Computational modeling and statistical evaluation of thermal behavior of cylindrical lithium-ion battery

Abstract

Understanding of thermal behavior of lithium-ion batteries under various operating conditions is crucial to develop robust battery thermal management system. Moreover, an accurate determination of parameter effects is essential for research, including battery thermal analysis and safety design. This article presents the battery temperature behavior of a 26,650 lithium-ion battery at multi-scale multi-dimension, combining computational modeling and statistical analysis. The thermal model was used to evaluate the complex effects of state-of-health, discharge current rate, heat transfer coefficient, and ambient temperature on the maximum battery temperature (Tmax) and the maximum battery temperature difference (dTmax). The current rate was the most significant parameter influencing both Tmax (delta = 0.85) and dTmax (delta = 14.95). The ambient temperature with the delta value of 0.51 was the second main factor affecting the Tmax, and the h with the delta value of 8.84 was the second crucial parameter for the dTmax. The effect of the SoH can be neglected compared to other discharge parameters. The statistical evaluation can potentially be used to assist battery thermal management systems. This study calls for more statistical methods to quantify battery temperature rise and correlate it with different charge/discharge parameters.