arXiv：cs.LG· Yidi Wang, Yunhe Zhang, Jiawei Gu, Ziyue Qiao, Pengyang Wang·· 4 小时前AI 评分31
Sculpt：面向城市风场预测的嵌套势框架，实现运动学可容许速度场
Shaping the Wind: Nested Potentials for Kinematically Admissible Urban Wind Prediction
AI 导读
研究者提出 Sculpt 嵌套势参数化框架，通过三维交错网格上体积矢量势的离散旋度生成无散度速度更新，并用共享标量势约束边界值以同时满足壁面不可穿透性，无需逐步压力投影。该框架将多分辨率体积势参数化以缓解旋度反向传播对大尺度梯度信号的衰减，并发布覆盖多种城市形态与入流条件的 LES 数据集 UrbanWindFlow，用于联合评估预测精度与运动学可容许性。
正文
Abstract:Predicting transient urban winds is fundamental to understanding urban microclimates and designing climate-resilient cities. Building-resolving large-eddy simulation produces detailed incompressible urban wind fields at substantial computational cost for each layout. Neural surrogates offer a faster alternative by learning to predict the evolution of velocity fields. However, minimizing velocity prediction error does not guarantee local mass conservation and wall impermeability, which together define kinematic admissibility. This limitation stems from an unconstrained output representation: geometry conditioning guides predictions but does not restrict them to admissible velocity fields. Correcting boundary violations in these outputs changes the flux balance in adjacent fluid cells and may consequently compromise local mass conservation. To address the challenge, we propose Sculpt, a nested potential framework that builds the coupled, geometry-dependent constraints directly into its parameterization. This nested parameterization generates divergence-free velocity updates through the discrete curl of a volume vector potential on the native three-dimensional staggered grid. A shared scalar potential constrains the vector potential's boundary values so that the same operator also enforces impermeability, without a per-step pressure projection. Because backpropagation through this curl attenuates large-scale gradient signals, we parameterize the volume potential at multiple resolutions to better capture large-scale flow structures. We introduce UrbanWindFlow, an LES dataset spanning urban morphologies and inflow conditions, to evaluate accuracy and kinematic admissibility together.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.07033 [cs.LG] |
| (or arXiv:2610.07033v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07033 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yidi Wang [view email]
[v1]
Sun, 4 Oct 2026 19:22:19 UTC (5,461 KB)
来源：arXiv：cs.LG · arxiv.org