State of charge estimation using extended Kalman filtering approach with sensitivity and energy management assessment for battery electric vehicles

Abstract

The accurate estimation of State of Charge (SoC) is critical for safe, reliable, and optimistic operation of Battery Electric Vehicles (BEVs). Nevertheless, achieving robust SoC estimation is a big challenge because of nonlinear battery dynamics, parameter variability, sensor noise, and uncertain initial conditions. Standard techniques for measuring the state of charge (SoC), e.g., Coulomb Counting (CC) and Open Circuit Voltage (OCV), tend to drift and are not flexible in rapidly changing operating conditions. In this paper, we propose a model-based SoC estimation framework using an Extended Kalman Filter (EKF) and a second-order Thevenin equivalent circuit model. Previous works propose more sophisticated and hybrid estimation algorithms; the main contribution of this work is a systematic and reproducible assessment of classical EKF performance under various controlled scenarios with respect to disturbances caused by sensor noise, bias, and initialization errors. The EKF implementation is tested on the standardized driving cycles based on the Worldwide Harmonized Light-Duty Vehicle Test Procedure (WLTP), the Urban Dynamometer Driving Schedule (UDDS), and the Highway Fuel Economy Driving Schedule (HWFET) in a MATLAB-based simulation platform. A comparative study with Coulomb Counting is performed under identical disturbance conditions, which enables the identification of regions of operation where the EKF offers robustness benefits. Findings show that Coulomb Counting performs well under ideal conditions but degrades under realistic disturbances, whereas the EKF maintains stable and accurate state estimation, making it more reliable for BEV energy management.