Battery monitoring and prognostics optimization techniques: Challenges and opportunities. (15th September 2022)

• Record Type: Journal Article
• Authors: Semeraro, Concetta; Caggiano, Mariateresa; Olabi, Abdul-Ghani; Dassisti, Michele
• Is Part Of: Energy. Volume 255(2022)
• Publication Date: 2022-09-15
• Subjects: Battery -- Model-based approaches -- Data-driven approaches -- Hybrid approaches -- Optimization techniques; ACKF Adaptive Cubature Kalman Filter -- AI Artificial Intelligence -- ANFIS Adaptive Neuro-Fuzzy Inference Syste -- ANN Artificial Neural Network -- ARNN Adaptive Recurrent Neural Network -- AUKF Adaptive Unscented Kalman Filter -- BA Bayesian Approach -- BLS Broad Learning System -- BMO Barnacles Mating Optimizer -- BMS Battery Management Systems -- BN Bayesian Network -- BPNN Back Propagation Neural Network -- B-LSTM Long-Short-Term Backpropagation Neural Memory -- CC-CV Costant Current-Costant Voltage -- CHMM Continuous Hidden Markov Model -- CKF Cubature Kalman Filter -- CNN Convolutional Neural Network -- CPE Constant Phase Element -- CWRNN Clockwork Recurrent Neural Network -- Cmaximum Maximum available capacity -- Cremaining Remaining capacity -- Ct Capacity estimate at time t -- C0 New battery's nominal capacity -- DAE-NN Denoising AutoEncoder Neural Network -- DBN Dynamic Bayesian Network -- DCNN Deep Convolution Neural Network -- DE Discharge Efficiency -- DL Deep Learning -- DOD Depth Of Discharge -- ECL Equivalent Circle Life -- ECM Equivalent circuit modelling -- EDPSO Exponential Decay Particle Swarm Optimization -- EIS Electrochemical Impedance Spectroscopy -- EKF Extended Kalman Filter -- ELM Extreme Learning Machines -- EMT Electrochemical modelling techniques -- EOCV End of Charge Voltage -- ERA Exponential Regression Algorithm -- ESC External Short Circuit -- FD Fault Detection -- FFNN Feed-Forward Neural Networks -- FOC Fractional-Order Calculus -- FOTM, SOTM First and the Second-Order Thèvenin ECM -- FUDS Federal Urban Drive Schedule -- GA-SVR Genetic Algorithm Optimized Support Vector Regression -- GM Grey Model -- GNL General Non-Linear -- GPR Gaussian process regression -- GRU Gated Recurrent Unit -- HPPC Hybrid Pulse Power Characterization -- IM Impedence Model -- IndRNN Independent Recurrent Neural Network -- KDE Kernel Density Estimation -- KF Kalman Filter -- LCA Linear Correlation Analysis -- LS Least-Squares -- LSE Least Squares Estimate -- LS-SVM Least-Squares Support Vector Machine -- LSTM Long Short-Term Memory -- MAE Mean-Absolute-Error -- MAPE Mean Absolute-Percentage Error -- MEM Minimalist Electrochemical Model -- ML Machine Learning -- MONESN Monotone Echo State Networks -- MSVM Multi-kernel SVM -- NASA National Aeronautics and Space Administration -- NB Naive Bayes -- NCA Nonparametric Correlation Analysis -- NN Neural Network -- PA Probabilistic algorith -- PBM Physics-Based Model -- PCA Partial Correlation Analysis -- PF Particle Filter -- PHM Prognostics and Health Management -- PNGV Partnership for a New Generation of Vehicles -- PNN Probabilistic Neural Network -- PRS Pseudo-Random Sequence -- RB Rule-Based -- RBF Radial Basis Functions -- RC Resistor-Capacitor -- RLM Recursive Levenberg-Marquardt -- RLS Recursive Least Squares -- RMSE Root Mean Square Error -- RNN Recurrent Neural Networks -- ROM Reduced-Order Model -- RUL Remaing Useful Life -- RUP Remaining Useful Performance -- RVFL Random Vector Functional Link -- RVM Relevance Vector Machine -- RWPF Random Walk Particle Filtering -- SOC State of charge -- SOE State of Energy -- SOH State of Health -- SPKF Sigma-Point Kalman Filtering -- SRCKF Square Root Kalman Cubing -- SSL Self-Supervised Learning -- SVM Support Vector Machines -- TM Thevenin model -- UKF Unscented Kalman Filter -- VRLA Valve-Regulated Lead-Acid -- ZEBRA Zero Emission Battery Research Activities; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2022.124538
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: In recent years, many researchers have been conducted on batteries' health monitoring and prognostics, mainly focusing on the batteries' state of charge (SOC). Accurately estimating the state of health (SOH) and predicting the remaining useful life (RUL) of battery components are very important for the prognosis and health management of the overall battery system. However, due to the non-linear dynamics caused by the electrochemical characteristics in batteries, the accurate estimations of SOC, SOH and RUL prediction are still challenging and many technologies have been developed to solve this challenge. This paper reviews and discusses state of the art in SOC and SOH and RUL estimation techniques for all battery types. A novel framework is developed and presented to compare all battery techniques based on three dimensions: battery performance (Z dimension), approaches (X dimension), and criteria (Y dimension) to fulfil. All studies are reviewed and discussed based on the dimensions and the criteria defined in the framework. Based on this investigation, this study summarizes at the end the key outcomes and suggests future research challenges. Graphical abstract: Novel framework for battery monitoring and prognostics optimization techniques comparison. Image 1 Highlights: Batteries' health monitoring and prognosis. Predicting the battery service life. Techniques for battery health monitoring and management. Framework to compare all battery techniques based on the performance and the criteria to fulfil.