Abstract: Accurate and reliable state of charge （SOC） estimation for large capacity battery energy storage system （LCBESS） is necessary for the battery management system （BMS）. Since the difference between batteries, it is difficult to obtain ideal prediction results by traditional methods. To solve the above problem, this paper proposes the prediction method of SOC for LCBESS based on improved unscented Kalman filter （UKF）, which adopts the genetic algorithm （GA） method to optimize the filter coefficients of UKF, in order to further improve the prediction accuracy of SOC. This paper simulates the series battery energy storage system under setting conditions. The simulation results show that the proposed improved unscented Kalman filter method can obtain the effective and reliable prediction results of SOC, which has a broad prospect of engineering application.

Key words: genetic algorithm, unscented Kalman filter （UKF）, prediction of state of charge, equivalent circuit