Averaged Bi-LSTM networks for RUL prognostics with non-life-cycle labeled dataset

Abstract LSTM network is an effective RNN model to predict the system RUL for its superior performance in sequential data processing. Usually, networks trained by life-cycle labeled dataset would possess similar RUL predicting accuracies, because the network training algorithm could ensure the network optimality for the whole training dataset. However, for networks trained by non-life-cycle labeled samples, the network uncertainty caused by different training conditions could lead to degradation prediction uncertainty for some local points. Further, the RUL predicting results that are computed by these uncertain local points may shows relatively large differences. Therefore, in order to obtain an accurate RUL prediction with networks trained by non-life-cycle labeled samples, our paper proposes a novel network model averaging method to deal with the network uncertainty. What is more, to learn the temporal correlation information of training samples sufficiently, we adopt the Bi-LSTM network to illustrate the application of the proposed network model averaging method. Finally, degradation values of Graphite/LiCoO2 battery are used to verify the effectiveness of the proposed method. The results show that the proposed method could improve the RUL prediction accuracy and reduce the prediction error.

[1]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[2]  Yandong Hou,et al.  Satellite lithium-ion battery remaining useful life estimation with an iterative updated RVM fused with the KF algorithm , 2017 .

[3]  Jürgen Schmidhuber,et al.  Learning to Forget: Continual Prediction with LSTM , 2000, Neural Computation.

[4]  Qiang Miao,et al.  Prognostics of lithium-ion batteries based on relevance vectors and a conditional three-parameter capacity degradation model , 2013 .

[5]  Jie Liu,et al.  Lithium-ion battery remaining useful life estimation based on fusion nonlinear degradation AR model and RPF algorithm , 2013, Neural Computing and Applications.

[6]  Li Lin,et al.  Remaining useful life estimation of engineered systems using vanilla LSTM neural networks , 2018, Neurocomputing.

[7]  Yong Yu,et al.  A Review of Recurrent Neural Networks: LSTM Cells and Network Architectures , 2019, Neural Computation.

[8]  Ramana V. Grandhi,et al.  A Bayesian statistical method for quantifying model form uncertainty and two model combination methods , 2014, Reliab. Eng. Syst. Saf..

[9]  Stephen Grossberg,et al.  Recurrent neural networks , 2013, Scholarpedia.

[10]  Li Lin,et al.  Fault diagnosis and remaining useful life estimation of aero engine using LSTM neural network , 2016, 2016 IEEE International Conference on Aircraft Utility Systems (AUS).

[11]  Yang Liu,et al.  A Novel Lifetime Estimation Method for Two-Phase Degrading Systems , 2019, IEEE Transactions on Reliability.

[12]  M. Steel,et al.  Benchmark Priors for Bayesian Model Averaging , 2001 .

[13]  F.O. Heimes,et al.  Recurrent neural networks for remaining useful life estimation , 2008, 2008 International Conference on Prognostics and Health Management.

[14]  Michael Buchholz,et al.  Health diagnosis and remaining useful life prognostics of lithium-ion batteries using data-driven methods , 2013 .

[15]  Xiang Li,et al.  Remaining useful life estimation in prognostics using deep convolution neural networks , 2018, Reliab. Eng. Syst. Saf..

[16]  Xiao-Sheng Si,et al.  Prognostics for Linear Stochastic Degrading Systems With Survival Measurements , 2020, IEEE Transactions on Industrial Electronics.

[17]  Dang-Bo Du,et al.  A Prognostic Model Based on DBN and Diffusion Process for Degrading Bearing , 2020, IEEE Transactions on Industrial Electronics.

[18]  Razvan Pascanu,et al.  On the difficulty of training recurrent neural networks , 2012, ICML.

[19]  Antonios Tsourdos,et al.  Prognostics of crack propagation in structures using time delay neural network , 2015, 2015 IEEE Conference on Prognostics and Health Management (PHM).

