State of Charge Estimation for Lithium-ion Batteries Based on Adaptive Fractional Extended Kalman Filter

Battery state of charge (SOC) estimation is crucial for battery management systems to ensure the reliability and safety of electric vehicles. To achieve accurate SOC estimation, the fractional-order model which can accurately describe the diffusion and polarization of batteries is established and parameterized by particle swarm optimization firstly. Then, the Kalman filter method that can realize optimal estimation of systems is combined with fractional calculus by utilizing fractional state function. Consequently, the fractional extended Kalman filter (FEKF) is built up, in which an adaptive variance update algorithm is adopted to improve the convergence speed and robustness. Finally, the proposed algorithms are applied to two dynamic working conditions and the experimental results indicate that the adaptive FEKF is efficient and accurate in SOC estimation.