Toward a Foundation-Model Paradigm for Aerodynamic Prediction in Three-Dimensional Design

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 pretraining 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 pretrained on SuperWing, a dataset of nearly 30,000 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 a 0.36% error on surface-flow prediction, reducing error by 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 pretrained models at https://github.com/tum-pbs/AeroTransformer . An interactive design tool is also built on the pretrained model and is available online at https://webwing.pbs.cit.tum.de .