基于改进深度残差网络的轴承故障诊断方法

摘要
图/表
参考文献(19)
相关文章 (15)

全文: PDF (1502 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要

针对滚动轴承在噪声环境中干扰大、工况复杂多变时诊断困难的问题，提出了一种改进深度残差网络的轴承故障诊断方法。对滚动轴承振动信号预处理，得到数据样本，分为训练集和测试集；将基于注意力机制的挤压与激励网络（squeeze-and-excitation networks，SENet）结构引入到残差神经网络残差块之中建立特征提取通道之间的联系，得到改进深度残差网络模型；再将标签化的训练集数据样本输入改进的诊断模型中进行训练；将训练好的诊断模型应用于测试集，输出每种故障的识别结果。在训练过程中，为了抑制过拟合，对原始训练样本进行加噪处理；同时引入了激活函数LReLU和Dropout技巧来提高模型的抗干扰能力。为了验证该模型的诊断性能，选用实验数据进行验证，结果表明该方法在载荷变化以及信号受到严重噪声污染时，依然拥有良好的故障诊断能力。

	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
	田科位
	董绍江
	姜保军
	裴雪武
	汤宝平
	胡小林
	赵兴新

关键词 ：滚动轴承, 轴承故障诊断, 深度残差网络, 挤压与激励网络

Abstract：A new fault diagnosis method based on an improved depth residual network was proposed to solve the problems of rolling bearings in noisy environment with large interference and complex and changeable working conditions.Firstly, the vibration signal of rolling bearing was preprocessed to obtain data samples, which were divided into training set and test set.The attention-mechanism-based Squeeze and Excitation Network (SENet) structure was introduced into the residual neural network residual block to establish the connection between the feature extraction channels.The improved deep residual network model was located there.The labeled training set data were input into the improved diagnostic model for training.Finally, the trained diagnosis model was applied to the test set to output the identification results of each fault.In order to suppress overfitting, the original training samples were denoised.Meanwhile, the activation function LReLU and Dropout technique were introduced to improve the anti-interference ability of the model.In order to verify the diagnostic performance of the model, experimental data were selected for verification.The results show that the method has good fault diagnosis capability when the load changes and the signal is seriously polluted by noise.

Key words： rolling bearing bearing fault diagnosis deep residual network extrusion and excitation network

收稿日期: 2020-07-13 出版日期: 2021-10-28

引用本文:

田科位，董绍江，姜保军，裴雪武，汤宝平，胡小林，赵兴新. 基于改进深度残差网络的轴承故障诊断方法[J]. 振动与冲击, 2021, 40(20): 247-254.
TIAN Kewei, DONG Shaojiang, JIANG Baojun, PEI Xuewu, TANG Baoping, HU Xiaolin, ZHAO Xingxin. A bearing fault diagnosis method based on an improved depth residual network. JOURNAL OF VIBRATION AND SHOCK, 2021, 40(20): 247-254.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2021/V40/I20/247

［1］JIA F, LEI Y G, LIN J, et al.Deep neural networks: A promising tool for fault characteristic mining and intelligent diagnosis of rotating machinery with massive data［J］.Mechanical Systems and Signal Processing, 2016, 72/73:303-315.
［2］SHAO H，JIANG H，LIN Y，et al.A novel method for intelligent fault diagnosis of rolling bearings using ensemble deep auto-encoders［J］.Mechanical Systems and Signal Processing, 2018,102:278-297.
［3］QU J, ZHANG Z, GONG T.A novel intelligent method for mechanical fault diagnosis based on dual-tree complex wavelet packet transform and multiple classifier fusion［J］.Neurocomputing, 2016, 171:837-853.
［4］ALI J B, FNAIECH N, SAIDI L, et al.Application of empirical mode decomposition and artificial neural network for automatic bearing fault diagnosis based on vibration signals［J］.Applied Acoustics, 2015, 89(3):16-27.
［5］史庆军，郭晓振，刘德胜.基于特征量融合和支持向量机的轴承故障诊断［J］.电子测量与仪器学报,2019(33)10:104-111.
SHI Qingjun, GUO Xiaozhen, LIU Desheng.Bearing fault diagnosis based on eigenvalue fusion and support vector machine ［J］.Journal of Electronic Measurement and Instrumentation, 2019,33(10):104-111.
［6］HE M, HE D.Deep learning based approach for bearing fault diagnosis［J］.IEEE Trans.Ind.Appl.,2017,53:3057-3065.
［7］ABDELJABER O, AVIC O, KIRANYAZ S, et al.Real-time vibration-based structural damage detection using one dimensional convolutional neural networks［J］.Journal of Sound and Vibration, Volume,2017,388:154-170.
［8］LIU H，ZHOU J，ZHENG Y，et al.Fault diagnosis of rolling bearings with recurrent neural network-based autoencoders［J］.ISA Trans,2018, 77:167-178.
［9］HAO S J, GE F X, LI Y M, et al.Multisensor bearing fault diagnosis based on one-dimensional convolutional long short-term memory networks［J］.Measurement,2020,159(5786):107802.
［10］HE K M, ZHANG X Y REN S Q, et al.Deep residual learning for image recognition［C］∥Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR).Piscataway, NJ, USA: IEEE, 2016.
［11］XU B, WANG N, CHEN T, et al.Empirical evaluation of rectified activations in convolution network［J］.I CML Deep Learn,2015:1-5
［12］HU J, LI S, GANG S.Squeeze-and-excitation networks［C］.In: Proceedings of the IEEE conference on computer vision and pattern recognition,2018.
［13］郭亮,高宏力,张一文,等.基于深度学习理论的轴承状态识别研究［J］.振动与冲击,2016,35(12):166-170.
GUO Liang, GAO Hongli, ZHANG Yiwen, et al.Research on bearing state recognition based on deep learning theory［J］.Journal of Vibration and Shock,2016,35(12):166-170.
［14］LIN M, CHEN Q, YAN S.Network in network［J］.arXiv:1312.4400, 2013.
［15］DONG S J, WU W L, HE K, et al.Rolling bearing performance degradation assessment based on improved convolutional neural network with anti-interference［J］.Measurement, 2020, 15:1-12.
［16］LOU X, LOPARO K A.Bearing fault diagnosis based on wavelet transform and fuzzy inference［J］.Mechanical Systems and Signal Processing, 2004, 18(5):1077-1095.
［17］ZHANG W, PENG G L, LI C H, et al.A deep convolutional neural network with new training methods for bearing fault diagnosis under noisy environment and different working load［J］.Mechanical Systems and Signal Processing,2018, 100:439-453.
［18］ZHANG W, PENG G L, LI C H, et al.A new deep learning model for fault diagnosis with good anti-noise and domain adaptation ability on raw vibration signals［J］.Sensors, 2017, 17(2):425.
［19］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(4):55-62.
YANG Ping, SU Yanchen, ZHANG Zhen.Research on rolling bearing fault diagnosis based on convolutional capsule network［J］.Journal of Vibration and Shock,2020,39(4):55-62.