Physics-Informed Neural Networks for Modeling Water Flows in a River Channel
Физически информированные нейронные сети для моделирования потоков воды в русле реки
2022-08-19
SCID: 54.1/rqscf5t3
Discuss with AI
Hydrological forecastingParameter identificationPhysics-informed neural networksRiver channel water flowSaint-Venant equations
Figures from the paper
Abstract (AI)
The impacts incurred by floods regularly affect the planet's population, inflicting social and economic problems. Optimal control strategies based on reservoir management may aid in controlling floods and mitigating the resulting damage. To this end, an accurate dynamic representation of water systems is needed. In practice, flood control strategies rely on hydrological forecasting models obtained from conceptual or data-driven methods. Encouraged by recent works, this research proposes a novel surrogate model for water flow in a river channel based on physics-informed neural networks (PINNs). This approach achieved promising results regarding the assimilation of real-data measurements and the parameter identification of differential equations that govern the underlying dynamics. This article investigates PINN performance in a simulated environment built directly from a configuration of the Saint-Venant equations. The objective is to create a suitable model with high prediction accuracy and scientifically consistent behavior for use in real-time applications. The experiments revealed promising results for hydrological modeling and presented alternatives to solve the main challenges found in conventional methods while assisting in synthesizing real-world representations.
Key Findings
1
Experiments based on a simulated Saint-Venant equation configuration showed promising hydrological modeling performance and high prediction accuracy.
2
PINNs are presented as an alternative for addressing challenges associated with conventional conceptual and data-driven hydrological forecasting methods.
3
The PINN successfully assimilated real-data measurements and identified parameters in the differential equations governing water-flow dynamics.
4
The model is intended to provide scientifically consistent, real-time representations that can support reservoir-based flood control strategies.
5
The study proposes a physics-informed neural network surrogate model for representing water flow dynamics in river channels.
Research Object
water flow in a river channel
Research Subject
the prediction accuracy, physical consistency, data assimilation, and differential-equation parameter identification of a PINN-based surrogate model for river-channel flow
Publication Details
Publication Date
2022-08-19
Journal
Publisher
ISSN
Access Type
Author Information
Download PDF
Subscribe to digest