Towards a Foundation-Model Paradigm for Aerodynamic Prediction in Three-dimensional Design

2026-04-20

AI Summary

This study addresses the challenge of achieving high-fidelity three-dimensional aerodynamic prediction, which is hindered by the prohibitive computational cost of generating simulation data and thus limits efficient optimization of complex configurations. To overcome this, the work introduces the foundational model paradigm to the field for the first time, proposing a two-stage modeling approach: a customized Transformer architecture, AeroTransformer, is first pre-trained on SuperWing—a large-scale, diverse geometric dataset comprising nearly 30,000 samples—and subsequently fine-tuned with only a small number of task-specific examples. Remarkably, with just 450 fine-tuning samples, the method achieves a surface flowfield prediction error as low as 0.36%, representing an 84.2% reduction compared to training from scratch. The authors publicly release the dataset, models, and an interactive online design tool, substantially reducing reliance on expensive simulations.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Search and Optimization: Sampling/Simulation-based Search
• Computer Vision: Large Vision Models

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Accurate machine-learning models for aerodynamic prediction are essential for accelerating shape optimization, yet remain challenging to develop for complex three-dimensional configurations due to the high cost of generating training data. This work introduces a methodology for efficiently constructing accurate surrogate models for design purposes by first pre-training a large-scale model on diverse geometries and then fine-tuning it with a few more detailed task-specific samples. A Transformer-based architecture, AeroTransformer, is developed and tailored for large-scale training to learn aerodynamics. The methodology is evaluated on transonic wings, where the model is pre-trained on SuperWing, a dataset of nearly 30000 samples with broad geometric diversity, and subsequently fine-tuned to handle specific wing shapes perturbed from the Common Research Model. Results show that, with 450 task-specific samples, the proposed methodology achieves 0.36% error on surface-flow prediction, reducing 84.2% compared to training from scratch. The influence of model configurations and training strategies is also systematically studied to provide guidance on effectively training and deploying such models under limited data and computational budgets. To facilitate reuse, we release the datasets and the pre-trained models at https://github.com/tum-pbs/AeroTransformer. An interactive design tool is also built on the pre-trained model and is available online at https://webwing.pbs.cit.tum.de.

Problem

• aerodynamic prediction
• three-dimensional design
• surrogate modeling
• data efficiency
• shape optimization

Innovation

• foundation model
• aerodynamic prediction
• Transformer architecture
• transfer learning
• surrogate modeling