• Login
  • Register
  • Search

Bearing Fault Diagnosis Based on Improved one-dimensional Convolutional Neural Network

Yiguang Wang, Ronaldo Juanatas, Jasmin Niguidula, Jonathan Caballero, Deye Jiang, Tao Ning

Abstract


Traditional bearing fault diagnosis mainly combines the characteristics of non-periodic, nonlinear and non-stationary vibration signals of bearing machinery, and uses one-dimensional convolutional neural network model for fault identifi cation and diagnosis. However, it relies too much on data collection, and has problems such as low diagnostic accuracy, poor diagnostic effectiveness and overfi tting of model, which urgently needs to be improved and optimized. Based on the classical one-dimensional convolutional neural network fault diagnosis method, this paper designs a new one-dimensional convolutional neural network model using Dropout technology. After adding the Dropout layer to the convolutional layer, the generalization ability of the bearing fault diagnosis model is greatly enhanced, and the probability of fault error diagnosis is reduced.

Keywords


One-dimensional convolutional neural network; Batch normalization layer; Bearing failure; Fault diagnosis

Full Text:

PDF

Included Database


References


[1] YU Miao. Application and Research of Improved Convolutional Neural Network in Bearing Fault diagnosis [D]. Dalian Jiaotong University, 2023.

[2] Wu Jigang, Wen Gang, Yang Kang. Aero engine fault diagnosis based on improved one-dimensional Convolutional neural Network [J]. Journal of Electronic Measurement and Instrumentation, 2023, 37 (03): 179-186. (in Chinese)

[3] HU Yu. Research on Bearing Fault Diagnosis Algorithm Based on one-dimensional Convolutional Neural Network [D]. Beijing University of Chemical Technology, 2022.

[4] Pan Yihao, Xiao Hong, Zhou Yubin, et al. Fault Diagnosis of Multi-axis Industrial Robot Based on Improved One-dimensional Convolutional Neural Network [J]. Modular Machine Tool & Automatic Processing Technology, 2021, (12): 10-14+18.

[5] Yang Yun, Ding Lei, ZHANG Haoyu. Research on Bearing Fault Diagnosis based on Improved one-dimensional Convolutional Neural Network and Bidirectional Gated cycle Unit [J]. Mechanical Science and Technology for Aerospace Engineering, 2023, 42 (04): 538-545.

[6] Cheng Liang, Dong Zijian, Wang Shumin, et al. Fault Diagnosis of Rolling Bearing Based on Improved One-dimensional Convolutional Neural Network [J]. Machine Design and Research, 2023, 39(3):126-130.

[7] Dong Shao-Jiang, PEI Xue-Wu, WU Wen-liang, et al. Rolling bearing fault Diagnosis Method Based on multi-layer noise reduction technology and improved Convolutional Neural Network [J]. Journal of Mechanical Engineering, 2021, 57(1):9.

[8] Meng Zhiqiang, Dong Shaojiang, Pan Xuejiao, et al. Fault Diagnosis of Rolling Bearing Based on Improved Convolutional Neural Network [J]. Modular Machine Tool & Automatic Machining Technology, 2020(2):5.

[9] Wang Qi, Deng Linfeng, Zhao Rongzhen. Fault Identification of Rolling Bearing based on improved one-dimensional Convolutional Neural Network [J]. Journal of Vibration and Shock, 2022, 41(3):8.

[10] Zhang Yuanxu, Cheng Changxin, Song Shengjian. Fault Diagnosis of Rolling Bearing Based on Improved RBF Neural Network [J]. Industrial Instrumentation & Automation, 2018(6):4.




DOI: http://dx.doi.org/10.18686/ahe.v7i36.12686

Refbacks
Copyright © 2020 Universe Scientific Publishing. All rights reserved.