Enhancing lattice kinetic schemes for fluid dynamics with Lattice-Equivariant Neural Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ortali, Giulio, Gabbana, Alessandro, Atmodimedjo, Imre, Corbetta, Alessandro
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910921119498240
author Ortali, Giulio
Gabbana, Alessandro
Atmodimedjo, Imre
Corbetta, Alessandro
author_facet Ortali, Giulio
Gabbana, Alessandro
Atmodimedjo, Imre
Corbetta, Alessandro
contents We present a new class of equivariant neural networks, hereby dubbed Lattice-Equivariant Neural Networks (LENNs), designed to satisfy local symmetries of a lattice structure. Our approach develops within a recently introduced framework aimed at learning neural network-based surrogate models Lattice Boltzmann collision operators. Whenever neural networks are employed to model physical systems, respecting symmetries and equivariance properties has been shown to be key for accuracy, numerical stability, and performance. Here, hinging on ideas from group representation theory, we define trainable layers whose algebraic structure is equivariant with respect to the symmetries of the lattice cell. Our method naturally allows for efficient implementations, both in terms of memory usage and computational costs, supporting scalable training/testing for lattices in two spatial dimensions and higher, as the size of symmetry group grows. We validate and test our approach considering 2D and 3D flowing dynamics, both in laminar and turbulent regimes. We compare with group averaged-based symmetric networks and with plain, non-symmetric, networks, showing how our approach unlocks the (a-posteriori) accuracy and training stability of the former models, and the train/inference speed of the latter networks (LENNs are about one order of magnitude faster than group-averaged networks in 3D). Our work opens towards practical utilization of machine learning-augmented Lattice Boltzmann CFD in real-world simulations.
format Preprint
id arxiv_https___arxiv_org_abs_2405_13850
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Enhancing lattice kinetic schemes for fluid dynamics with Lattice-Equivariant Neural Networks
Ortali, Giulio
Gabbana, Alessandro
Atmodimedjo, Imre
Corbetta, Alessandro
Computational Physics
Machine Learning
Fluid Dynamics
We present a new class of equivariant neural networks, hereby dubbed Lattice-Equivariant Neural Networks (LENNs), designed to satisfy local symmetries of a lattice structure. Our approach develops within a recently introduced framework aimed at learning neural network-based surrogate models Lattice Boltzmann collision operators. Whenever neural networks are employed to model physical systems, respecting symmetries and equivariance properties has been shown to be key for accuracy, numerical stability, and performance. Here, hinging on ideas from group representation theory, we define trainable layers whose algebraic structure is equivariant with respect to the symmetries of the lattice cell. Our method naturally allows for efficient implementations, both in terms of memory usage and computational costs, supporting scalable training/testing for lattices in two spatial dimensions and higher, as the size of symmetry group grows. We validate and test our approach considering 2D and 3D flowing dynamics, both in laminar and turbulent regimes. We compare with group averaged-based symmetric networks and with plain, non-symmetric, networks, showing how our approach unlocks the (a-posteriori) accuracy and training stability of the former models, and the train/inference speed of the latter networks (LENNs are about one order of magnitude faster than group-averaged networks in 3D). Our work opens towards practical utilization of machine learning-augmented Lattice Boltzmann CFD in real-world simulations.
title Enhancing lattice kinetic schemes for fluid dynamics with Lattice-Equivariant Neural Networks
topic Computational Physics
Machine Learning
Fluid Dynamics
url https://arxiv.org/abs/2405.13850