基于扩展卡尔曼滤波的储能电池能量和功率状态联合估计方法

Joint energy and power state estimation method for energy storage battery based on extended Kalman filter

电池储能是碳中和目标的有力抓手,准确估计其能量状态(state of energy,SOE)和峰值功率状态(state of power,SOP)是电池储能高效可靠运行的关键和基础。由于电池的电化学反应过程十分复杂,作为隐性状态量的SOE和SOP精确值难以获得。为此,本工作提出了一种基于模型SOE和SOP联合估计方法。应用Thevenin等效电路模型,采用递归最小二乘法建立了在线参数辨识算法,获得准确的模型参数。为解决恒定功率需求下的功率预测难题,提出了多步功率预测法,提高了SOP的预测精度,并结合扩展卡尔曼滤波算法,进一步提出了多状态联合估计方法。实验验证了算法的可行性,结果表明,在存在较大初始误差的情况下,所提出的方法电压、SOE最大预测误差均<2%,实现了准确的SOP预测。

Battery energy storage is a powerful target for carbon neutrality.Accurate estimation of its state of energy(SOE)and state of power(SOP)is the key and foundation for the effective and reliable operation of battery energy storage.It is challenging to determine the precise values of SOE and SOP as recessive state quantities due to the intricacy of the electrochemical reaction process in batteries.Therefore,a model-based joint estimation method of SOE and SOP is suggested in this paper.Recursive least squares are utilized to create an online parameter identification technique using the Thevenin equivalent circuit model,and accurate model parameters are achieved.To address the prediction problem under constant power demand,a multi-step power prediction method is proposed to enhance the prediction accuracy of SOP.An additional joint estimation approach of SOE and SOP is suggested in conjunction with the expanded Kalman filter algorithm.The feasibility of the algorithm is verified by experiments.The findings demonstrate that,even in the presence of significant starting errors,the suggested method's maximum voltage and SOE prediction errors are both less than 2%, resulting in precise SOP prediction.