AUTHOR=Guglielmo Gianmarco , Montessori Andrea , Tucny Jean-Michel , La Rocca Michele , Prestininzi Pietro 

TITLE=A priori physical information to aid generalization capabilities of neural networks for hydraulic modeling

JOURNAL=Frontiers in Complex Systems

VOLUME=Volume 2 - 2024

YEAR=2025

URL=https://www.frontiersin.org/journals/complex-systems/articles/10.3389/fcpxs.2024.1508091

DOI=10.3389/fcpxs.2024.1508091

ISSN=2813-6187

ABSTRACT=The application of Neural Networks to river hydraulics and flood mapping is fledgling, despite the field suffering from data scarcity, a challenge for machine learning techniques. Consequently, many purely data-driven Neural Networks have shown limited capabilities when tasked with predicting new scenarios. In this work, we propose introducing physical information into the training phase in the form of a regularization term. Whereas this idea is formally borrowed from Physics-Informed Neural Networks, the proposed methodology does not necessarily resort to PDEs, making it suitable for scenarios with significant epistemic uncertainties, such as river hydraulics. The method enriches the information content of the dataset and appears highly versatile. It shows improved predictive capabilities for a highly controllable, synthetic hydraulic problem, even when extrapolating beyond the boundaries of the training dataset and in data-scarce scenarios. Therefore, our study lays the groundwork for future employment on real datasets from complex applications.