Deep Gaussian process regression for lithium-ion battery health prognosis and degradation mode diagnosis

Abstract Accurate state of health estimation and end of life prediction is critical for safe and reliable operation of lithium-ion batteries. This paper proposes a deep Gaussian process algorithm for lithium-ion battery health monitoring. The proposed algorithm uses Gaussian processes to model mapping between the layers, and matrix-variate Gaussian distribution to model correlation between nodes of a given layer. Deep architecture of the proposed algorithm enables capacity estimation using the partial charge-discharge time-series data, in the form of voltage, temperature and current, eliminating need for input feature extraction. The algorithm also estimates statistical correlation between capacity and the elapsed time from the first cycle, which enables end of life prediction. Hidden layer output of the proposed algorithm correlates strongly with the battery internal states, enabling on-board diagnosis of battery degradation mechanisms. The paper demonstrates effectiveness of the proposed algorithm using aging dataset of batteries cycled at different C-rates, temperatures and usage profiles. The proposed algorithm predicts capacity and end of life with coefficient of determinant greater than 0.9 and mean absolute error less than 0.1, demonstrating high prediction accuracy of the proposed approach.

[1]  Mohammad Farrokhi,et al.  State-of-Charge Estimation for Lithium-Ion Batteries Using Neural Networks and EKF , 2010, IEEE Transactions on Industrial Electronics.

[2]  M. Safari,et al.  Life Simulation of a Graphite/LiFePO4 Cell under Cycling and Storage , 2012 .

[3]  Michael A. Osborne,et al.  Gaussian Process Regression for In Situ Capacity Estimation of Lithium-Ion Batteries , 2017, IEEE Transactions on Industrial Informatics.

[4]  P. Bruce,et al.  Degradation diagnostics for lithium ion cells , 2017 .

[5]  Taejung Yeo,et al.  A novel multistage Support Vector Machine based approach for Li ion battery remaining useful life estimation , 2015 .

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

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

[8]  I. Villarreal,et al.  Critical review of state of health estimation methods of Li-ion batteries for real applications , 2016 .

[9]  Stephen J. Roberts,et al.  A tutorial on variational Bayesian inference , 2012, Artificial Intelligence Review.

[10]  John Newman,et al.  Stress generation and fracture in lithium insertion materials , 2005 .

[11]  Naoki Baba,et al.  Numerical simulation of thermal behavior of lithium-ion secondary batteries using the enhanced single particle model , 2014 .

[12]  Han-Lim Choi,et al.  Deep Gaussian Process-Based Bayesian Inference for Contaminant Source Localization , 2018, IEEE Access.

[13]  Thomas M. Cover,et al.  Elements of Information Theory , 2005 .

[14]  Jay Lee,et al.  Review and recent advances in battery health monitoring and prognostics technologies for electric vehicle (EV) safety and mobility , 2014 .

[15]  K. Goebel,et al.  Prognostics in Battery Health Management , 2008, IEEE Instrumentation & Measurement Magazine.

[16]  B. Saha,et al.  Uncertainty Management for Diagnostics and Prognostics of Batteries using Bayesian Techniques , 2008, 2008 IEEE Aerospace Conference.

[17]  Andreas Jossen,et al.  Calculation of the state of safety (SOS) for lithium ion batteries , 2016 .

[18]  A. O'Hagan,et al.  Bayesian emulation of complex multi-output and dynamic computer models , 2010 .

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

[20]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[21]  Mihai V. Micea,et al.  Online State-of-Health Assessment for Battery Management Systems , 2011, IEEE Transactions on Instrumentation and Measurement.

[22]  Zonghai Chen,et al.  A novel Gaussian process regression model for state-of-health estimation of lithium-ion battery using charging curve , 2018 .

[23]  Andreas Jossen,et al.  Comprehensive Modeling of Temperature-Dependent Degradation Mechanisms in Lithium Iron Phosphate Batteries , 2017 .

[24]  M. Verbrugge,et al.  Cycle-life model for graphite-LiFePO 4 cells , 2011 .

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

[26]  Michael A. Osborne,et al.  Gaussian process regression for forecasting battery state of health , 2017, 1703.05687.

[27]  Zafer Sahinoglu,et al.  Battery State-of-Charge Estimation Based on Regular/Recurrent Gaussian Process Regression , 2018, IEEE Transactions on Industrial Electronics.

[28]  Yi-Jun He,et al.  State of health estimation of lithium‐ion batteries: A multiscale Gaussian process regression modeling approach , 2015 .

[29]  Jean-Michel Vinassa,et al.  Behavior and state-of-health monitoring of Li-ion batteries using impedance spectroscopy and recurrent neural networks , 2012 .

[30]  M. Broussely,et al.  Aging mechanism in Li ion cells and calendar life predictions , 2001 .

[31]  Sun Zechang,et al.  A new SOH prediction concept for the power lithium-ion battery used on HEVs , 2009, 2009 IEEE Vehicle Power and Propulsion Conference.

[32]  Mark W. Verbrugge,et al.  Battery Cycle Life Prediction with Coupled Chemical Degradation and Fatigue Mechanics , 2012 .

[33]  Zoubin Ghahramani,et al.  Sparse Gaussian Processes using Pseudo-inputs , 2005, NIPS.

