Intelligent Bearing Fault Diagnosis Based on Adaptive Deep Belief Network Under Variable Working Conditions

doi:10.16183/j.cnki.jsjtu.2021.161

Abstract

Abstract:

In engineering, working environment and operating state are constantly changing, which decreases the correct rate of equipment fault diagnosis, resulting in the loss of time and cost. The structure of the deep belief network is investigated for the time-varying factors in the mechanical system. In combination with the signal decomposition technology of fixed learning step size, the original characteristics of the sensor data are retained. In addition, the deep key information of the signal is repeatedly extracted layer by layer. The data loss technology is integrated to optimize the network structure to avoid over-fitting problems. Further, considering the domain adaptive method in transfer learning, the memory characteristics of different levels of deep belief networks are solidified. Therefore, a domain adaptive deep belief network with shift-invariant features (SIF-DADBN) is proposed for rolling bearing fault diagnosis. By identifying the characteristic information of similar fault signals with variable working conditions, the accuracy and generalization of bearing intelligent fault diagnosis are both improved. Based on the public data set of rolling bearings, the average correct rate of the fault diagnosis technology is found to be as high as 95.65%. Compared with five other methods, the effectiveness and accuracy of SIF-DADBN under variable working conditions are verified.

Key words: variable working conditions, fault diagnosis, deep belief network, domain adaptation, dropout

CLC Number:

TH17

MA Hangyu, ZHOU Di, WEI Yujie, WU Wei, PAN Ershun. Intelligent Bearing Fault Diagnosis Based on Adaptive Deep Belief Network Under Variable Working Conditions[J]. Journal of Shanghai Jiao Tong University, 2022, 56(10): 1368-1377.

