【摘要】 锂离子电池荷电状态(SOC)与开路电压(OCV)的关系曲线(OCV曲线)是影响其SOC估计精度的核心因素。针对小电流OCV(LO)测试耗时短但数据精度较低的问题，提出一种OCV模型及其优化方法。该方法基于LO测试的OCV数据，采用道格拉斯-普克算法和分段线性函数建立OCV模型。并将OCV曲线上的4个OCV点作为变量，建立了其他OCV点的随动模型，使曲线能够运用粒子群优化算法进行优化。基于优化后的OCV曲线，动态工况下的端电压估计绝对平均误差降低83.5%，采用自适应扩展卡尔曼滤波的SOC估计误差小于0.3%。该方法能够基于耗时短的LO测试获取准确OCV曲线，降低锂离子电池研究与应用的测试成本。更多还原

【Abstract】 The relationship curve between state-of-charge(SOC) and open-circuit voltage(OCV) of lithium-ion batteries is the core factor affecting SOC estimation accuracy. An OCV model and its optimization method are proposed to address the issue of short test times but low accuracy of the data in the low-current OCV(LO) test. Based on OCV data from the LO test, the method creates an OCV model using the Douglas-Peucker algorithm and piecewise linear functions. Since four OCV points on the OCV curve are utilized as variables and a follow-up model is created for other OCV points, the curve could be optimized using particle swarm optimization. Based on the optimized OCV curve, the absolute average error of terminal voltage estimation under dynamic conditions is reduced by 83.5%, and the SOC estimation error using adaptive extended Kalman filtering is less than 0.3%. Accurate OCV curves based on the short-time LO test can be obtained by this method, reducing the testing cost in the research and application of lithium-ion batteries.更多还原