用混合模型加速高分辨率3D物理模拟，支持精准预测与随机生成。
P3D: Scalable Neural Surrogates for High-Resolution 3D Physics Simulations with Global Context
- 采用CNN-Transformer混合结构，兼顾局部细节与全局依赖。
- 在512³分辨率下实现湍流模拟，速度与精度优于现有方法。
- 可训练为扩散模型，生成不同雷诺数下的湍流样本。
我们提出一种可扩展的框架，用于学习高分辨率3D物理模拟的确定性与概率性神经代理模型。设计了一种面向3D物理模拟的混合CNN-Transformer骨干网络，在速度和精度上显著优于现有架构。该网络可在模拟域的小块数据上预训练，通过融合获得全局解，且可通过快速可扩展的序列到序列模型引入长程依赖。此设计使大模型在高分辨率数据集上的训练得以降低内存与计算需求。我们在一组基准方法上评估了该骨干网络，目标是同时学习14类不同偏微分方程（PDE）在3D中的动力学行为。我们展示了模型在空间分辨率达512³的各向同性湍流模拟中的可扩展性。最后，通过将其训练为扩散模型，成功生成了不同雷诺数下高度湍流3D通道流的概率样本，准确捕捉了底层流动统计特性。
原文摘要 · Abstract (English)
We present a scalable framework for learning deterministic and probabilistic neural surrogates for high-resolution 3D physics simulations. We introduce a hybrid CNN-Transformer backbone architecture targeted for 3D physics simulations, which significantly outperforms existing architectures in terms of speed and accuracy. Our proposed network can be pretrained on small patches of the simulation domain, which can be fused to obtain a global solution, optionally guided via a fast and scalable sequence-to-sequence model to include long-range dependencies. This setup allows for training large-scale models with reduced memory and compute requirements for high-resolution datasets. We evaluate our backbone architecture against a large set of baseline methods with the objective to simultaneously learn the dynamics of 14 different types of PDEs in 3D. We demonstrate how to scale our model to high-resolution isotropic turbulence with spatial resolutions of up to $512^3$. Finally, we demonstrate the versatility of our network by training it as a diffusion model to produce probabilistic samples of highly turbulent 3D channel flows across varying Reynolds numbers, accurately capturing the underlying flow statistics.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。