PulseAugur
实时 10:20:04

Faster by Design: Interactive Aerodynamics via Neural Surrogates Trained on Expert-Validated CFD

研究人员开发了一种新的基于图的神经算子,称为 Gauge-Invariant Spectral Transformer (GIST),旨在加速赛车设计的空气动力学模拟。该模型在一个由专家验证的高保真赛车空气动力学数据集上进行训练,解决了先前在简单形状上训练的 AI 模型的局限性。GIST 展现了最先进的准确性,并有潜力在工业赛车工作流程中实现交互式设计空间探索。 AI

影响 加速赛车空气动力学模拟,实现更快的迭代设计和探索。

排序理由 介绍特定领域新模型和数据集的学术论文。

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

Faster by Design: Interactive Aerodynamics via Neural Surrogates Trained on Expert-Validated CFD

报道来源 [1]

  1. arXiv cs.LG TIER_1 English(EN) · Nicholas Thumiger, Andrea Bartezzaghi, Mattia Rigotti, Cezary Skura, Thomas Frick, Elisa Serioli, Fabrizio Arbucci, A. Cristiano I. Malossi ·

    Faster by Design: Interactive Aerodynamics via Neural Surrogates Trained on Expert-Validated CFD

    arXiv:2604.18491v2 Announce Type: replace Abstract: Computational Fluid Dynamics (CFD) is central to race-car aerodynamic development, yet its cost -- tens of thousands of core-hours per high-fidelity evaluation -- severely limits the design space exploration feasible within real…