Figures/Tables 16

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Tab.1

Fig.7

Fig.8

Fig.9

Fig.10

Fig.11

Fig.12

Tab.2

Fig.13

Fig.14

References 26

[1]	孙旺, 李彦明, 杜文辽, 等. 基于蚁群神经网络的泵车主泵轴承性能评估[J]. 上海交通大学学报, 2012, 46(4): 596-600.
	SUN Wang, LI Yanming, DU Wenliao, et al. State performance evaluation for the main pump bearing of pump truck based on ant colony optimization of neural network[J]. Journal of Shanghai Jiao Tong University, 2012, 46(4): 596-600.
[2]	雷亚国, 贾峰, 周昕, 等. 基于深度学习理论的机械装备大数据健康监测方法[J]. 机械工程学报, 2015, 51(21): 49-56. doi: 10.3901/JME.2015.21.049
	LEI Yaguo, JIA Feng, ZHOU Xin, et al. A deep learning-based method for machinery health monitoring with big data[J]. Journal of Mechanical Engineering, 2015, 51(21): 49-56. doi: 10.3901/JME.2015.21.049
[3]	WEN X Q, XU Z A. Wind turbine fault diagnosis based on ReliefF-PCA and DNN[J]. Expert Systems with Applications, 2021, 178: 115016. doi: 10.1016/j.eswa.2021.115016 URL
[4]	HASHIM M A, NASEF M H, KABEEL A E, et al. Combustion fault detection technique of spark ignition engine based on wavelet packet transform and artificial neural network[J]. Alexandria Engineering Journal, 2020, 59(5): 3687-3697. doi: 10.1016/j.aej.2020.06.023 URL
[5]	逯程, 徐廷学, 王虹. 基于属性粒化聚类与回声状态网络的末制导雷达故障诊断[J]. 上海交通大学学报, 2018, 52(9): 1112-1119.
	LU Cheng, XU Tingxue, WANG Hong. Fault diagnosis of terminal guidance radar based on attribute granulation clustering and echo state network[J]. Journal of Shanghai Jiao Tong University, 2018, 52(9): 1112-1119.
[6]	王昊, 邱思琦, 王丽亚. 结合深度自编码与强化学习的轴承健康评估[J]. 工业工程与管理, 2021, 26(3): 89-95.
	WANG Hao, QIU Siqi, WANG Liya. Health assessment of bearings based on deep auto-encoder and reinforcement learning[J]. Industrial Engineering and Management, 2021, 26(3): 89-95.
[7]	马萍, 张宏立, 范文慧. 基于局部与全局结构保持算法的滚动轴承故障诊断[J]. 机械工程学报, 2017, 53(2): 20-25. doi: 10.3901/JME.2017.02.020
	MA Ping, ZHANG Hongli, FAN Wenhui. Fault diagnosis of rolling bearings based on local and global preserving embedding algorithm[J]. Chinese Journal of Mechanical Engineering, 2017, 53(2): 20-25.
[8]	MINHAS A S, KANKAR P K, KUMAR N, et al. Bearing fault detection and recognition methodology based on weighted multiscale entropy approach[J]. Mechanical Systems and Signal Processing, 2021, 147: 107073. doi: 10.1016/j.ymssp.2020.107073 URL
[9]	MA Y L, CHENG J S, WANG P, et al. Rotating machinery fault diagnosis based on multivariate multiscale fuzzy distribution entropy and Fisher score[J]. Measurement, 2021, 179: 109495. doi: 10.1016/j.measurement.2021.109495 URL
[10]	沈飞, 陈超, 徐佳文, 等. 谱质心迁移在变工况轴承故障诊断的应用[J]. 仪器仪表学报, 2019, 40(5): 99-108.
	SHEN Fei, CHEN Chao, XU Jiawen, et al. Application of spectral centroid transfer inbearing fault diagnosis under varying working conditions[J]. Chinese Journal of Scientific Instrument, 2019, 40(5): 99-108.
[11]	LI Q, SHEN C Q, CHEN L, et al. Knowledge mapping-based adversarial domain adaptation: A novel fault diagnosis method with high generalizability under variable working conditions[J]. Mechanical Systems and Signal Processing, 2021, 147: 107095. doi: 10.1016/j.ymssp.2020.107095 URL
[12]	ZHAO B, ZHANG X M, LI H, et al. Intelligent fault diagnosis of rolling bearings based on normalized CNN considering data imbalance and variable working conditions[J]. Knowledge-Based Systems, 2020, 199: 105971. doi: 10.1016/j.knosys.2020.105971 URL
[13]	赵小强, 梁浩鹏. 使用改进残差神经网络的滚动轴承变工况故障诊断方法[J]. 西安交通大学学报, 2020, 54(9): 23-31.
	ZHAO Xiaoqiang, LIANG Haopeng. Fault diagnosis method for rolling bearing under variable working conditions using improved residual neural network[J]. Journal of Xi’an Jiaotong University, 2020, 54(9): 23-31.
[14]	HINTON G E, SALAKHUTDINOV R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786): 504-507. doi: 10.1126/science.1127647 pmid: 16873662
[15]	李艳峰, 王新晴, 张梅军, 等. 基于奇异值分解和深度信度网络多分类器的滚动轴承故障诊断方法[J]. 上海交通大学学报, 2015, 49(5): 681-686.
	LI Yanfeng, WANG Xinqing, ZHANG Meijun, et al. An approach to fault diagnosis of rolling bearing using SVD and multiple DBN classifiers[J]. Journal of Shanghai Jiao Tong University, 2015, 49(5): 681-686.
[16]	MA S, CHU F L. Ensemble deep learning-based fault diagnosis of rotor bearing systems[J]. Computers in Industry, 2019, 105: 143-152. doi: 10.1016/j.compind.2018.12.012
[17]	WANG G M, QIAO J F, LI X L, et al. Improved classification with semi-supervised deep belief network[J]. IFAC PapersOnLine, 2017, 50(1): 4174-4179. doi: 10.1016/j.ifacol.2017.08.807 URL
[18]	CHE C C, WANG H W, NI X M, et al. Domain adaptive deep belief network for rolling bearing fault diagnosis[J]. Computers & Industrial Engineering, 2020, 143: 106427. doi: 10.1016/j.cie.2020.106427 URL
[19]	JAITLY N, HINTON G. Learning a better representation of speech soundwaves using restricted boltzmann machines[C]//2011 IEEE International Conference on Acoustics, Speech and Signal Processing. Prague, Czech Republic: IEEE, 2011: 5884-5887.
[20]	HINTON G.E. A practical guide to training restricted Boltzmann machines[J]. Momentum, 2010, 9(1), 926-947.
[21]	HINTON G E. Training products of experts by minimizing contrastive divergence[J]. Neural Computation, 2002, 14(8): 1771-1800. doi: 10.1162/089976602760128018 pmid: 12180402
[22]	CHEN Y, YANG C L, ZHANG Y. Deep domain similarity Adaptation Networks for across domain classification[J]. Pattern Recognition Letters, 2018, 112: 270-276. doi: 10.1016/j.patrec.2018.08.006 URL
[23]	GRETTON A, BORGWARDT K M, RASCH M J, et al. A kernel two-sample test[J]. Journal of Machine Learning Research, 2012, 13(1): 723-773
[24]	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.
[25]	张振良, 刘君强, 黄亮, 等. 基于半监督迁移学习的轴承故障诊断方法[J]. 北京航空航天大学学报, 2019, 45(11): 2291-2300.
	ZHANG Zhenliang, LIU Junqiang, HUANG Liang, et al. A bearing fault diagnosis method based on semi-supervised and transfer learning[J]. Journal of Beijing University of Aeronautics and Astronautics, 2019, 45(11): 2291-2300.
[26]	WANG G M, QIAO J F, BI J, et al. TL-GDBN: Growing deep belief network with transfer learning[J]. IEEE Transactions on Automation Science and Engineering, 2019, 16(2): 874-885. doi: 10.1109/TASE.2018.2865663 URL

数据集	电机转速/ (r·min^-1)	样本数/个	数据点/个	训练集/个		测试集/个
数据集	电机转速/ (r·min^-1)	样本数/个	数据点/个	源域	目标域	测试集/个
A	1 730	40+9×40	2 000	40+9×40
B	1 750	40+9×40	2 000		10+9×10
C	1 772	40+9×40	2 000			10+9×10
D	1 797	40+9×40	2 000

序号	方法	节点设置	数据点/个	训练集+测试集/(个+个)
1	SIF-DADBN	200、200、160、100、50、10	2 000	500+100
2	Raw-DBN	2 000、200、160、100、0、10	2 000	500+100
3	FFT-DBN^[18]	12、200、160、100、50、10	2 000	500+100
4	TL^[25]	200、10	2 000	500+100
5	SVM	200、10	2 000	500+100
6	TL-GDBN^[26]	200、200、160、100、50、10	2 000	500+100