Large-scale graph-machine-learning surrogate models for 3D-flowfield prediction in external aerodynamics

Advanced Modeling and Simulation in Engineering Sciences

Table 1 Features of the graph and predicted variables

	Node features		Edge features
	Distances	Output values
Total tensor dimension	\( \|Nodes\| \times 6 \)	\( \|Nodes\| \times 7 \)	\( \|Edges\| \times 1 \)
Description	\(\bullet \|Nodes\| \times 3 \rightarrow \) node	\(\bullet \|Nodes\| \times 1 \rightarrow p\)	\(\bullet \|Edges\| \times 1 \rightarrow \)
	distance from the	\(\bullet \|Nodes\| \times 3 \rightarrow U\)	edge length
	origin (0,0,0)	\(\bullet \|Nodes\| \times 1 \rightarrow k\)
	\(\bullet \|Nodes\| \times 3 \rightarrow \) wall
	distance from the	\(\bullet \|Nodes\| \times 1 \rightarrow \nu _{t}\)
	closest wall face	\(\bullet \|Nodes\| \times 1 \rightarrow \omega \)
	center