Lithium-ion batteries are used in a wide variety of applications, with growing expectations for even higher power and energy densities. Optimising battery design and operation is crucial, especially in extreme conditions e.g. charging/discharging at high rates. Large-format pouch cells are often the preferred choice in such scenarios, since their high surface-area-to-volume ratio enables effective heat removal. Unfortunately, this also leads to non-uniform temperature distributions especially at higher Crates. Recent literature, recognising the limitations of lumped modelling, indicates a rise in interest in predicting battery performance at high C-rates. Selecting and parameterising electrochemical-thermal models for this scenario is non-trivial [1,2].

In this study, we explore the use of dynamic mode decomposition, an established data-driven technique for simplifying battery models at low C-rates [3], for predicting surface temperature in pouch cells at high C-rates. Our aim is to develop a method using minimal amounts of measured data in time that could provide a promising alternative or complement to computationally intensive porous electrode theory models that have a large number of parameters. Hence, it can be used both used to speed up existing advanced analytical model parametrisation studies based on numerical multiphysics studies as well as interpolating experimental performance at different conditions. We show the effectiveness of our method using data from lock-in thermography experiments imaging surface temperatures on an 8 Ah A123 Systems LFP/Gr pouch cell during cycling experiments. Results show that it is possible to extrapolate multiphysics simulations very accurately (with mean error of 0.04% at 8C) from only onethird of the simulation data, leading to faster parameterisation. Our study also shows that it is possible to generate thermal performance data at different C-rates with moderate accuracy (i.e. 0.4% mean error). Hence, this method has potential to be used to estimate results of costly IR thermography experiments at intermediate points by performing simpler cycling tests.

References
1. Chu, H. N.; Kim, S. U.; Rahimian, S. K.; Siegel, J. B.; Monroe, C. W. Parameterization of Prismatic Lithium–Iron–Phosphate Cells through a Streamlined Thermal/Electrochemical Model. Journal of Power Sources 2020, 453, 227787. https://doi.org/10.1016/j.jpowsour.2020.227787.
2. Lin, J.; Chu, H. N.; Howey, D. A.; Monroe, C. W. Multiscale Coupling of Surface Temperature with Solid Diffusion in Large Lithium-Ion Pouch Cells. Communications Engineering 2022, 1 (1), 1–10. https://doi.org/10.1038/s44172-022-00005-8.
3. Kanbur, B. B.; Kumtepeli, V.; Duan. F. Thermal performance prediction of the battery surface via dynamic mode decomposition. Energy, 2022, 201, 117642. https://doi.org/10.1016/j.energy.2020.117642