Intelligent fault diagnosis for rolling bearing based on improved convolutional neural network

被引：0
|
作者：
Gong W.-F. [1 ,2 ]
Chen H. [1 ]
Zhang Z.-H. [1 ]
Zhang M.-L. [2 ]
Guan C. [1 ]
Wang X. [2 ]
机构：
[1] Key Laboratory of High Performance Ship Technology of Ministry of Education in China, School of Energy and Power Engineering, Wuhan University of Technology, Wuhan
[2] Beihai Campus, Guilin University of Electronic and Technology, Beihai
来源：
Zhendong Gongcheng Xuebao/Journal of Vibration Engineering | 2020年 / 33卷 / 02期
关键词：
Convolutional neural network; Deep learning; Fault diagnosis; Global average pooling; Rolling bearing;
D O I：
10.16385/j.cnki.issn.1004-4523.2020.02.021
中图分类号：
学科分类号：
摘要：
Convolutional neural network (CNN) has gradually become a research hotspot in the field of intelligent fault diagnosis. The full connection layer structure of the traditional CNN model has a problem of excessive training parameters, which makes the network training and testing time longer. To solve this problem, a novel method called improved CNN is proposed for the rapid fault diagnosis of rolling bearing. The proposed method improves the traditional CNN model structure by introducing the global average pooling technology to replace the fully connected network structure. The proposed method can effectively reduce the model training parameter quantity and test time. The deep learning training techniques are applied, such as data enhancement, dropout, and adaptive learning rate, etc. The proposed method is applied to the diagnosis of rolling bearing experimental data sets and compared with traditional intelligent diagnosis methods. The results show that the accuracy of fault identification of improved CNN algorithm is 99.04%. The accuracy of incipient fault diagnosis and test time consumption are obviously better than traditional CNN and other intelligent algorithms. The proposed method does not need any manual feature extraction operations on the raw data during the whole fault diagnosis process. The end-to-end algorithm structure of proposed method has good operability and versatility. © 2020, Nanjing Univ. of Aeronautics an Astronautics. All right reserved.
引用
收藏
页码：400 / 413
页数：13
相关论文
共 31 条
  • [11] Li H., Zhang Q., Qin X., Et al., Fault diagnosis method for rolling bearings based on short-time Fourier transform and convolution neural network, Journal of Vibration and Shock, 37, 19, pp. 124-131, (2018)
  • [12] Jegadeeshwaran R., Sugumaran V., Fault diagnosis of automobile hydraulic brake system using statistical features and support vector machines, Mechanical Systems and Signal Processing, 52-53, 1, pp. 436-446, (2015)
  • [13] Yang Y., Yu D., Cheng J., A roller bearing fault diagnosis method based on EMD energy entropy and ANN, Journal of Sound & Vibration, 294, 1, pp. 269-277, (2006)
  • [14] Verstraete D., Ferrada A., Droguett E.L., Et al., Deep learning enabled fault diagnosis using time-frequency image analysis of rolling element bearings, Shock and Vibration, 2017, pp. 1-17, (2017)
  • [15] Hinton G.E., Salakhutdinov R.R., Reducing the dimensionality of data with neural networks, Science, 313, 5786, pp. 504-507, (2006)
  • [16] Noda K., Yamaguchi Y., Nakadai K., Et al., Audio-visual speech recognition using deep learning, Applied Intelligence, 42, 4, pp. 722-737, (2015)
  • [17] Gao Z., Research and application of image classification method based on deep convolutional neural network, (2018)
  • [18] Young T., Hazarika D., Poria S., Et al., Recent trends in deep learning based natural language processing, IEEE Computational Intelligence Magazine, 13, 3, pp. 55-75, (2017)
  • [19] Jia F., Lei Y., Lin J., Et al., Deep neural networks: A promising tool for fault characteristic mining and intelligent diagnosis of rotating machinery with massive data, Mechanical Systems & Signal Processing, 72-73, pp. 303-315, (2016)
  • [20] Tamilselvan P., Wang P., Failure diagnosis using deep belief learning based health state classification, Reliability Engineering & System Safety, 115, 7, pp. 124-135, (2013)
1234