State of Charge Estimation of a LiFePO4 Battery Using a Dual Extended Kalman Filter

State of Charge (SOC) is a key metric in battery management systems (BMS), particularly for lithium iron phosphate ($\text{LiFePO}_{4}$) batteries. This paper proposes a method to enhance SOC estimation accuracy using a third-order equivalent circuit model (ECM) for modeling internal resistance. By integrating ECM parameters with a Dual Extended Kalman Filter (DEKF), the method improves SOC prediction accuracy under varying conditions. The ECM was calibrated using voltage and current discharge data from a 48 V, $100 \text{Ah} \text{LiFePO}_{4}$ battery, processed in MATLAB 2023b. The DEKF outperforms the conventional Extended Kalman Filter (EKF), providing more reliable SOC estimates, especially under dynamic load scenarios. Simulation results show a correlation between higher discharge currents and accelerated capacity loss, leading to a faster SOC decline and increased internal resistance. These insights contribute to the development of smarter, SOC-aware battery management systems for $\text{LiFePO}_{4}$ technology.