Inverse airfoil design method for generating varieties of smooth airfoils using conditional WGAN-gp

TL;DR

This paper introduces a conditional Wasserstein GAN with gradient penalty that directly generates smooth airfoil shapes meeting specific lift coefficients, eliminating the need for post-processing smoothing methods.

Contribution

The study presents a novel CWGAN-GP model capable of producing smooth, shape-compliant airfoils directly, advancing airfoil design automation.

Findings

Generated airfoils are smooth without additional smoothing.

Shapes meet specified lift coefficient requirements.

Model outperforms traditional GANs in shape quality.

Abstract

Machine learning models are recently utilized for airfoil shape generation methods. It is desired to obtain airfoil shapes that satisfies required lift coefficient. Generative adversarial networks (GAN) output reasonable airfoil shapes. However, shapes obtained from ordinal GAN models are not smooth, and they need smoothing before flow analysis. Therefore, the models need to be coupled with Bezier curves or other smoothing methods to obtain smooth shapes. Generating shapes without any smoothing methods is challenging. In this study, we employed conditional Wasserstein GAN with gradient penalty (CWGAN-GP) to generate airfoil shapes, and the obtained shapes are as smooth as those obtained using smoothing methods. With the proposed method, no additional smoothing method is needed to generate airfoils. Moreover, the proposed model outputs shapes that satisfy the lift coefficient requirements.