Gear Degradation Trend Prediction by Meta-learning Gated Recurrent Unit Networks under Few Samples

doi:10.3901/JME.2022.03.149

Abstract

Abstract: To address the low accuracy of gear degradation trend prediction due to over-fitting in deep neural network training under small gear sample conditions, a meta-learning gated recurrent unit (MLGRU) network is proposed for gear degradation trend prediction. The gated recurrent neural network is built on the framework of meta-learning, and the network is optimized through multi-task training to reduce over-fitting of the gear prediction model, so that the MLGRU networks can converge quickly under small-sample conditions. Deep learning of gear evolution patterns and prediction of degradation trends by gated recurrent unit network. The experimental results show that the proposed meta-learning gated neural networks can fully learn the gear evolution mechanism and predict the gear degradation trend.

Key words: gears, degradation trend prediction, meta-learning, gated recurrent neural networks, few samples

CLC Number:

TG156

YU Xiaoxia, DENG Lei, TANG Baoping, XIA Yi, LI Qikang. Gear Degradation Trend Prediction by Meta-learning Gated Recurrent Unit Networks under Few Samples[J]. Journal of Mechanical Engineering, 2022, 58(3): 149-156.

References

[1] XIANG Sheng,QIN Yi,ZHU Caichao,et al. LSTM networks based on attention ordered neurons for gear remaining life prediction[J]. ISA Transactions,2020,106:343-354.
[2] 叶壮,余建波. 基于多通道加权卷积神经网络的齿轮箱振动信号特征提取[J]. 机械工程学报,2021,57(1):111-112. YE Zhuang,YU Jianbo. Feature extraction of gearbox vibration signals based on multi-channels weighted convolutional neural network[J]. Journal of Mechanical Engineering,2021,57(1):111-112.
[3] XIANG Sheng,QIN Yi,ZHU Caichao,et al. Long short-term memory neural network with weight amplification and its application into gear remaining useful life prediction[J]. Engineering Applications of Artificial Intelligence,2020,103587(91):1-3.
[4] KULKARNI S,ACHENBACH J. Structural health monitoring and damage prognosis in fatigue[J]. Struct. Heal. Monit., 2008,7:37-49.
[5] BOSMAN R,SCHIPPER D. Mild wear maps for boundary lubricated contacts[J]. Wear,2012,280:54-62.
[6] BAJPAI P,KAHRAMAN A,ANDERSON NE,A surface wear prediction methodology for parallel-axis gear pairs[J]. J. Tribol., 2004,126:597-605.
[7] KUMAR A,GANDHI C,ZHOU Y,et al. Latest developments in gear defect diagnosis and prognosis:A review[J]. Measurement,2020,158(107735):2-7.
[8] 王鹏,邓蕾,汤宝平,等. 基于自编码器和门限循环单元神经网络的滚动轴承退化趋势预测[J]. 振动与冲击,2020,39(17):107-108. WANG Peng,DENG Lei,TANG Baoping,et al. Degradation trend prediction of rolling bearing based on auto-encoder and GRU neural networkJ]. Journal of Vibration and Shock,2020,39(17):107-108.
[9] 金晓航,孙毅,单继宏,等. 风力发电机组故障诊断与预测技术研究综述[J]. 仪器仪表学报,2017,38(5):1041-1053. JIN Xiaohang,SUN Yi,SHAN Jihong,et al. Fault diagnosis and prognosis for wind turbines:An overview[J]. Chinese Journal of Scientific Instrument,2017,38(5):1041-1053.
[10] YAN Haoran,QIN Yi,XIANG Sheng,et al. Long-term gear life prediction based on ordered neurons LSTM neural networks[J],Measurement Journal,2020,165(108205):1-10.
[11] YU Yong,HU Changhua,SI Xiaosheng,et al. Averaged Bi-LSTM networks for RUL prognostics with non-life-cycle labeled dataset[J]. Neurocomputing,2020,402:134-147.
[12] SHE Daoming,JIA Minping. A BiGRU method for remaining useful life prediction of machinery[J]. Measurement,2021,167(108277):2-8.
[13] ELFORJANI M. Diagnosis and prognosis of real world wind turbine gears[J]. Renewable Energy,2020,147:1676-1693.
[14] FINN C,ABBEEL P,LEVINE S. Model-agnostic meta-learning for fast adaptation of deep networks[C]//International Conference on Machine Learning. PMLR,2017:1126-1135.
[15] JAVED K,GOURIVEAU R,ZERHOUNI N,et al. Enabling health monitoring approach based on vibration data for accurate prognostics[J]. IEEE Transactions on Industrial Electronics,2014,62(1):647-656.
[16] LEI Yaguo,LI Naipeng,GUO Liang,et al. Machinery health prognostics:A systematic review from data acquisition to RUL prediction[J]. Mechanical Systems and Signal Processing,2018,104:799-834.
[17] CUI Yiqian,SHI Junyou,WANG Zili. Discrete event logistics systems (DELS) simulation modeling incorporating two-step remaining useful life (RUL) estimation[J]. Computers in Industry,2015,72:68-81.
[18] WANG Saisai,CHEN Jie,WANG Hua,et al. Degradation evaluation of slewing bearing using HMM and improved GRU[J]. Measurement,2019,146:385-395.
[19] CAMBRON P,MASSON C,TAHAN A,et al. Control chart monitoring of wind turbine generators using the statistical inertia of a wind farm average[J]. Renewable Energy,2018,116:88-98.
[20] GUO Liang,LI Naipeng,JIA Feng,et al. A recurrent neural network based health indicator for remaining useful life prediction of bearings[J]. Neurocomputing,2017,240:98-109.
[21] CHO K,VAN MERRIËNBOER B,GULCEHRE C,et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation[J]. arXiv preprint arXiv:1406.1078,2014.
[22] LI Hai,WANG Wei,LI Ziwei,et al. A novel approach for predicting tool remaining useful life using limited data[J]. Mechanical Systems and Signal Processing,2020,143(106832):1-20.
[23] CHUNG J,GULCEHRE C,CHO K H,et al. Empirical evaluation of gated recurrent neural networks on sequence modeling[J]. arXiv preprint arXiv:1412.3555,2014:1-5.