Transported Memory Networks accelerating Computational Fluid Dynamics
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912247237836800 |
|---|---|
| author | Schulz, Matthias Jouan, Gwendal Berger, Daniel Gavranovic, Stefan Hartmann, Dirk |
| author_facet | Schulz, Matthias Jouan, Gwendal Berger, Daniel Gavranovic, Stefan Hartmann, Dirk |
| contents | In recent years, augmentation of differentiable PDE solvers with neural networks has shown promising results, particularly in fluid simulations. However, most approaches rely on convolutional neural networks and custom solvers operating on Cartesian grids with efficient access to cell data. This particular choice poses challenges for industrial-grade solvers that operate on unstructured meshes, where access is restricted to neighboring cells only. In this work, we address this limitation using a novel architecture, named Transported Memory Networks. The architecture draws inspiration from both traditional turbulence models and recurrent neural networks, and it is fully compatible with generic discretizations. Our results show that it is point-wise and statistically comparable to, or improves upon, previous methods in terms of both accuracy and computational efficiency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_18591 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Transported Memory Networks accelerating Computational Fluid Dynamics Schulz, Matthias Jouan, Gwendal Berger, Daniel Gavranovic, Stefan Hartmann, Dirk Machine Learning Fluid Dynamics In recent years, augmentation of differentiable PDE solvers with neural networks has shown promising results, particularly in fluid simulations. However, most approaches rely on convolutional neural networks and custom solvers operating on Cartesian grids with efficient access to cell data. This particular choice poses challenges for industrial-grade solvers that operate on unstructured meshes, where access is restricted to neighboring cells only. In this work, we address this limitation using a novel architecture, named Transported Memory Networks. The architecture draws inspiration from both traditional turbulence models and recurrent neural networks, and it is fully compatible with generic discretizations. Our results show that it is point-wise and statistically comparable to, or improves upon, previous methods in terms of both accuracy and computational efficiency. |
| title | Transported Memory Networks accelerating Computational Fluid Dynamics |
| topic | Machine Learning Fluid Dynamics |
| url | https://arxiv.org/abs/2502.18591 |