beach, USA: IEEE, 2017. 4077−4087
    [19] Sung F, Yang Y, Li Z, Tao X, Hospedales T M. Learning to compare: Relation network for few-shot learning. In: Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA: IEEE, 2018.1199−1208
    [20] Bishay M, Zoumpourlis G, Patras I. TARN: Temporal attentive relation network for few-shot and zero-shot action recognition. arXiv preprint arXiv: 1907.09021, 2019
    [21] 王晓茹, 张珩. 基于注意力机制和图卷积的小样本分类网络. 计算机工程与应用, 2021, 57(19):164-170 doi: 10.3778/j.issn.1002-8331.2104-0275

    Wang Xiao-Ru, Zhang Heng. Relation network based on attention mechanism and graph convolution for few-shot learning. Computer Engineering and Applications, 2021, 57(19):164-170 doi: 10.3778/j.issn.1002-8331.2104-0275
    [22] Wu J Y, Zhao Z B, Sun C, Yan R Q, Chen X F. Few-shot transfer learning for intelligent fault diagnosis of machine. Measurement, 2020, 166: Article No.108202
    [23] 吕枫, 王义, 阮胡林, 秦毅, 王平. 深度嵌入关系空间下齿轮箱标记样本扩充及其半监督故障诊断方法. 仪器仪表学报, 2021, 42(2):55-65

    Lv Feng, Wang Yi, Ruan Hu-lin, Qin Yi, Wang Ping. Labeled sample augmentation based on deep embedding relation space for semi-supervised fault diagnosis of gearbox. Chinese Journal of Scientific Instrument, 2021, 42(2):55-65
    [24] Savitzky A, Golay M J. Smoothing and differentiation of data by simplified least squares procedures. Analytical Chemistry. 2002, 36(8) : 1627-1639
    [25] Nectoux P, Gouriveau R, Medjaher K, Ramasso E, Morello B, Zerhouni N, et al. PRONOSTIA: An ex-perimental platform for bearings accelerated degradation tests. IEEE International Conference on Prognostics and Health M-anagement. USA: Denver, 2012: 1−8
    [26] Soualhi A, Medjaher K, Zerhouni N.Bearing health monitoring based on hilbert–huang transform, support vector machine, and regression. IEEE Transactions on Instrumentation and Measurement. 2015, 64(1):52-62 doi: 10.1109/TIM.2014.2330494
    [27] Singleton R K, Strangas E G, Aviyente S. Extended kalman filtering for remaining-useful-life estimation of bearings. IEEE Transactions on Industrial Electronics. 2015, 62(3):1781-1790 doi: 10.1109/TIE.2014.2336616
    [28] 周哲韬, 刘路, 宋晓, 陈凯. 基于Transformer模型的滚动轴承剩余使用寿命预测方法. 北京航空航天大学学报: 1-17[2022-03-15]. DOI: 10.13700/j.bh.1001-5965.2021.0247

    Zhou Zhe-tao, Liu Lu, Song Xiao, Chen Kai. Remaining useful life prediction method of rolling bearing based on transformer model. Journal of Beijing University of Aeronautics and Astronautics: 1-17[2022-03-15]. DOI: 10.13700/j.bh.1001-5965.2021.0247
    [29] Hinchi A Z, Tkiouat M. Rolling element bearing remaining useful life estimation based on a convolutional long-short-term memory network. Procedia Computer Science, 2018, 127:123-132 doi: 10.1016/j.procs.2018.01.106
  • 加载中
计量
  • 文章访问数:  108
  • HTML全文浏览量:  22
  • 被引次数: 0
出版历程
  • 收稿日期:  2021-12-15
  • 录用日期:  2022-03-13
  • 网络出版日期:  2022-05-30

目录

    /

    返回文章
    返回