用图神经网络构建洪水预测代理模型，实现秒级快速预报。
Toward an Operational GNN-Based Multimesh Surrogate for Fast Flood Forecasting
- 基于投影网格与多网格连接设计图神经网络，保持高精度监督。
- 结合流量特征和推演训练，6小时预报仅需0.4秒，比原模型快2700倍。
- 适合需要实时洪水模拟的市政应急与防灾决策场景。
实际洪水预报仍依赖高保真二维水力求解器，但其计算耗时严重制约大规模城市洪泛区的快速决策支持。与此同时，基于AI的代理模型在计算物理领域展现出加速高成本仿真的潜力。本文以法国下塔特河为研究区域，从生产级的高分辨率非结构化有限元网格Telemac2D模型（节点数超4×10⁵）出发，构建涵盖多种典型水文过程与峰值流量的合成洪水事件数据库。在此基础上，提出一种基于投影网格与多网格连通性的图神经网络代理模型。投影策略保障训练可行性并保留原始仿真高精度监督，多网格结构扩展有效空间感受野而不增加网络深度。进一步研究了显式流量特征Q(t)及前向推演训练对长时自回归滚动预测的影响。实验表明：在边界驱动场景中，条件化Q(t)至关重要；多网格连接在模型充分条件化后带来额外增益；前向推演进一步提升滚动稳定性。综合配置（含Q(t)、多网格连接、前向推演）表现最优。该模型在代理网格上的水力变量及插值至25m规则网格的淹没图上均优于基准方案。在本案例中，代理模型在单张NVIDIA A100 GPU上约0.4秒完成6小时预报，相较56核CPU运行的参考模拟（约180分钟）提速2700倍。结果表明，图基代理模型可作为工业级水力求解器在业务洪水制图中的实用补充。
原文摘要 · Abstract (English)
Operational flood forecasting still relies on high-fidelity two-dimensional hydraulic solvers, but their runtime can be prohibitive for rapid decision support on large urban floodplains. In parallel, AI-based surrogate models have shown strong potential in several areas of computational physics for accelerating otherwise expensive high-fidelity simulations. We address this issue on the lower Têt River (France), starting from a production-grade Telemac2D model defined on a high-resolution unstructured finite-element mesh with more than $4\times 10^5$ nodes. From this setup, we build a learning-ready database of synthetic but operationally grounded flood events covering several representative hydrograph families and peak discharges. On top of this database, we develop a graph-neural surrogate based on projected meshes and multimesh connectivity. The projected-mesh strategy keeps training tractable while preserving high-fidelity supervision from the original Telemac simulations, and the multimesh construction enlarges the effective spatial receptive field without increasing network depth. We further study the effect of an explicit discharge feature $Q(t)$ and of pushforward training for long autoregressive rollouts. The experiments show that conditioning on $Q(t)$ is essential in this boundary-driven setting, that multimesh connectivity brings additional gains once the model is properly conditioned, and that pushforward further improves rollout stability. Among the tested configurations, the combination of $Q(t)$, multimesh connectivity, and pushforward provides the best overall results. These gains are observed both on hydraulic variables over the surrogate mesh and on inundation maps interpolated onto a common $25\,\mathrm{m}$ regular grid and compared against the original high-resolution Telemac solution. On the studied case, the learned surrogate produces 6-hour predictions in about $0.4\,\mathrm{s}$ on a single NVIDIA A100 GPU, compared with about $180\,\mathrm{min}$ on 56 CPU cores for the reference simulation. These results support graph-based surrogates as practical complements to industrial hydraulic solvers for operational flood mapping.
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。