Physics-Informed Neural Network Surrogate Models for River Stage Prediction
Суррогатные модели на основе физически информированных нейронных сетей для прогнозирования уровня воды в реке
2025-03-21
SCID: 54.1/k9rb4ry3
Discuss with AI
HEC-RASPhysics-Informed Neural NetworksSaint-Venant equationsriver stage predictionsurrogate models
Figures from the paper
Abstract (AI)
This work investigates the feasibility of using Physics-Informed Neural Networks (PINNs) as surrogate models for river stage prediction, aiming to reduce computational cost while maintaining predictive accuracy. Our primary contribution demonstrates that PINNs can successfully approximate HEC-RAS numerical solutions when trained on a single river, achieving strong predictive accuracy with generally low relative errors, though some river segments exhibit higher deviations. By integrating the governing Saint-Venant equations into the learning process, the proposed PINN-based surrogate model enforces physical consistency and significantly improves computational efficiency compared to HEC-RAS. We evaluate the model's performance in terms of accuracy and computational speed, demonstrating that it closely approximates HEC-RAS predictions while enabling real-time inference. These results highlight the potential of PINNs as effective surrogate models for single-river hydrodynamics, offering a promising alternative for computationally efficient river stage forecasting. Future work will explore techniques to enhance PINN training stability and robustness across a more generalized multi-river model.
Key Findings
1
Embedding the Saint-Venant equations enforces physical consistency in the learned river-stage predictions.
2
Generalization to multiple rivers remains future work, with training stability and robustness identified as challenges.
3
Physics-Informed Neural Networks can successfully approximate HEC-RAS river-stage solutions when trained on a single river.
4
The PINN surrogate substantially reduces computational cost relative to HEC-RAS and supports real-time inference.
5
The surrogate model achieves strong predictive accuracy with generally low relative errors, although deviations are higher in some river segments.
Research Object
single-river hydrodynamics and river stage dynamics represented by HEC-RAS numerical solutions
Research Subject
the predictive accuracy, physical consistency, computational efficiency, and real-time inference performance of PINN-based surrogate models for river stage
Publication Details
Publication Date
2025-03-21
Journal
Publisher
ISSN
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest