Operator Learning for Surrogate Modeling of Wave-Induced Forces from Sea Surface Waves

TL;DR

This paper introduces a Deep Operator Network surrogate for the SWAN wave model, enabling efficient and accurate predictions of wave-induced forces in various scenarios, including realistic coastal simulations.

Contribution

The work demonstrates the effectiveness of DeepONets as surrogates for complex wave models, reducing computational costs while maintaining high accuracy.

Findings

High accuracy in predicting radiation stress components and wave height in steady-state simulations.

Successful application to multiple 1-D and 2-D numerical examples with variable conditions.

Effective surrogate modeling in a realistic coastal scenario in Duck, NC.

Abstract

Wave setup plays a significant role in transferring wave-induced energy to currents and causing an increase in water elevation. This excess momentum flux, known as radiation stress, motivates the coupling of circulation models with wave models to improve the accuracy of storm surge prediction, however, traditional numerical wave models are complex and computationally expensive. As a result, in practical coupled simulations, wave models are often executed at much coarser temporal resolution than circulation models. In this work, we explore the use of Deep Operator Networks (DeepONets) as a surrogate for the Simulating WAves Nearshore (SWAN) numerical wave model. The proposed surrogate model was tested on three distinct 1-D and 2-D steady-state numerical examples with variable boundary wave conditions and wind fields. When applied to a realistic numerical example of steady state wave simulation in Duck, NC, the model achieved consistently high accuracy in predicting the components of the radiation stress gradient and the significant wave height across representative scenarios.