[34]  Oleg Wasynczuk,et al.  Physically-based reduced-order capacity loss model for graphite anodes in Li-ion battery cells , 2017 .

[35]  Michael Pecht,et al.  Quantitative Analysis of Lithium-Ion Battery Capacity Prediction via Adaptive Bathtub-Shaped Function , 2013 .

[36]  Piyush M. Tagade,et al.  A Gaussian process emulator approach for rapid contaminant characterization with an integrated multizone-CFD model , 2013, 1305.2565.

[37]  Yu Peng,et al.  Prognostics for state of health estimation of lithium-ion batteries based on combination Gaussian process functional regression , 2013, Microelectron. Reliab..

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

[39]  Kai Goebel,et al.  Comparison of prognostic algorithms for estimating remaining useful life of batteries , 2009 .

[40]  Marc Peter Deisenroth,et al.  Doubly Stochastic Variational Inference for Deep Gaussian Processes , 2017, NIPS.

[41]  Datong Liu,et al.  Lithium-ion battery remaining useful life estimation with an optimized Relevance Vector Machine algorithm with incremental learning , 2015 .

[42]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[43]  G. Amatucci,et al.  Surface degradation of Li1–xNi0.80Co0.15Al0.05O2 cathodes: Correlating charge transfer impedance with surface phase transformations , 2016 .

[44]  Simo Srkk,et al.  Bayesian Filtering and Smoothing , 2013 .

[45]  Kai Goebel,et al.  Modeling Li-ion Battery Capacity Depletion in a Particle Filtering Framework , 2009 .

[46]  Tom Gorka,et al.  Method for estimating capacity and predicting remaining useful life of lithium-ion battery , 2014, 2014 International Conference on Prognostics and Health Management.

[47]  A. Rukhin.  Matrix Variate Distributions , 1999, The Multivariate Normal Distribution.

[48]  Taejung Yeo,et al.  Bayesian calibration for electrochemical thermal model of lithium-ion cells , 2016 .

[49]  M. Doyle,et al.  Simulation and Optimization of the Dual Lithium Ion Insertion Cell , 1994 .

[50]  M. Safari,et al.  Multimodal Physics-Based Aging Model for Life Prediction of Li-Ion Batteries , 2009 .

[51]  Mohammad Rezvani,et al.  A Comparative Analysis of Techniques for Electric Vehicle Battery Prognostics and Health Management (PHM) , 2011 .

[52]  Dirk Uwe Sauer,et al.  Comparative study of a structured neural network and an extended Kalman filter for state of health determination of lithium-ion batteries in hybrid electricvehicles , 2013, Eng. Appl. Artif. Intell..

[53]  M. Doyle,et al.  Modeling of Galvanostatic Charge and Discharge of the Lithium/Polymer/Insertion Cell , 1993 .

[54]  Bhaskar Saha,et al.  Prognostics Methods for Battery Health Monitoring Using a Bayesian Framework , 2009, IEEE Transactions on Instrumentation and Measurement.

[55]  Jorge F. Silva,et al.  Particle-Filtering-Based Prognosis Framework for Energy Storage Devices With a Statistical Characterization of State-of-Health Regeneration Phenomena , 2013, IEEE Transactions on Instrumentation and Measurement.

[56]  Ali Emadi,et al.  Long Short-Term Memory Networks for Accurate State-of-Charge Estimation of Li-ion Batteries , 2018, IEEE Transactions on Industrial Electronics.

[57]  Simon F. Schuster,et al.  Calendar Aging of Lithium-Ion Batteries I. Impact of the Graphite Anode on Capacity Fade , 2016 .

[58]  P. Balbuena,et al.  Lithium-ion batteries : solid-electrolyte interphase , 2004 .

[59]  Neil D. Lawrence,et al.  Deep Gaussian Processes , 2012, AISTATS.

[60]  Lingling Li,et al.  Remaining Useful Life Prediction for Lithium-Ion Batteries Based on Gaussian Processes Mixture , 2016, PloS one.

[61]  Ralph E. White,et al.  Development of First Principles Capacity Fade Model for Li-Ion Cells , 2004 .

[62]  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.

[63]  Jay Lee,et al.  A review on prognostics and health monitoring of Li-ion battery , 2011 .

[64]  Ralph E. White,et al.  Extension of Physics-Based single Particle Model for Higher Charge-Discharge Rates , 2013 .

[65]  Dong Zhang,et al.  Hybrid electrochemical modeling with recurrent neural networks for li-ion batteries , 2017, 2017 American Control Conference (ACC).

[66]  Song-Yul Choe,et al.  Development of a physics-based degradation model for lithium ion polymer batteries considering side reactions , 2015 .

[67]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[68]  David A. Stone,et al.  Nonlinear observers for predicting state-of-charge and state-of-health of lead-acid batteries for hybrid-electric vehicles , 2005, IEEE Transactions on Vehicular Technology.

[69]  Huei Peng,et al.  On-board state of health monitoring of lithium-ion batteries using incremental capacity analysis with support vector regression , 2013 .

[70]  Kang Xu,et al.  EIS study on the formation of solid electrolyte interface in Li-ion battery , 2006 .