Sub-domain adaptive unsupervised bearing fault diagnosis

PDF(2046 KB)

Journal of Vibration and Shock ›› 2021, Vol. 40 ›› Issue (15) : 34-40.

WU Jingran1,2, LIU Jianhua2, CUI Ran1,2

Author information +

History +

Abstract

Here, aiming at the problem of deep unsupervised bearing fault diagnosis network only aligning global distribution without considering the fine-grained information of each category in source domain and target domain, a sub-domain adaptive unsupervised end-to-end bearing fault diagnosis network was proposed. Firstly, 1-D convolutional neural network was used to do feature extraction, and multi-classification functions were used to construct classifiers. Then, by minimizing the local maximum average difference and loss function of classifier, the distribution of related sub-domains was aligned. Finally, the effectiveness of the proposed method was verified with the bearing fault data set of Jiangnan University. The results showed that the recognition accuracy of the proposed method is obviously higher than those of the other 5 popular domain adaptive fault diagnosis methods when the target domain data is unlabeled. The t-distributed random neighbor embedding results showed that the proposed method can effectively align category information of source domain and target domain; the feasibility and effectiveness of the proposed method are verified.

Key words

bearing fault diagnosis / transfer learning / unsupervised domain adaptation

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

WU Jingran, LIU Jianhua, CUI Ran. Sub-domain adaptive unsupervised bearing fault diagnosis[J]. Journal of Vibration and Shock, 2021, 40(15): 34-40

References

［1］李涛,段礼祥,张东宁, 等.自适应卷积神经网络在旋转机械故障诊断中的应用［J］.振动与冲击,2020,39(16):275-282.
LI Tao, DUAN Lixiang, ZHANG Dongning, et al. Application of adaptive convolutional neural network in rotating machinery fault diagnosis ［J］. Journal of Vibration and Shock, 2020,39(16):275-282.
［2］吴静然, 丁恩杰, 崔冉,等. 采用多尺度注意力机制的旋转机械故障诊断方法［J］. 西安交通大学学报, 2020(2):51-58.
WU Jingran, DING Enjie, CUI Ran, et al. A diagnostic approach for rotating machinery using multi-scale feature attention-based mechanism ［J］. Journal of Xi’an Jiaotong University, 2020(2):51-58.
［3］雷亚国, 贾峰, 孔德同, 等. 大数据下机械智能故障诊断的机遇与挑战［J］. 机械工程学报, 2018, 54(5): 94-104.
LEI Yaguo, JIA Feng, KONG Detong, et al. Opportunities and challenges of machinery intelligent fault diagnosis in big data era［J］. Journal of Mechanical Engineering, 2018, 54(5): 94-104.
［4］徐卓飞,刘凯,张海燕, 等. 基于经验模式分解和主元分析的滚动轴承故障诊断方法研究［J］.振动与冲击,2014,33(23):133-139.
XU Zhuofei,LIU Kai, ZHANG Haiyan, et al. A fault diagnosis method for rolling bearings based on empirical mode decomposition and principal component analysis ［J］. Journal of Vibration and Shock, 2014,33(23): 133-139.
［5］王新,闫文源. 基于变分模态分解和SVM的滚动轴承故障诊断［J］.振动与冲击,2017,36(18):252-256.
WANG Xin,YAN Wenyuan. Fault diagnosis of rolling element bearings based on EMD and MKD ［J］. Journal of Vibration and Shock, 2017,36(18):252-256.
［6］朱浩,宁芊,雷印杰, 等.基于注意力机制-Inception-CNN模型的滚动轴承故障分类［J］.振动与冲击,2020,39(19):84-93.
ZHU Hao,NING Qian，LEI Yinjie, et al. Fault diagnosis of rolling element bearings based on EMD and MKD ［J］. Journal of Vibration and Shock, 2020,39(19):84-93.
［7］
ZHANG Wei, LI Chuanhao, PENG Gaoliang, 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.
［8］LI X, ZHANG W, DING Q, et al. Multi-Layer domain adaptation method for rolling bearing fault diagnosis［J］. Signal Process, 2019,157: 180-197.
［9］LIU Han, ZHOU Jianzhong, XU Yanhe, et al. Unsupervised fault diagnosis of rolling bearings using a deep neural network based on generative adversarial networks［J］. Neurocomputing,2018,315:412-424.
［10］GUO L, LEI Y, XING S, et al. Deep convolutional transfer learning network: A new method for intelligent fault diagnosis of machines with unlabeled data［J］. IEEE Transactions on Industrial Electronics, 2019, 66(9):7316-7325.
［11］ZHU Y, ZHUANG F, WANG J, et al. Deep subdomain adaptation network for image classification［J］. IEEE Transactions on Neural Networks and Learning Systems, 2020(99):1-10.
［12］KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Imagenet classification with deep convolutional neural networks［C］. in: Proc Adv. Neural Inf. Process. Syst, 2012: 1097-1105.
［13］NAIR V, HINTON G E. Rectified linear units improve restricted boltzmann machines vinod nair［C］. International Conference on International Conference on Machine Learning. Omnipress, 2010:807-814.
［14］GOODFELLOW I, BENGIO Y, COURVILLE A. Deep learning［M］. The MIT Press, 2016.
［15］TZENG E, HOFFMAN J, ZHANG N, et al. Deep domain confusion: maximizing for domain invariance［OL］. arXiv preprint arXiv:14123474, 2014.
［16］ZHAO Z, ZHANG Q, YU X, et al. Unsupervised deep transfer learning for intelligent fault diagnosis: An Open Source and Comparative Study［OL］. Transfer Learning for PHM, 2019.
［17］HINTON G E, SRIVASTAVA N, KRIZHEVSKY A, et al. Improving neural networks by preventing co-adaptation of feature detectors［J］. Computer Science, 2012, 3(4): 212-223.
［18］李可.江南大学轴承故障数据集［EB/OL］. http:∥mad-net.org:8765/explore.html?t=0.%205831516555847212.
［19］SUN B, FENG J, SAENKO K. Return of frustratingly easy domain adaptation［C］. in: Proceedings of the Thirtieth AAAI Conference on ArtificialInte lligence,2016.
［20］LONG M, ZHU H, WANG J, et al. Deep transfer learning with joint adaptation networks［C］. in: Proceedingsofthe 34thInternational Conferenceon MachineLearning- Volume70,JMLR. org,2017:2208-2217.
［21］GANIN Y, USTINOVA E, AJAKAN H, et al. Domain-adversarial training of neural networks［J］. Journal of Machine Learning Research, 2017, 17(1):2096-2030. 
［22］LONG M, CAO Z, WANG J, et al. Conditional adversarial domain adaptation［C］. NeuralInf. Process. Syst, 2018:1640-1650. 
［23］LAURENS V D M, HINTON G. Visualizing data using t-SNE［J］. Journal of Machine Learning Research, 2008, 2605(9):2579-2605.