SOC Estimation of LiFePO4 Battery with Dual Adaptive Unscented Kalman Filter

The traditional unscented kalman filter (UKF) algorithm is used to estimate the state of charge (SOC) of the lithium battery, it has better estimation effect, and less amount of calculation. However, this algorithm also has its limitations, It can only be used when the statistical properties of observation noise and process noise are known. This paper proposes an adaptive unscented Kalman filter (AUKF) algorithm to solve these problems based on an adaptive algorithm and an unscented kalman filter strategy. The AUKF can use the noise statistical estimator to revise the unknown or inaccurate noise statistical properties in real time. According to the experimental results, it has been shown that the AUKF stronger convergence speed and stability than the traditional UKF.