Physics-informed neural networks improve avalanche flow predictions
Physics-Informed Neural Networks for Depth-Averaged Avalanche Dynamics
Machine Learning
Summary
Predicting how avalanches move is important to keep people safe in mountainous areas. The authors used a special type of AI called physics-informed neural networks to model avalanche flows based on known physical laws. They tested their method first in simple one-dimensional cases and then on two-dimensional lab experiments of granular flows on slopes. Their approach combined physics knowledge with some real measurements to make accurate predictions of flow height and speed. This could help improve avalanche hazard assessments.
What this means in practice
- •For mountain hazard assessment teams: Provide more accurate and timely predictions of avalanche motion using physics-informed neural networks combined with limited field data.
- •For civil engineering modelers: Incorporate hybrid physics-informed neural network models to simulate and design structures resilient to granular flow hazards like avalanches and landslides.
Authors
Pradyumn Singh Sikarwar, Vishal Sharma, Gaurav Bhutani
Abstract
Accurate prediction of avalanche motion is essential for hazard assessment in mountainous terrain. This study develops and evaluates a physics-informed neural network (PINN) framework for the Savage-Hutter model of depth-averaged granular flow, progressing from 1D analytical verification to 2D experimental validation. First, three 1D problems of increasing complexity were verified against the analytical solution: height prediction with prescribed velocity, velocity prediction with prescribed height, and coupled prediction of both fields using the conservative formulation. The decoupled tests accurately reconstructed the spatio-temporal evolution of each field when the other was prescribed. The coupled formulation learned both fields without prescribed data, achieving mean height and velocity RMSEs of 0.043 and 0.079 in non-dimensional units. A hyperparameter sensitivity study evaluated the effects of network depth, width, collocation density, learning rate, and epochs. The framework was then extended to 2D and validated against laboratory experiments of a cylindrical granular pile collapsing on an inclined plane, with TITAN2D providing numerical comparisons. Purely physics-based training converged to the trivial zero solution; augmenting the loss with 10 sparse training points from final deposit profiles produced a physics-informed, data-assisted hybrid framework. Peak flow depth, depth-averaged velocity, RMSE, and wetted-area IoU evaluated global and local agreement. Global height RMSE ranged from 2.7 to 6.7 mm across four experimental cases, while mean wetted-area IoU ranged from 69 to 81 %, demonstrating consistent performance across variations in pile mass and slope angle.