TripNet: Learning Large-scale High-fidelity 3D Car Aerodynamics with Triplane Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Qian, Elrefaie, Mohamed, Dai, Angela, Ahmed, Faez
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913854286462976
author Chen, Qian
Elrefaie, Mohamed
Dai, Angela
Ahmed, Faez
author_facet Chen, Qian
Elrefaie, Mohamed
Dai, Angela
Ahmed, Faez
contents Surrogate modeling has emerged as a powerful tool to accelerate Computational Fluid Dynamics (CFD) simulations. Existing 3D geometric learning models based on point clouds, voxels, meshes, or graphs depend on explicit geometric representations that are memory-intensive and resolution-limited. For large-scale simulations with millions of nodes and cells, existing models require aggressive downsampling due to their dependence on mesh resolution, resulting in degraded accuracy. We present TripNet, a triplane-based neural framework that implicitly encodes 3D geometry into a compact, continuous feature map with fixed dimension. Unlike mesh-dependent approaches, TripNet scales to high-resolution simulations without increasing memory cost, and enables CFD predictions at arbitrary spatial locations in a query-based fashion, independent of mesh connectivity or predefined nodes. TripNet achieves state-of-the-art performance on the DrivAerNet and DrivAerNet++ datasets, accurately predicting drag coefficients, surface pressure, and full 3D flow fields. With a unified triplane backbone supporting multiple simulation tasks, TripNet offers a scalable, accurate, and efficient alternative to traditional CFD solvers and existing surrogate models.
format Preprint
id arxiv_https___arxiv_org_abs_2503_17400
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TripNet: Learning Large-scale High-fidelity 3D Car Aerodynamics with Triplane Networks
Chen, Qian
Elrefaie, Mohamed
Dai, Angela
Ahmed, Faez
Fluid Dynamics
Machine Learning
Surrogate modeling has emerged as a powerful tool to accelerate Computational Fluid Dynamics (CFD) simulations. Existing 3D geometric learning models based on point clouds, voxels, meshes, or graphs depend on explicit geometric representations that are memory-intensive and resolution-limited. For large-scale simulations with millions of nodes and cells, existing models require aggressive downsampling due to their dependence on mesh resolution, resulting in degraded accuracy. We present TripNet, a triplane-based neural framework that implicitly encodes 3D geometry into a compact, continuous feature map with fixed dimension. Unlike mesh-dependent approaches, TripNet scales to high-resolution simulations without increasing memory cost, and enables CFD predictions at arbitrary spatial locations in a query-based fashion, independent of mesh connectivity or predefined nodes. TripNet achieves state-of-the-art performance on the DrivAerNet and DrivAerNet++ datasets, accurately predicting drag coefficients, surface pressure, and full 3D flow fields. With a unified triplane backbone supporting multiple simulation tasks, TripNet offers a scalable, accurate, and efficient alternative to traditional CFD solvers and existing surrogate models.
title TripNet: Learning Large-scale High-fidelity 3D Car Aerodynamics with Triplane Networks
topic Fluid Dynamics
Machine Learning
url https://arxiv.org/abs/2503.17400