Finite Volume-Informed Neural Network Framework for 2D Shallow Water Equations: Rugged Loss Landscapes and the Importance of Data Guidance

Xiaofeng Liu

arXiv:2605.11001 · 2026-05-13 공개 · arXiv · PDF

surrogate-modeling loss-landscapes shallow-water-equations finite-volume pinns data-guided roe-riemann-solver unstructured-meshes

Abstract

Physics-informed neural networks (PINNs) are a simple surrogate-modelling paradigm for partial differential equations, but their standard strong-form residual formulation is ill suited to the shallow water equations (SWE). It cannot enforce local conservation, handle discontinuities, or leverage the boundary-conforming unstructured meshes used in real-world applications. We introduce ``Data-Guided FVM-PINN'', a framework that replaces the strong-form residual with a differentiable, well-balanced Roe Riemann-solver finite-volume (FVM) loss evaluated on unstructured meshes. The major finding is that physics-only FVM-PINN training often fails on realistic 2D problems: the network collapses to a trivial low-momentum state that nearly satisfies the FVM-PINN residual but bears no resemblance to the true flow. A loss-landscape diagnostic shows that the FVM-PINN loss at zero momentum is only about $7\times$ larger than at the trained solution, a shallow basin that an ordinary optimizer falls into; adding even sparse data turns this into a $310\times$ separation, breaking the degeneracy. On a 2D block-in-channel benchmark, just $200$ random velocity measurements drop the velocity-field $L_2$ error by $22\times$ versus physics-only; $50$ measurements still deliver a $7\times$ reduction. A controlled ablation isolates the contribution of the FVM-PINN loss: it reduces velocity-field $L_2$ by $\sim$$23\%$ in the sparse-data regime and is essentially neutral when dense reference data is available. On a real-world Savannah River reach ($1306$ cells, $3600$~s simulation, five Manning zones), the framework constructs an accurate surrogate from SRH-2D anchor data, with time-window decomposition reducing error monotonically via progressive initial-condition handoff.

한국어 요약

📋 한 줄 요약

**[과학 머신러닝 / PINN]** 2D 천수방정식을 위해 강형식 잔차 대신 Roe Riemann 솔버 기반 유한체적법(FVM) 손실을 사용하고, 손실 풍경의 평탄한 분지를 깨기 위해 희박한 데이터로 가이드하는 Data-Guided FVM-PINN 프레임워크를 제안한다.

🎯 핵심 기여도

💡 핵심 아이디어

물리 손실만으로는 해 공간에 가짜 최소점이 존재하므로, 희박하지만 의미 있는 데이터를 함께 사용하여 최적화기를 올바른 해 분지로 유도해야 한다. 손실 풍경 자체를 진단 도구로 사용하여 학습 실패 원인을 정량화한다.

🔬 기술적 접근법

📊 주요 결과

💭 의의 및 한계

**의의**: PINN이 실제 수리/홍수 시뮬레이션에 적용되기 어렵던 문제(보존, 불연속, 메쉬)를 FVM과 결합해 해결하고, 데이터-물리 결합의 필요성을 손실 풍경으로 정량화한다. **한계**: 여전히 정답에 가까운 희소 데이터가 필요하며, 진정한 무측정 환경에서는 trivial 해로 붕괴할 위험이 있다.

🚀 실용적 활용