REVIEW 1 cited by
Direct Prediction of Steady-State Flow Fields in Meshed Domain with Graph Networks
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Direct Prediction of Steady-State Flow Fields in Meshed Domain with Graph Networks
read the original abstract
We propose a model to directly predict the steady-state flow field for a given geometry setup. The setup is an Eulerian representation of the fluid flow as a meshed domain. We introduce a graph network architecture to process the mesh-space simulation as a graph. The benefit of our model is a strong understanding of the global physical system, while being able to explore the local structure. This is essential to perform direct prediction and is thus superior to other existing methods.
Forward citations
Cited by 1 Pith paper
-
Spatio-Temporal Prediction of Unsteady Airfoil Aerodynamics Using Augmented Graph Neural Ordinary Differential Equations with Exogenous Controls
GNODE, a graph-neural-network-plus-neural-ODE surrogate with augmented latent dimensions, predicts unsteady transonic airfoil flows with more stable and accurate rollouts than an autoregressive GNN baseline.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.