Semi-empirical modeling and experimental evaluation of partial immersive thermal management performance for 21700 cylindrical cells targeted towards automotive battery packs

Thermal management of lithium-ion cells in battery packs is of paramount importance, ensuring that their extended service life and safety are guaranteed. This paper explores the potential benefits that partial direct thermal management may offer for automotive battery packs. It presents a systematic modeling and experimentation campaign of a battery pack composed of 21700 cylindrical cells, evaluating its thermal response under six thermal management conditions at the same electrical load. Static air and forced air-cooling are performed, setting a baseline for the analysis. The effectiveness of a single-phase dielectric liquid coolant for direct thermal management is evaluated with varying immersion levels of the cells. The level of the fluid is increased from 25% to 100% of the battery cell height, following 25% increments. For each condition, the pack is tested with a discharge rate of 0.5C, with temperature measurements recorded at the positive terminal, negative terminal and the body of the central cell of the pack. The work presents the development of a multiphysics model considering electrochemical and thermal physics coupling, based on a semi-empirical lumped battery representation. The model results to have a good fit with the experimental data, with a maximum error of 1.4 °C when simulating the partial immersion conditions. The experimental outcomes indicate that an immersion ratio of 50% is an agreeable trade-off between maximum temperature reached, maximum temperature difference between the positive and negative terminals, and coolant volume for the studied battery pack.