2411.13194

Total: 1

#1 Solver-in-the-loop approach to turbulence closure [PDF] [Copy] [Kimi] [REL]

Authors: André Freitas, Kiwon Um, Mathieu Desbrun, Michele Buzzicotti, Luca Biferale

We present a novel methodology for modeling the influence of the unresolved scales of turbulence for sub-grid modeling. Our approach employs the differentiable physics paradigm in deep learning, allowing a neural network to interact with the differential equation evolution and performing an a posteriori optimization by incorporating the solver into the training iteration (an approach known as solver-in-the-loop), thus departing from the conventional a priori instantaneous training approach. Our method ensures that the model is exposed to equations-informed input distributions, accounting for prior corrections and often leading to more accurate and stable time evolution. We present results of our methodology applied to a shell model of turbulence, and we discuss further potential applications to Navier-Stokes equations.

Subjects: Fluid Dynamics ; Chaotic Dynamics

Publish: 2024-11-20 10:51:43 UTC