Fault Diagnosis of Rolling Bearing with Roller Spalling Based on Two-Step Transfer Learning on Unbalanced Dataset

doi:10.16183/j.cnki.jsjtu.2022.008

Abstract

Abstract:

Under operating conditions, bearings have a substantial service life with short failure time periods, which leads to unbalanced dataset and greatly affects the accuracy of deep learning model fault diagnosis. To address this problem, a fault diagnosis method of rolling bearing unbalanced dataset based on two-step transfer learning is proposed in this paper. First, a small amount of data in the source and target domains is used to generate the transition dataset by conditional gradient penalized generative adversarial network (CWGAN-GP). Then, the constructed convolutional neural network model is migrated twice between the source domain dataset, the transition dataset, and the target domain dataset. Finally, a small amount of data from the target domain is used to fine-tune the transferred model to obtain the final fault diagnosis model. The experimental results show that the method has a good diagnostic recognition effect on rolling bearing spalling class faults with unbalanced dataset under different working conditions.

Key words: transfer learning, fault diagnosis, unbalanced dataset, generative adversarial network

CLC Number:

TH17

GUO Junfeng, WANG Miaosheng, WANG Zhiming. Fault Diagnosis of Rolling Bearing with Roller Spalling Based on Two-Step Transfer Learning on Unbalanced Dataset[J]. Journal of Shanghai Jiao Tong University, 2023, 57(11): 1512-1521.

