“…For example, the development of a reliable battery management system (BMS) is challenging considering the difficulties in accurate monitoring of key battery states like the state of charge (SOC) and state of health (SOH). [1][2][3][4] The SOC estimation strategies in the literature can be categorized into two types: open-loop and closed-loop. 5 The Coulomb counting (CC) method 6 is a typical open-loop strategy to obtain the SOC (SOC t = SOC initial − ηI(t)Δt/Cn, SOC t is the SOC at the time t, SOC initial is Nomenclature: ANN, Artificial neural network; AEKF, Adaptive extended Kalman filter; BMS, Battery management system; CC, Coulomb counting; CKF, Cubature Kalman filter; CRR, Capacity retention ratio; EVs, Electrified vehicles; EKF, Extended Kalman filter; ECM, Equivalent circuit model; FBCRLS, Frisch scheme-based bias compensating recursive least squares; GA, Genetic algorithm; HPPC, Hybrid pulse power characterization; HWFET, Highway fuel economy driving schedule; KF, Kalman filter; ML, Maximum likelihood; MAE, Mean absolute error; NMC, LiNiMnCoO 2 ; NYCC, New york city cycle; OECM, Offset-free equivalent circuit model; OCV, Open circuit voltage; PF, Particle filter; RTLS, Recursive total least squares; RLS, Recursive least squares; RMSE, Root mean square error; SOC, State of charge; SOH, State of health; SVM, Supporting vector machine; SPKF, Sigma-point Kalman filter; SFTP, Supplemental federal test procedure; UKF, Unscented Kalman filter; UPF, Unscented particle filter; UDDS, Urban dynamometer driving schedule; VICO, Voltage-input/current-output; WTM, Wavelet transform matrix the starting point of the SOC, η is the Coulombic efficiency and is usually set to be 1, Δt is the time interval, Cn is the actual capacity).…”