有色噪声条件下的动力电池SOC估算

SOC Estimation of Power Battery Within Colored Noise

由于纯电动汽车的工作环境十分复杂,电流量测中的有色噪声信号会造成动力电池荷电状态(SOC)估算结果的不精确甚至发散,为解决该问题,在此提出一种增广矩阵扩展卡尔曼联合递推增广最小二乘(AMEKFRELS)算法,针对有色噪声进行建模,在线辨识并实时修正系统参数。仿真与实验结果表明,与简化有色噪声为白噪声的估测算法相比,该算法响应速度快,估测精度高,能够满足动力电池实际应用需要。

Because pure electric vehicles working environment is very complex, the colored noise signal in current t will cause the battery state of charge（SOC） estimation inaccurate or even divergent,to solve this prob- lem, an method which joint augmented matrix extended Kalman filter with recursive extended least squares （AMEKF- RELS） algorithm is proposed, aiming at modeling colored noise,identifying and correcting system parameters online.Both the simulation and experimental results reveal that compared with simplifying the colored noise to white noise ,the AMEKF-RELS algorithm has faster response, higher accuracy, which can meet practical application of power battery.