Impact of Battery Aging on the Performance of Low-Complexity State-of-Charge Estimators: A Real-Data Analysis *
Year: 2026
Type: conference-paper
Abstract: This paper proposes a low-complexity State-of-Charge (SoC) estimation strategy for lithium-ion batteries under aging. The approach reformulates a second-order equivalent circuit model into a higher-order linear time-varying structure, enabling the use of a classical Kalman filter with linear output (KFLO). This avoids both the linearization errors of the EKF and the computational overhead of the UKF. A hybrid identification scheme updates internal resistance online via event-driven current variations while recursively correcting cycle efficiency. Experimental validation on progressively aged cells shows that the estimator maintains robust accuracy, achieving RMSE below 2.5% and MAE below 1.3% even for aged batteries, outperforming EKF and UKF benchmarks. Results demonstrate the method's suitability for resourceconstrained BMS applications requiring long-term reliability.
Language: English
Subfield: Automotive Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7201848578
