基于迁移学习与深度残差网络的滚动轴承快速故障诊断算法

PDF(3545 KB)

振动与冲击 ›› 2022, Vol. 41 ›› Issue (3) : 154-164.

论文

刘飞1，陈仁文1，邢凯玲2，丁汕汕1，张迈一1

作者信息 +

Fast fault diagnosis algorithm for rolling bearing based on transfer learning and deep residual network

LIU Fei1, CHEN Renwen1, XING Kailing2, DING Shanshan1, ZHANG Maiyi1

Author information +

文章历史 +

摘要

针对现有基于深度学习的滚动轴承故障诊断算法训练参数量大，训练时间长且需要大量训练样本的缺点，提出了一种基于迁移学习（Transfer Learning，TL）与深度残差网络（Deep Residual Network，ResNet）的快速故障诊断算法（TL-ResNet）。首先开发了一种将短时傅里叶变换（STFT）与伪彩色处理相结合的振动信号转三通道图像数据的方法；然后将在ImageNet数据集上训练的ResNet18模型作为预训练模型，通过迁移学习的方法，应用到滚动轴承故障诊断领域当中；最后对滚动轴承在不同工况下的故障诊断问题，提出了采用小样本迁移的方法进行诊断。在凯斯西储大学（CWRU）与帕德博恩大学（PU）数据集上进行了实验，TL-ResNet的诊断准确率分别为99.8%与95.2%，且在CWRU数据集上TL-ResNet的训练时间仅要1.5s，这表明本算法优于其他的基于深度学习的故障诊断算法与经典算法，可用于实际工业环境中的快速故障诊断。

Abstract

Aiming at the shortcomings of the existing deep learning-based rolling bearing fault diagnosis algorithms that the training parameters are large, the training time is long, and a large number of training samples are required, a fast fault diagnosis algorithm (TL-ResNet) based on Transfer Learning (TL) and Deep Residual Network (ResNet) is proposed. First developed a method of converting vibration signals into three-channel image data by combining short-time Fourier transform (STFT) with pseudo-color processing; then the ResNet18 model trained on the ImageNet dataset is used as a pre-training model, and applied to the field of rolling bearing fault diagnosis through the method of transfer learning; finally, the method of small sample transfer is proposed for the fault diagnosis of rolling bearing under different working conditions. Experiments were performed on the Case Western Reserve University（CWRU）and Universität Paderborn（PU） datasets. The diagnosis accuracy of TL-ResNet is 99.8% and 95.2%, respectively, and the training time of TL-ResNet on the CWRU dataset is only 1.5s, which shows that this algorithm is superior to other deep learning-based fault diagnosis algorithms and classic algorithms, it can be used for rapid fault diagnosis in actual industrial environments.

导出引用

刘飞1，陈仁文1，邢凯玲2，丁汕汕1，张迈一1. 基于迁移学习与深度残差网络的滚动轴承快速故障诊断算法[J]. 振动与冲击, 2022, 41(3): 154-164

LIU Fei1, CHEN Renwen1, XING Kailing2, DING Shanshan1, ZHANG Maiyi1. Fast fault diagnosis algorithm for rolling bearing based on transfer learning and deep residual network[J]. Journal of Vibration and Shock, 2022, 41(3): 154-164

