Faster by Design: Interactive Aerodynamics via Neural Surrogates Trained on Expert-Validated CFD

2026-04-20

AI Summary

High-fidelity computational fluid dynamics (CFD) simulations are prohibitively expensive, limiting the efficient exploration of design spaces in race car aerodynamics. To address this challenge, this work introduces the first expert-validated, industrial-scale high-fidelity CFD dataset for an LMP2 race car and proposes GIST, a graph neural operator that enables interactive surrogate-model-driven design for the first time. GIST leverages spectral embedding to encode mesh connectivity, ensuring discretization invariance and supporting linear scalability to large-scale meshes. Experiments demonstrate that GIST achieves state-of-the-art accuracy on both public benchmarks and the newly curated race car dataset, with predictions sufficiently accurate for early-stage aerodynamic design. These results validate the feasibility of replacing conventional CFD with a surrogate model for real-time aerodynamic queries.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Search and Optimization: Sampling/Simulation-based Search
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT

Abstract

Computational Fluid Dynamics (CFD) is central to race-car aerodynamic development, yet its cost -- tens of thousands of core-hours per high-fidelity evaluation -- severely limits the design space exploration feasible within realistic budgets. AI-based surrogate models promise to alleviate this bottleneck, but progress has been constrained by the limited complexity of public datasets, which are dominated by smoothed passenger-car shapes that fail to exercise surrogates on the thin, complex, highly loaded components governing motorsport performance. This work presents three primary contributions. First, we introduce a high-fidelity RANS dataset built on a parametric LMP2-class CAD model and spanning six operating conditions (map points) covering straight-line and cornering regimes, generated and validated by aerodynamics experts at Dallara to preserve features relevant to industrial motorsport. Second, we present the Gauge-Invariant Spectral Transformer (GIST), a graph-based neural operator whose spectral embeddings encode mesh connectivity to enhance predictions on tightly packed, complex geometries. GIST guarantees discretization invariance and scales linearly with mesh size, achieving state-of-the-art accuracy on both public benchmarks and the proposed race-car dataset. Third, we demonstrate that GIST achieves a level of predictive accuracy suitable for early-stage aerodynamic design, providing a first validation of the concept of interactive design-space exploration -- where engineers query a surrogate in place of the CFD solver -- within industrial motorsport workflows.

Problem

• Computational Fluid Dynamics
• surrogate models
• aerodynamic design
• motorsport
• design space exploration

Innovation

• neural surrogate
• computational fluid dynamics
• graph neural operator
• spectral embedding
• interactive design