Review of "grey box" lifetime modeling for lithium-ion battery: Combining physics and data-driven methods. (1st December 2022)

• Record Type: Journal Article
• Authors: Guo, Wendi; Sun, Zhongchao; Vilsen, Søren Byg; Meng, Jinhao; Stroe, Daniel Ioan
• Is Part Of: Journal of energy storage. Volume 56:Part A(2022)
• Publication Date: 2022-12-01
• Subjects: ANN artificial neural network -- BMS battery management system -- BTMS battery temperature management system -- BOL beginning of life -- CDKF central differential Kalman filter -- CT computerized tomography -- DAE differential algebraic equations -- DNN deep neural network -- DRA discrete-time realization algorithm -- DRT distribution function of relaxation times -- ECM equivalent circuit model -- EIS electrochemical impedance spectroscopy -- EKF extended Kalman filter -- EM electrochemical model -- EnKF ensemble Kalman filter -- EODV end-of-discharge voltage -- EST energy storage technology -- ETNN electrochemical-thermal-neural network -- EV electric vehicle -- FLW finite length warburg -- FNN feedforward neural network -- FSW finite space warburg -- GA genetic algorithm -- GHPF Gauss-Hermite particle filter -- GITT galvanostatic intermittent titration technique -- GPR Gaussian process regression -- IEKF iterative extended Kalman filter -- IMM interacting-multiple-model -- KF Kalman filter -- LASSO least absolute shrinkage and selection operator -- LiB lithium-ion battery -- LS least squares -- LSTM long short-term memory -- Mask R-CNN mask regional convolutional neural network -- MC Monte Carlo -- MCMC Markov Chain Monte Carlo -- ML machine learning -- MLP multi-layer perceptron -- MOA multi optimization analysis -- NODE neural ordinary differential equations -- OCV open circuit voltage -- ODE ordinary differential equation -- P2D Persudo two dimension -- PCDNN physics-constrained deep neural network -- PC polynomial chaos -- PDEs partial differential equations -- PDF probability density function -- PEM point estimate method -- PF particle filter -- PHM prognostics and health management -- PINN physics-informed neural network -- PITT potentiostatic intermittent titration technique -- RBPF Rao–Blackwellized PF -- RLS recursive least squares -- RMSE root mean square error -- RNN recurrent neural network -- ROM reduced-order model -- RUL remaining useful life -- RVM relevance vector machines -- SEI solid electrolyte interphase -- SEM Scanning Electron Microscope -- SOA single optimization analysis -- SOC state-of-charge -- SOH state of health -- SPKF Sigma Point Kalman filter -- SPM single-particle model -- SPMT SPM coupled with thermal effects -- UKF unscented Kalman filter -- UODE universal ordinary differential equation -- UQ uncertainty qualification -- XPS X-ray photoelectron spectroscopy -- XRD X-ray Diffraction; Lithium-ion batteries -- Lifetime modeling -- Physics-based models -- Data-driven methods -- Data-driven assisted physical models -- Physics-guided data-driven; Energy storage -- Periodicals; Energy storage -- Research -- Periodicals; 621.3126
• DOI: 10.1016/j.est.2022.105992
• Languages: English
• ISSNs: 2352-152X

Abstract

Abstract: Lithium-ion batteries are a popular choice for a wide range of energy storage system applications. The current motivation to improve the robustness of lithium-ion battery applications has stimulated the need for in-depth research into aging effects and the establishment of lifetime prediction models. This paper reviews different combination approaches of physics-based models and data-driven models. The three basic physics-based battery lifetime models are introduced, and requirements and features are compared from an application perspective. Then, state-of-the-art approaches for integrating physics and data-driven methods are systematically reviewed. Flowcharts present each approach to offer the readers a clear understanding. Next, the publication trends are represented by line graphs, and pie charts, including data-driven assisted physical models and physics-guided data-driven, different physical model applications, and data-driven approaches. It is concluded that electrochemical models have great potential to describe complex aging behavior under various conditions. Moreover, machine learning is a promising tool to overcome mechanistic absence and highly nonlinear performance, occupying 78 % of all data-driven methods. Physics-guided data-driven approach started to emerge as an innovative lifetime prediction method after 2020. The application advantages and limitations are compared according to the description of different methods. Furthermore, future perspectives are discussed, with opportunities and challenges. The Prospect of applying physics-guided machine learning looks forward to more inspiration. Graphical abstract: Unlabelled Image Highlights: Physics-based lifetime modeling for lithium-ion batteries is classified into three broad categories. The requirements and capabilities of these models are compared from an application perspective. The combination of physical and data-driven approaches is divided into two main categories. The first one is data-driven assisted physical models, termed as physical model prediction is the primary driver, and data-driven methods assist it. The other one is physics guided data-driven, where a physical model is used to guide and constrain data-driven predictions. The different approaches are illustrated with flowcharts. The publication trend of selected papers is presented as line graphs. Different trends in the application of physical models and trends in the application of data-driven methods are discussed. The requirements, advantages, and disadvantages of different integration methods are compared. Readers can select an appropriate method based on their available resources. Future development based on physic guided data-driven is proposed. Considering EM-PINN is recognized as a promising direction. It is challenging to simultaneously overcome the high complexity of EMs and combine it with machine learning to improve the computational efficiency of online applications.