Comparing recurrent and convolutional neural networks for predicting wave propagation

TL;DR

This paper compares recurrent and convolutional neural networks for predicting surface wave propagation governed by Saint-Venant equations, demonstrating improved long-term prediction and generalization with faster inference.

Contribution

It provides a systematic comparison showing convolutional networks perform as well as recurrent ones for wave prediction and assesses their generalization in different settings.

Findings

Convolutional networks match recurrent networks in accuracy.

Both models outperform previous methods in long-term prediction.

Networks generalize well to longer time frames and different physical conditions.

Abstract

Dynamical systems can be modelled by partial differential equations and numerical computations are used everywhere in science and engineering. In this work, we investigate the performance of recurrent and convolutional deep neural network architectures to predict the surface waves. The system is governed by the Saint-Venant equations. We improve on the long-term prediction over previous methods while keeping the inference time at a fraction of numerical simulations. We also show that convolutional networks perform at least as well as recurrent networks in this task. Finally, we assess the generalisation capability of each network by extrapolating in longer time-frames and in different physical settings.