Articles | Volume 15, issue 1
Research article
19 Feb 2024
Research article |  | 19 Feb 2024

A convolutional neural-network-based diagnostic framework for industrial bearing

Bowen Yu and Chunli Xie

Related subject area

Subject: Mechanisms and Robotics | Techniques and Approaches: Mathematical Modeling and Analysis
Design and analysis of a dual-rope crawler rope-climbing robot
Jinhang Wang, Lairong Yin, Ronghua Du, Long Huang, and Juan Huang
Mech. Sci., 15, 31–45,,, 2024
Short summary
Type synthesis of non-overconstrained and overconstrained two rotation and three translation (2R3T) parallel mechanisms with three branched chains
Yu Rong, Xingchao Zhang, Tianci Dou, and Hongbo Wang
Mech. Sci., 14, 567–577,,, 2023
Short summary
Motion planning and control strategy of a cable-driven body weight support gait training robot
Tao Qin, Qianpeng Wang, Wei Su, Chao Wei, Yanduo Zhang, and Jianwei Zhang
Mech. Sci., 14, 413–427,,, 2023
Short summary
Optimal resource allocation method and fault-tolerant control for redundant robots
Yu Rong, Tianci Dou, and Xingchao Zhang
Mech. Sci., 14, 399–412,,, 2023
Short summary
Ellipsoid contact analysis and application in the surface conjugate theory of face gears
Xiaomeng Chu, Yali Liu, and Hong Zeng
Mech. Sci., 14, 305–314,,, 2023
Short summary

Cited articles

Chen, W. and Shi, K.: Multi-scale Attention Convolutional Neural Network for time series classification, Neural Networks, 136, 126–140,, 2021. 
Chen, X., Zhang, B., and Gao, D.: Bearing fault diagnosis base on multi-scale CNN and LSTM model, J. Intell. Manuf., 32, 971–987,, 2021. 
Goodfellow, I., Bengio, Y., and Courville, A.: Deep learning, MIT press, ISBN 9780262035613, 2016. 
He, K., Zhang, X., Ren, S., and Sun, J.: Deep residual learning for image recognition, Proceedings of the IEEE conference on computer vision and pattern recognition, Las Vegas, Nevada, USA, 26 June–1 July 2016, IEEE, 770–778,, 2016.​​​​​​​ 
Hinton, G. E., Srivastava, N., Krizhevsky, A., Sutskever, I., and Salakhutdinov, R. R.: Improving neural networks by preventing co-adaptation of feature detectors, arXiv [preprint],, 3 July 2012. 
Short summary
In this study, we introduce an innovative diagnostic framework tailored to industrial bearings that facilitates automated feature extraction and enables end-to-end fault detection. Our validation across multiple bearing vibration datasets confirms the framework's superiority in handling complex and non-stationary signals in industrial environments.