Multiscale heat transfer in a battery pack

Battery packs undergoing thermal runaway pose a serious safety hazard, and accurate modeling of these systems is crucial for system design and safety. However, the multiscale nature of these systems presents a significant challenge, as accurately resolving spatial and temporal scales spanning multiple orders of magnitude is computationally intensive and pushes the limits of theoretical models.

In this project, we developed a non-intrusive two-way coupled hybrid algorithm for simulating heat transfer in battery packs undergoing thermal runaway. Our hybrid algorithm combines fine-scale (cell-scale) and continuum-scale (battery pack-scale) models in the same numerical simulation to achieve predictive accuracy while keeping computational costs manageable.

We developed two methods with different orders of accuracy to enforce continuity at the coupling boundary, and derived weak formulations of the fine-scale and upscaled governing equations for numerical implementation in FEniCS. Through our simulations, we demonstrated that our hybrid algorithm can accurately predict the average temperature fields within errors determined by homogenization theory.

Furthermore, we demonstrated the computational efficiency of our hybrid algorithm against cell-scale simulations, highlighting the significant computational savings achieved without compromising the accuracy of the simulation results.

Our project presents a significant advancement in the field of battery pack modeling, providing a non-intrusive hybrid algorithm that can accurately simulate thermal runaway in these systems. Our findings have important implications for battery pack design and safety, providing a valuable tool for improving the efficiency and safety of these systems.