基于Hermite插值的锂电池参数辨识方法研究

Research on Parameter Identification Method for Lithium Battery Based on Hermite Interpolation

针对传统最小二乘法建模时拟合的各个参数在测试点的准确数据不能全部落在拟合曲线上的问题,提出利用Hermite插值法建立等效电路模型,并拟合各模型参数曲线,建立三元锂电池等效戴维南模型(Thevenin)。基于三元锂电池复合脉冲功率试验(HPPC)数据,使用Matlab中的lsqcurvefit函数辨识出模型的四个参数,并在Simulink模块验证所建模型的有效性。对电池进行动态工况放电试验。试验结果表明,利用Hermite插值法建立的Thevenin模型具有非常高的精度。仿真验证表明,端电压最大误差为0.07 V,用于荷电状态(SOC)估算时的最大误差低于2%。

In order to solve the problem that the exact data of the parameters fitted by the traditional least squares method did not all fall on the fitting curve,it was proposed to establish the Thevenin model for ternary lithium battery by using the Hermite interpolation method to set up the equivalent circuit model and fitting the parameter curve of each model.Based on hybrid pulse power characteristic(HPPC) experiment data of ternary lithium battery,four parameters of the model were identified by lsqcurvefit function in Matlab,and the validity of the model was verified by Simulink module.The experimental results showed that the Thevenin model established by Hermite interpolation method had very high accuracy.The maximum error of terminal voltage was 0.07 V,and the maximum error for state of charge(SOC) estimation was less than 2%.