Sensitivity analysis for parameter classification of energy balance-integrated single particle model for battery cells

다국어 초록 (Multilingual Abstract)

With the increasing use of electric vehicles (EVs), there is a growing interest in the thermal management ofEVs. In this study, we first reduced the computational complexity of single particle model (SPM) for the battery cell byintroducing a 4th order approximation for Li-ion concentration in the solid phase. In addition, by integrating it with anenergy balance, the constructed model can calculate the battery temperature along with the terminal voltage and stateof charge. To develop a model compatible with the experimental data requires parameter estimation. However, the estimationaccuracy for each parameter depends on its sensitivity. We investigated the influence of 16 parameters on themeasured data under general experimental conditions (constant C-rate discharge) through simulations and sensitivityanalysis. We classified the radius of the particle, total active surface areas, electrode maximum concentration, and a heattransfer coefficient as dominant parameters. When dominant parameters were estimated using the virtual experimentaldata, the percent error was smaller than 3.1%. For the parameters with minor influence, the estimation error waslarge even with the excellent agreement of the experimental data. We confirmed which parameter could be estimatedusing the C-rate experimental data accurately and which parameter should be estimated with additional experiments.