基于端电压残差的超级电容融合建模方法研究

A Supercapacitor Fusion Modeling Method Based on the Terminal Voltage Residual

针对单一模型在超级电容整个工作区段内无法持续保持最优的问题,依托Rint模型、Thevenin模型和GNL模型,提出了1种基于端电压残差科学切换模型函数的超级电容融合建模方法。为了更加精确地建立3个基础模型,基于HPPC(Hybrid Pulse Power Characteristic)实验数据,采用遗传算法分别对这3个模型进行了离线参数辨识,在UDDS(Urban Dynamometer Driving Schedule)标准工况下,获得3个模型的端电压残差数据集。然后在每一时刻对3个模型的端电压残差进行对比,选择最优模型作为当前时刻的模型,以此获得融合模型。再对融合模型和3个基础模型之间的最大误差、平均误差和均方根误差进行对比研究。结果表明,提出的融合建模方法精度和可靠性更高,有利于车载电源系统的优化控制。

In order to solve the problem that a single model cannot keep optimal in the whole working section of the supercapacitor.Relying on Rint model,Thevenin model and GNL model,an supercapacitor fusion modeling method based on terminal voltage residuals switching model function is proposed.In order to establish the three basic models more accurately,genetic algorithm is adopted for off-line parameter identification of these three models based on Hybrid Pulse Power Characteristic(HPPC)experimental data.Then,the terminal voltage residual data of the three models are obtained under the Urban Dynamometer Driving Schedule(UDDS)condition.Finally,by comparing the terminal voltage residual of the three models at each moment,the optimal model is selected as the current moment model to get the fusion model.By comparing the maximum error,mean error and root mean square error between the fusion model and the three basic models.The results show that the proposed fusion modeling method has higher accuracy and reliability,which is beneficial to the optimization control of the on-board power supply system.