参考文献

[1]Austin H. Bonnett,Chuck Yung. Increased Efficiency Versus Increased Reliability[J]. IEEE Industry Applications Magazine,2008,14(1).
[2]周小龙,刘薇娜,姜振海,等.基于改进HHT和马氏距离的齿轮故障诊断[J].振动与冲击,2017,36(22):218-224.
ZHOU Xiaolong , LIU Weina , JIAN Zhanhai ,et al. Gear fault diagnosis based on improved HHT and Mahalanobis distance[J]. Journal of Vibration and Shock, 2017, 36(22):218-224.
[3]陈俊洵,程龙生,胡绍林,等.基于EMD的改进马田系统的滚动轴承故障诊断[J].振动与冲击,2017,36(05):151-156.
CHEN Junxun, CHENG Longsheng, HU Shaolin, et al. Fault diagnosis of rolling bearings using modified Mahalanobis-Taguchi system based on EMD[J]. Journal of Vibration and Shock, 2017,36(05):151-156.
[4] Xia M , Li T , Xu L , et al. Fault Diagnosis for Rotating Machinery Using Multiple Sensors and Convolutional Neural Networks[J]. IEEE/ASME Transactions on Mechatronics, 2017, PP(99):1-1.
[5]张龙,毛志德,杨世锡,等.基于包络谱带通峭度的改进谱峭度方法及在轴承诊断中的应用[J].振动与冲击, 2018, 37(23): 171-179.
ZHANG Long , MAO Zhide , YANG Shixi , et al. An improved Kurtogram based on band-pass envelope spectral Kurtosis with its application in bearing fault diagnosis[J]. Journal of Vibration and Shock, 2018, 37(23): 171-179.
[6]姜景升,王华庆,柯燕亮,等.基于LTSA与K-最近邻分类器的故障诊断[J].振动与冲击,2017,36(11):134-139.
JIANG Jingsheng, WANG Huaqing, KE Yanliang, et al. Fault Diagnosis Based on LTSA and K-Nearest Neighbor Classifier[J]. Journal of Vibration and Shock,2017,36(11):134-139.
[7]何大伟,彭靖波,胡金海,等.基于改进FOA优化的CS-SVM轴承故障诊断研究[J].振动与冲击,2018,37(18):108-114.
HE Dawei, PENG Jingbo, HU Jinhai et al. Bearing fault diagnosis based on a modified CS-SVM model optimized by an improved FOA algorithm[J]. Journal of Vibration and Shock, 2018,37(18):108-114.
[8]李锋,汤宝平,郭胤.Laplacian双联最小二乘支持向量机用于早期故障诊断[J].振动与冲击,2017,36(16):85-92.
LI Feng, TANG Baoping, GUO Yin. Early fault diagnosis using Laplacian twin least squares support vector machine[J]. Journal of Vibration and Shock,2017,36(16):85-92.
[9] Haidong S , Hongkai J , Ying L , 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(MAR.1):278-297.
[10] Sun W , Shao S , Zhao R , et al. A sparse auto-encoder-based deep neural network approach for induction motor faults classification[J]. Measurement, 2016:171-178.
[11] Ding X , He Q . Energy-Fluctuated Multiscale Feature Learning With Deep ConvNet for Intelligent Spindle Bearing Fault Diagnosis[J]. IEEE Transactions on Instrumentation and Measurement, 2017, PP(8):1-10.
[12] Zhao R , Wang D , Yan R , et al. Machine Health Monitoring Using Local Feature-based Gated Recurrent Unit Networks[J]. IEEE Transactions on Industrial Electronics, 2017:1-1.
[13] 雷亚国, 杨彬, 杜兆钧,等. 大数据下机械装备故障的深度迁移诊断方法[J]. 机械工程学报, 2019.
LEI Yaguo , YANG Bin , DU Zhaojun , et al. Deep Transfer Diagnosis Method for Machinery in Big Data Era[J]. Journal of Mechanical Engineering, 2019.
[14] Siyu S , Stephen M A , Ruqiang Y , et al. Highly Accurate Machine Fault Diagnosis Using Deep Transfer Learning[J]. IEEE Transactions on Industrial Informatics, 2019, PP:1-1.
[15] Wen L , Gao L , Li X . A New Deep Transfer Learning Based on Sparse Auto-Encoder for Fault Diagnosis[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2017:1-9.
[16] Shen F , Chen C , Yan R , et al. Bearing fault diagnosis based on SVD feature extraction and transfer learning classification[C]// Prognostics & System Health Management Conference. IEEE, 2016.
[17]李恒,张氢,秦仙蓉,等.基于短时傅里叶变换和卷积神经网络的轴承故障诊断方法[J].振动与冲击,2018,37(19):124-131.
LI Heng, ZHANG Qin, QIN Xianrong, et al. Fault diagnosis method for rolling bearings based on short-time Fourier transform and convolution neural network[J]. Journal of Vibration and Shock, ,2018,37(19):124-131.
[18] Zhao Z , Zhang Q , Yu X , et al. Unsupervised Deep Transfer Learning for Intelligent Fault Diagnosis: An Open Source and Comparative Study[J]. 2019.
[19] Ian Goodfellow, Yoshua Bengio, Aaron Courville. Deep Learning[M]. The MIT Press, 2016.
[20] He K , Zhang X , Ren S , et al. Deep Residual Learning for Image Recognition[C]// IEEE Conference on Computer Vision & Pattern Recognition. IEEE Computer Society, 2016.
[21] Pan S J , Yang Q . A Survey on Transfer Learning[J]. IEEE Transactions on Knowledge & Data Engineering, 2010, 22(10):1345-1359.
[22] Yosinski J , Clune J , Bengio Y , et al. How transferable are features in deep neural networks?[C]// International Conference on Neural Information Processing Systems. MIT Press, 2014.
[23] Zhao Z , Li T , Wu J , et al. Deep learning algorithms for rotating machinery intelligent diagnosis: An open source benchmark study[J]. ISA Transactions, 2020.
[24] Lessmeier C , Kimotho J K , Zimmer D , et al. Condition Monitoring of Bearing Damage in Electromechanical Drive Systems by Using Motor Current Signals of Electric Motors: A Benchmark Data Set for Data-Driven Classification[C]// European Conference of the Prognostics and Health Management Society. 2016.