HypNO用图神经网络精准模拟交通流中的激波与传播。
HypNO: A Graph-Based Neural Operator with Physics-Informed Message Passing for Hyperbolic Conservation Laws
- 构建时空细胞图，用物理约束消息传递建模
- 在LWR和ARZ模型上准确预测激波与解的突变
- 适合需要高保真非线性波动模拟的研究者
我们提出HypNO，一种基于图的神经算子，用于标量双曲守恒律。HypNO直接作用于有限体积单元的时空图，采用邻接因子化、物理信息驱动的消息传递机制，以尊重激波附近的迎风性和熵适定性。我们在Lighthill-Whitham-Richards（LWR）和Aw-Rascle-Zhang（ARZ）交通流模型上测试该架构，这些模型因同时存在全局传输与激波形成而成为算子学习方法的严苛压力测试。HypNO在多种初始条件下均能准确预测解的快照，并成功捕捉解中的激波与间断。
原文摘要 · Abstract (English)
We introduce HypNO, a graph-based neural operator for scalar hyperbolic conservation laws. HypNO operates directly on a space-time graph of finite-volume cells and uses adjacency-factored, physics-informed message passing to respect upwinding and entropy admissibility near shocks. We benchmark the architecture on the Lighthill-Whitham-Richards (LWR) and Aw-Rascle-Zhang (ARZ) traffic-flow models, a stress test for operator-learning methods because of their simultaneous global transport and shock formation. HypNO predicts solution snapshots accurately across a range of initial conditions while capturing the shocks and discontinuities of the solution.
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。