Predicting drag on rough surfaces by transfer learning of empirical correlations

TL;DR

This paper introduces a transfer learning framework that leverages empirical correlations to improve neural network predictions of drag on rough surfaces, especially when limited high-fidelity data is available.

Contribution

It presents a novel transfer learning approach that incorporates empirical correlations to enhance neural network performance in drag prediction with scarce data.

Findings

Transfer learning significantly improves drag prediction accuracy.

Empirical correlations provide valuable approximate physics knowledge.

Framework is effective with limited high-fidelity simulation data.

Abstract

Recent developments in neural networks have shown the potential of estimating drag on irregular rough surfaces. Nevertheless, the difficulty of obtaining a large high-fidelity dataset to train neural networks is deterring their use in practical applications. In this study, we propose a transfer learning framework to model the drag on irregular rough surfaces even with a limited amount of direct numerical simulations. We show that transfer learning of empirical correlations, reported in the literature, can significantly improve the performance of neural networks for drag prediction. This is because empirical correlations include `approximate knowledge' of the drag dependency in high-fidelity physics. The `approximate knowledge' allows neural networks to learn the surface statistics known to affect drag more efficiently. The developed framework can be applied to applications where acquiring a large dataset is difficult, but empirical correlations have been reported.