基于扩展卡尔曼粒子滤波算法的锂电池SOC估计

SOC Estimation for Li-Ion Battery Based on Extended Kalman Particle Filter

锂电池荷电状态用来描述电池剩余电量的多少,进而反映电动汽车的续驶里程,是电池管理系统中的核心参数。电池循环次数、瞬间大电流以及温度等因素都会使电池特性发生变化,使用扩展卡尔曼滤波算法对电池荷电状态进行估计,会有较大的误差甚至导致算法不收敛。为了有效地抑制发散以及噪声的影响,基于锂电池混合噪声模型,应用扩展卡尔曼粒子滤波算法对锂电池荷电状态和电流漂移噪声进行同步估计。最后根据充放电试验数据进行仿真分析,结果证明了该算法的优越性。

As the key parameter for power battery management, the SOC of Li-ion battery described the residual capacity,and indicated the remainder driving range of electric vehicles. The cycles, instantaneous high current, abnormal temperatures and other factors would change cell characteristics, which might introduce larger errors even divergence over time if the extended Kalman filter algorithm were applied to the SOC estimation. To suppress the divergence and noise, this paper proposed a method based on EKPF algorithm to realize accurate SOC and the current drift estimation on the Li-ion battery mixed noise model. Finally, the superiority of this method was validated by simulation results.