Figures/Tables 12

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Tab.1

Fig.8

Fig.9

Fig.10

Fig.11

References 20

[1]	张根保, 李浩, 冉琰, 等. 一种用于轴承故障诊断的迁移学习模型[J]. 吉林大学学报(工学版), 2020, 50(5): 1617-1626.
	ZHANG Genbao, LI Hao, RAN Yan, et al. A transfer learning model for bearing fault diagnosis[J]. Journal of Jilin University (Engineering and Technology Edition), 2020, 50(5): 1617-1626.
[2]	周生通, 朱经纬, 周新建, 等. 组合载荷作用下动车牵引电机转子系统弯扭耦合振动特性[J]. 交通运输工程学报, 2020, 20(1): 159-170.
	ZHOU Shengtong, ZHU Jingwei, ZHOU Xinjian, et al. Bending-torsional coupling vibration characteristics of EMU traction motor rotor system under combined loads[J]. Journal of Traffic and Transportation Engineering, 2020, 20(1): 159-170.
[3]	韩毅. 地铁车辆滚动轴承振动信号的时域分析[J]. 城市轨道交通研究, 2021, 24 (Sup.1): 57-62.
	HAN Yi. Time domain analysis of vibration signal for metro vehicle rolling bearing[J]. Urban Mass Transit, 2021, 24 (Sup.1): 57-62.
[4]	李舜酩, 侯钰哲, 李香莲. 滚动轴承振动故障时频域分析方法综述[J]. 重庆理工大学学报(自然科学), 2021, 35(10): 85-93.
	LI Shunming, HOU Yuzhe, LI Xianglian. Review on time-frequency-domain analysis methods for vibration faults of rolling bearings[J]. Journal of Chongqing University of Technology (Natural Science), 2021, 35(10): 85-93.
[5]	张士强. 基于深度学习的故障诊断技术研究[D]. 哈尔滨: 哈尔滨工业大学, 2018.
	ZHANG Shiqiang. Research on fault diagnosis technology based on deep learning[D]. Harbin:Harbin Institute of Technology, 2018.
[6]	ZHAO R, YAN R Q, CHEN Z H, et al. Deep learning and its applications to machine health monitoring[J]. Mechanical Systems and Signal Processing, 2019, 115: 213-237. doi: 10.1016/j.ymssp.2018.05.050
[7]	TAMILSELVAN P, WANG P F. Failure diagnosis using deep belief learning based health state classification[J]. Reliability Engineering & System Safety, 2013, 115: 124-135. doi: 10.1016/j.ress.2013.02.022 URL
[8]	HOANG D T, KANG H J. Rolling element bearing fault diagnosis using convolutional neural network and vibration image[J]. Cognitive Systems Research, 2019, 53: 42-50. doi: 10.1016/j.cogsys.2018.03.002 URL
[9]	ZHAO X L, JIA M P, LIU Z. Fault diagnosis framework of rolling bearing using adaptive sparse contrative auto-encoder with optimized unsupervised extreme learning machine[J]. IEEE Access, 2019, 8: 99154-99170. doi: 10.1109/Access.6287639 URL
[10]	ZHANG X, HUANG T, WU B, et al. Multi-model ensemble deep learning method for intelligent fault diagnosis with high-dimensional samples[J]. Frontiers of Mechanical Engineering, 2021, 16(2): 340-352. doi: 10.1007/s11465-021-0629-3
[11]	SHAO S Y, WANG P, YAN R Q. Generative adversarial networks for data augmentation in machine fault diagnosis[J]. Computers in Industry, 2019, 106: 85-93. doi: 10.1016/j.compind.2019.01.001 URL
[12]	WANG J, HE Q B. Wavelet packet envelope manifold for fault diagnosis of rolling element bearings[J]. IEEE Transactions on Instrumentation and Measurement, 2016, 65(11): 2515-2526. doi: 10.1109/TIM.2016.2566838 URL
[13]	SOUALHI A, MEDJAHER K, ZERHOUNI N. Bearing health monitoring based on Hilbert-Huang transform, support vector machine, and regression[J]. IEEE Transactions on Instrumentation and Measurement, 2015, 64(1): 52-62. doi: 10.1109/TIM.2014.2330494 URL
[14]	CHEN J L, LI Z P, PAN J, et al. Wavelet transform based on inner product in fault diagnosis of rotating machinery: A review[J]. Mechanical Systems and Signal Processing, 2016, 70/71: 1-35. doi: 10.1016/j.ymssp.2015.08.023 URL
[15]	GOODFELLOW I, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial networks[J]. Communications of the ACM, 2020, 63(11): 139-144. doi: 10.1145/3422622 URL
[16]	ZHENG M, LI T, ZHU R, et al. Conditional Wasserstein generative adversarial network-gradient penalty-based approach to alleviating imbalanced data classification[J]. Information Sciences, 2020, 512: 1009-1023. doi: 10.1016/j.ins.2019.10.014 URL
[17]	ARJOVSKY M, CHINTALA S, BOTTOU L. Wasserstein gan[EB/OL]. (2017-12-06)[2022-01-20]. https://arxiv.org/abs/1701.07875
[18]	ADDAGARLA S K. Real time multi-scale facial mask detection and classification using deep transfer learning techniques[J]. International Journal of Advanced Trends in Computer Science and Engineering, 2020, 9(4): 4402-4408. doi: 10.30534/ijatcse/2020/33942020 URL
[19]	PENG C, LI L L, CHEN Q, et al. A fault diagnosis method for rolling bearings based on parameter transfer learning under imbalance data sets[J]. Energies, 2021, 14(4): 944. doi: 10.3390/en14040944 URL
[20]	TAN B, ZHANG Y, PAN S J, et al. Distant domain transfer learning[C]// Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence. San Francisco, California, USA: ACM, 2017: 2604-2610.

序号	故障类型	源域 0 HP	目标域
序号	故障类型	源域 0 HP	0 HP	1 HP	2 HP
1	Normal	400	400	400	400
2	I(1)	400	400	400	400
3	I(2)	400	400	400	400
4	I(3)	400	400	400	400
5	B(1)	400	400	400	400
6	B(2)	400	400	400	400
7	B(3)	400	400	400	400
8	O(1)	400	400	400	400
9	O(2)	400	400	400	400
10	O(3)	400	400	400	400