VortexNet: A Graph Neural Network-Based Multi-Fidelity Surrogate Model for Field Predictions
Date:
This talk introduces VortexNet, a graph neural network (GNN)-based surrogate model designed to bridge the fidelity gap between low-fidelity (LF) and high-fidelity (HF) aerodynamic predictions. VortexNet learns corrections to vortex lattice method (VLM) panel-wise loading coefficient fields using data from high-fidelity CFD simulations, enabling pressure coefficient field predictions across a range of delta wing geometries and free-stream conditions. The talk covers the model architecture, its prediction accuracy and generalizability across configurations, and its potential role in conceptual design and multidisciplinary design optimization (MDO).
Related paper: AIAA 2025-0494