IEEE Transactions on Industrial Electronics · 2021 · 20 citations · 31 references
EngineeringChemical EngineeringElectrochemical EngineeringStochastic ModelingTime Series AnalysisBattery SystemsBattery TechnologyBayesian FilteringKalman FilteringState EstimationState Of Charge EstimationBattery ModelingBayesian EstimationComputational EfficiencyEquivalent Circuit ModelingExtended Kalman FilterLithium-ion BatteriesLithium-ion Battery State Of Charge EstimationState Of Charge Prediction
The state-of-charge (SOC) estimation and remaining-dischargeable-time (RDT) prediction are critical and challenging to safe operation of Li-ion batteries. The main challenges are the limited accuracy of traditional equivalent circuit model and computation-inefficiency of electrochemical battery models. To address this problem, this article proposes a Lebesgue-sampling-based extended Kalman filter (LS-EKF) approach that integrates the high fidelity of a simplified first principle (SFP) model with the low computation of Lebesgue sampling (LS) in a Bayesian estimation framework. In this framework, the SFP model is first introduced along with its design and validation. The LS-EKF is employed with the SFP model to estimate the SOC and predict the RDT. The proposed method is verified with a series of experiments under different operating conditions. The results and comparisons demonstrate the effectiveness of the proposed method in terms of the accuracy of estimation and prediction, as well as the computational efficiency.
31
A few useful things to know about machine learning
Pedro Domingos · Communications of the ACM · 2012 · 3.2K citations