AirfoilGen: A valid-by-construction and performance-aware latent diffusion model for airfoil generation

2026-05-19

AI Summary

This study addresses the challenge in existing deep generative methods for airfoil design, which often struggle to simultaneously ensure geometric validity and precise aerodynamic performance control. To overcome this limitation, the authors propose AirfoilGen—a performance-conditioned latent diffusion model based on a novel circle-sweep representation. This approach enforces geometric validity through circle-sweep constraints and integrates target lift-to-drag coefficients directly into the latent space for conditional generation. The work contributes a new parameterization scheme for airfoils via circle-sweep representation, a performance-aware latent diffusion framework, and a large-scale dataset comprising over 200,000 airfoils. Experimental results demonstrate that AirfoilGen significantly improves geometric validity and achieves a 98.41% accuracy in aerodynamic performance control, substantially outperforming current state-of-the-art methods.

Abstract

Airfoil shape design is a fundamental task in aerospace engineering, with a direct impact on flight stability and fuel consumption. Deep learning has recently emerged as a promising tool for this task, but existing deep generative approaches remain limited in both geometric validity and physical controllability. They offer little control over the generated shapes, yielding invalid geometries, and they typically do not condition effectively on aerodynamic performance. To address these issues, this paper proposes AirfoilGen, a valid-by-construction and performance-aware latent diffusion model for airfoil. It first introduces a novel airfoil representation scheme, the circle sweeping representation, to constrain the generative process so that output shapes respect essential airfoil characteristics. It then enables explicit control over aerodynamic performance (e.g., lift and drag coefficients) by operating in a learned latent space: a transformer model encodes airfoil shapes into vector embeddings, and a conditional diffusion model denoises Gaussian noise into these latent embeddings while incorporating target aerodynamic performance. In addition, this paper presents a new dataset of over 200,000 airfoils, which is substantially larger than the widely used UIUC airfoil dataset (1,650 airfoils) and more suitable for training modern deep generative models. Experiments demonstrate that AirfoilGen enables airfoil generation with far greater geometric validity and aerodynamic performance controllability than previously achievable, with an average performance-conditioning accuracy of 98.41%.

Problem

• airfoil generation
• geometric validity
• aerodynamic performance
• deep generative models
• conditional generation

Innovation

• latent diffusion model
• valid-by-construction
• performance-aware generation
• circle sweeping representation
• airfoil design