Thermal management of Lithium-ion batteries (Li-ion) is a critical challenge for the widespread adoption of electric vehicles in the automotive market. The heat transfer intensity offered by the immersion Battery Thermal Management Systems (BTMS) directly affect the capabilities of the Li-ion cells in terms of power and energy performances, lifespan and safety. The heat generated inside the cells results from complex electrochemical processes occurring at small scale within cell layers. However, for battery packs designers, the heat transfer processes of interest take place at pack level, for instance to assess temperature homogeneity across all cells. Reproducing both the heat transfer and electrochemical phenomena through numerical simulation at high fidelity therefore requires bridging a considerable scale gap, from the individual jelly-roll layers up to the full battery stack.

In this work, we develop a modeling framework coupling two physical domains. At inner layers scale, the Doyle-Fuller-Newman (DFN) model is implemented to reproduce the electrochemical dynamics between three generic layers of the jelly-roll (positive, negative electrodes and electrolyte) : ion intercalation, charge transfer kinetics, and temperature-dependent transport properties. At pack scale, a Computational Fluid Dynamics (CFD) model is implemented, with a Conjugate Heat Transfer (CHT) approach allowing to resolve the temperature field inside the batteries on one hand, and flow and temperature of the dielectric cooling fluid on the other hand. The coupling is performed through the temperature field : temperature fields from CFD update temperature dependent electrochemical properties of the DFN model. Heat generation is computed through ohmic, reversible, and irreversible mechanisms within the cell [3]. Numerically, the DFN model is implemented in the open source PyBaMM python library [2] and the CFD model is implemented in StarCCM+.

This coupling framework is first validated against experimental measurements of temperature, voltage and electrical current evolutions, over specific charging and discharging cycles. This comprehensive experimental approach is useful to validate both electrochemical and thermal predictions of the numerical framework. Overall, this approach stands as a useful tool for BTMS design. It allows to assess directly the effects of electrochemical parameters (such as electrode thickness and conductivity, as well as SEI layer molar concentration and thickness, among others) on the heating rates submitted to the cells, and then overall heat transfer performances of the BTMS. Conversely, one can assess which cooling strategy (mass flow, inlet temperature, type of coolant, pack and cells geometries) lead to smaller constraints on the cells and increase their lifespan, safety and reduce hazards of thermal runaway.

References

[1] Brosa Planella, F., et al. (2022). A continuum of physics-based lithium-ion battery models

reviewed. Progress in Energy, 4(4), 042003.

[2] Sulzer, V., Marquis, S.G., Timms, R., Robinson, M., Chapman, S.J. (2021). Python Battery Mathematical Modelling (PyBaMM). Journal of Open Research Software, 9(1), 14.

[3] Hu, X., Ping, R., Rong, D., Geng, M. (2025). Experimental and numerical study on the heat generation characteristics of 18650 commercial lithium-ion batteries over the full life cycle.