Application of zonal Reduced-Order-Modeling to tire rolling simulation

Physic-based simulation remains a key enabler for real-world ever-growing complex industrial systems especially when crucial decisions are needed. While classical approaches have proven their accuracy and robustness over the years and come with a rich mathematical foundation, they suffer from several limitations depending of the underlying physics and use cases. For instance, especially concerning the resolution of Partial Differential Equations (PDEs) in 3 dimensions (3D), classical approaches are known to be computationally expensive. However, it turns out that simple pure data-driven approaches, while allegedly much more efficient from a computational point of view, do not necessarily hold up well regarding physical considerations. In this work, our aim is to investigate the tradeoff between accuracy and computational cost to design efficient and robust physical simulation methods under industrial constraints. In particular, as it is not easy to generate a large dataset through numerical simulations for such a problem, our aim is to design an approach addressing the data scarcity issue. To do so, we propose to hybridize a standard Finite Element Method (FEM) physics-based solver with a zonal Reduced Order Model (ROM) approach to simulate a rolling tire.