Enhanced deep gated recurrent unit and complex wavelet packet energy moment entropy for early fault prognosis of bearing

被引：188

作者：

Shao Haidong ^{[1
,2
,3
]}

Cheng Junsheng ^{[1
,2
]}

Jiang Hongkai ^{[4
]}

Yang Yu ^{[1
,2
]}

Wu Zhantao ^{[1
,2
]}

机构：

[1] Hunan Univ, State Key Lab Adv Design & Mfg Vehicle Body, Changsha 410082, Peoples R China

[2] Hunan Univ, Coll Mech & Vehicle Engn, Changsha 410082, Peoples R China

[3] Lulea Univ Technol, Dept Civil Environm & Nat Resources Engn, S-97187 Lulea, Sweden

[4] Northwestern Polytech Univ, Sch Aeronaut, Xian 710072, Peoples R China

来源：

KNOWLEDGE-BASED SYSTEMS | 2020年 / 188卷

基金：

中国国家自然科学基金;

关键词：

Enhanced deep gated recurrent unit; Bearing; Early fault prognosis; Energy moment entropy; Modified training algorithm; CONVOLUTIONAL NEURAL-NETWORK; BELIEF NETWORK; DIAGNOSIS; MANIFOLD; ENCODER;

D O I：

10.1016/j.knosys.2019.105022

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Early fault prognosis of bearing is a very meaningful yet challenging task to improve the security of rotating machinery. For this purpose, a novel method based on enhanced deep gated recurrent unit and complex wavelet packet energy moment entropy is proposed in this paper. First, complex wavelet packet energy moment entropy is defined as a new monitoring index to characterize bearing performance degradation. Second, deep gated recurrent unit network is constructed to capture the nonlinear mapping relationship hidden in the defined monitoring index. Finally, a modified training algorithm based on learning rate decay strategy is developed to enhance the prognosis capability of the constructed deep model. The proposed method is applied to analyze the simulated and experimental signals of bearing. The results demonstrate that the proposed method is more superior in sensibility and accuracy to the existing methods. (C) 2019 Elsevier B.V. All rights reserved.

引用

页数：14

共 38 条

[1]

[Anonymous], P 34 INT C MACH LEAR

[2]

[Anonymous], 2015, Deep learn. nat., DOI [10.1038/nature14539, DOI 10.1038/NATURE14539]

[3]

[Anonymous], 5 INT C LEARN REPR F

[4] Early fault diagnosis of rotating machinery based on wavelet packets-Empirical mode decomposition feature extraction and neural network [J].

Bin, G. F. ;

Gao, J. J. ;

Li, X. J. ;

Dhillon, B. S. .

MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2012, 27 :696-711

[5] LSTM: A Search Space Odyssey [J].

Greff, Klaus ;

Srivastava, Rupesh K. ;

Koutnik, Jan ;

Steunebrink, Bas R. ;

Schmidhuber, Juergen .

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2017, 28 (10) :2222-2232

[6] A recurrent neural network based health indicator for remaining useful life prediction of bearings [J].

Guo, Liang ;

Li, Naipeng ;

Jia, Feng ;

Lei, Yaguo ;

Lin, Jing .

NEUROCOMPUTING, 2017, 240 :98-109

[7] A novel adversarial learning framework in deep convolutional neural network for intelligent diagnosis of mechanical faults [J].

Han, Te ;

Liu, Chao ;

Yang, Wenguang ;

Jiang, Dongxiang .

KNOWLEDGE-BASED SYSTEMS, 2019, 165 :474-487

[8] Vibration signal classification by wavelet packet energy flow manifold learning [J].

He, Qingbo .

JOURNAL OF SOUND AND VIBRATION, 2013, 332 (07) :1881-1894

[9] Rotating machinery prognostics: State of the art, challenges and opportunities [J].

Heng, Aiwina ;

Zhang, Sheng ;

Tan, Andy C. C. ;

Mathew, Joseph .

MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2009, 23 (03) :724-739

[10] Intelligent fault diagnosis of rolling bearings using an improved deep recurrent neural network [J].

Jiang, Hongkai ;

Li, Xingqiu ;

Shao, Haidong ;

Zhao, Ke .

MEASUREMENT SCIENCE AND TECHNOLOGY, 2018, 29 (06)

← 1 2 3 4 →