Physics-Informed Neural Networks for Depth-Averaged Avalanche Dynamics

License: See LICENSE file · Updated 2026-09-28
My rating

This paper presents a physics-informed neural network framework for modeling avalanche motion using the Savage-Hutter model of depth-averaged granular flow. The framework is first verified against analytical solutions for three one-dimensional problems of increasing complexity, then evaluated through two-dimensional experimental validation. Its intended application is avalanche hazard assessment in mountainous terrain.

Tags: