Characteristic‐Aligned Control‐Volume Physics‐Informed Neural Network for Flood‐Wave Propagation and Roughness Estimation

Физически информированная нейронная сеть с контрольными объёмами, согласованными с характеристиками, для распространения паводковых волн и оценки шероховатости
Zhongjing Wang, Ziyan Lu, Tao Wang, Jingjing Duan, Yongnan Zhu, Xusheng He, Anjun Deng
2026-10-01

Saint-Venant equationscontrol-volume conservationflood-wave propagationphysics-informed neural networksroughness estimation
Abstract Physics‐informed neural networks (PINNs) provide a promising framework for river hydrodynamic simulation and parameter inversion under sparse observations. However, spatial heterogeneity in channel geometry and roughness causes strong variations in flood‐wave propagation and parameter sensitivity, challenging local conservation and joint state‐parameter estimation. This study proposes a Characteristic‐aligned Control‐volume Physics‐Informed Neural Network for Roughness Estimation (CC‐PINN‐RE). It constructs dynamic diamond‐shaped space‐time control volumes from local characteristic speeds of the Saint‐Venant equations, enforces mass and momentum conservation in integral form, and combines adaptive control‐volume management with local‐sensitivity‐weighted roughness regularization. The method is evaluated on three one‐dimensional channel cases of increasing complexity against strong‐form PINN‐RE and fixed‐control‐volume cvPINN‐RE. CC‐PINN‐RE achieves the lowest errors in eight of nine relative metrics for water depth, discharge, and roughness. In Case 3, where channel width, bed slope, and roughness vary spatially, it reduces the corresponding errors by 57.0%, 82.2%, and 57.3% relative to PINN‐RE and reduces the maximum absolute error in peak water depth by 69.7%. Ablation experiments with three random seeds support the contributions of the three proposed components. Observation‐deployment experiments show that increasing the number of monitoring stations is generally more effective than increasing temporal sampling frequency at individual stations. Under 0%–2% observational noise in Case 3, the model remains stable. These results indicate that CC‐PINN‐RE provides a physically consistent framework for integrating characteristic propagation, local conservation, and parameter inversion, offering a useful basis for joint hydrodynamic reconstruction and spatially varying roughness estimation in heterogeneous river channels.
1
Across three increasingly complex one-dimensional channel cases, CC-PINN-RE achieves the lowest errors in eight of nine relative metrics versus strong-form PINN-RE and fixed-volume cvPINN-RE.
2
CC-PINN-RE dynamically constructs characteristic-aligned diamond-shaped space-time control volumes using local Saint-Venant wave speeds.
3
In the heterogeneous Case 3, CC-PINN-RE reduces width, slope, and roughness errors by 57.0%, 82.2%, and 57.3%, respectively, versus PINN-RE, while reducing peak-depth maximum error by 69.7%.
4
More monitoring stations generally improve performance more effectively than increased temporal sampling at individual stations, and the model remains stable under 0%–2% observational noise.
5
The method enforces integral mass and momentum conservation and applies local-sensitivity-weighted regularization for spatially varying roughness estimation.

One-dimensional flood-wave propagation and spatially heterogeneous river channels governed by the Saint-Venant equations

Hydrodynamic state reconstruction and spatially varying channel-roughness estimation, including water depth, discharge, mass–momentum conservation, and parameter sensitivity

Publication Details
Publication Date
2026-10-01
Journal
Publisher
ISSN
Cited by
0
Access Type
Author Information
Authors
Zhongjing Wang
Ziyan Lu
Tao Wang
Jingjing Duan
Yongnan Zhu
Xusheng He
Anjun Deng
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%