[20]  Xin Zhang,et al.  Remaining useful life prediction of lithium-ion battery using an improved UPF method based on MCMC , 2017, Microelectron. Reliab..

[21]  Takehisa Yairi,et al.  A review on the application of deep learning in system health management , 2018, Mechanical Systems and Signal Processing.

[22]  Xianliang Zhang,et al.  A new statistical precipitation downscaling method with Bayesian model averaging: a case study in China , 2015, Climate Dynamics.

[23]  Yong Sun,et al.  A review on degradation models in reliability analysis , 2010, WCE 2010.

[24]  Hongwen He,et al.  Long Short-Term Memory Recurrent Neural Network for Remaining Useful Life Prediction of Lithium-Ion Batteries , 2018, IEEE Transactions on Vehicular Technology.

[25]  Chetan Gupta,et al.  Long Short-Term Memory Network for Remaining Useful Life estimation , 2017, 2017 IEEE International Conference on Prognostics and Health Management (ICPHM).

[26]  Adrian E. Raftery,et al.  Bayesian model averaging: a tutorial (with comments by M. Clyde, David Draper and E. I. George, and a rejoinder by the authors , 1999 .

[27]  Wasserman,et al.  Bayesian Model Selection and Model Averaging. , 2000, Journal of mathematical psychology.

[28]  David He,et al.  Using deep learning based approaches for bearing remaining useful life prediction , 2016 .

[29]  Mohamed Elforjani,et al.  Prognosis of Bearing Acoustic Emission Signals Using Supervised Machine Learning , 2018, IEEE Transactions on Industrial Electronics.

[30]  M. Pecht,et al.  Cycle life testing and modeling of graphite/LiCoO 2 cells under different state of charge ranges , 2016 .

[31]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[32]  Bin Zhang,et al.  Lithium-ion battery remaining useful life prediction with Deep Belief Network and Relevance Vector Machine , 2017, 2017 IEEE International Conference on Prognostics and Health Management (ICPHM).

[33]  xing-hui Zhang,et al.  Hidden Markov models in bearing fault diagnosis and prognosis , 2010, 2010 Second International Conference on Computational Intelligence and Natural Computing.

[34]  Jan Drewes Achenbach,et al.  Structural Health Monitoring and Damage Prognosis in Fatigue , 2008 .

[35]  Michael Osterman,et al.  Prognostics of lithium-ion batteries based on DempsterShafer theory and the Bayesian Monte Carlo me , 2011 .

[36]  Haidong Shao,et al.  Rolling bearing fault detection using continuous deep belief network with locally linear embedding , 2018, Comput. Ind..

[37]  Afnizanfaizal Abdullah,et al.  A Review of Deep Learning Architectures and Their Application , 2017, AsiaSim 2017.

[38]  Weiwen Peng,et al.  Bayesian Degradation Analysis With Inverse Gaussian Process Models Under Time-Varying Degradation Rates , 2017, IEEE Transactions on Reliability.

[39]  Hong-Zhong Huang,et al.  A Bidirectional LSTM Prognostics Method Under Multiple Operational Conditions , 2019, IEEE Transactions on Industrial Electronics.

[40]  Bhaskar Saha,et al.  An Adaptive Recurrent Neural Network for Remaining Useful Life Prediction of Lithium-ion Batteries , 2010 .

[41]  Wei Liang,et al.  Remaining useful life prediction of lithium-ion battery with unscented particle filter technique , 2013, Microelectron. Reliab..

[42]  Kai Zhang,et al.  Remaining Useful Life Prediction of Lithium-Ion Batteries Using Neural Network and Bat-Based Particle Filter , 2019, IEEE Access.

[43]  Xiaoli Li,et al.  Deep Convolutional Neural Network Based Regression Approach for Estimation of Remaining Useful Life , 2016, DASFAA.

[44]  Bo Peng,et al.  Cycle life estimation of lithium-ion polymer batteries using artificial neural network and support vector machine with time-resolved thermography , 2017, Microelectron. Reliab..