Going with the Speed of Sound: Pushing Neural Surrogates into Highly-turbulent Transonic Regimes

2025-11-26

AI Summary

Neural surrogate models for aerospace transonic flow are limited by existing datasets—mostly 2D, neglecting critical 3D effects (e.g., tip vortices) and strong nonlinear compressibility features. Method: We introduce the first large-scale 3D wing transonic CFD dataset (~30,000 simulations), systematically varying geometric and freestream parameters while explicitly resolving complex physics including high-turbulence regimes and shock–boundary-layer interactions. We propose end-to-end architectures—AB-UPT and Transolver—that jointly model volumetric and surface flow fields, ensuring physical consistency in predicting lift/drag coefficients and Pareto-optimal aerodynamic fronts. Contribution/Results: Experiments demonstrate strong out-of-distribution generalization: models accurately predict aerodynamic performance for unseen wing configurations, enabling efficient transonic aerodynamic optimization with significantly reduced computational cost.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Search and Optimization: Sampling/Simulation-based Search
• Computer Vision: 3D Computer Vision

Application Category

• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

The widespread use of neural surrogates in automotive aerodynamics, enabled by datasets such as DrivAerML and DrivAerNet++, has primarily focused on bluff-body flows with large wakes. Extending these methods to aerospace, particularly in the transonic regime, remains challenging due to the high level of non-linearity of compressible flows and 3D effects such as wingtip vortices. Existing aerospace datasets predominantly focus on 2D airfoils, neglecting these critical 3D phenomena. To address this gap, we present a new dataset of CFD simulations for 3D wings in the transonic regime. The dataset comprises volumetric and surface-level fields for around $30,000$ samples with unique geometry and inflow conditions. This allows computation of lift and drag coefficients, providing a foundation for data-driven aerodynamic optimization of the drag-lift Pareto front. We evaluate several state-of-the-art neural surrogates on our dataset, including Transolver and AB-UPT, focusing on their out-of-distribution (OOD) generalization over geometry and inflow variations. AB-UPT demonstrates strong performance for transonic flowfields and reproduces physically consistent drag-lift Pareto fronts even for unseen wing configurations. Our results demonstrate that AB-UPT can approximate drag-lift Pareto fronts for unseen geometries, highlighting its potential as an efficient and effective tool for rapid aerodynamic design exploration. To facilitate future research, we open-source our dataset at https://huggingface.co/datasets/EmmiAI/Emmi-Wing.

Problem

• Extending neural surrogates to 3D transonic aerospace flows with wingtip vortices
• Addressing the gap in 3D datasets for compressible turbulent transonic regimes
• Evaluating neural surrogates' generalization for aerodynamic drag-lift optimization

Innovation

• New dataset for 3D wings in transonic regime
• Evaluates neural surrogates like AB-UPT on OOD generalization
• AB-UPT approximates drag-lift Pareto fronts for unseen geometries