A Study on the Stability Limits of Graph Neural Network Surrogates for Advection-Diffusion

Master Thesis (2026)
Author(s)

M. Maassen van den Brink (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Contributor(s)

A. Heinlein – Mentor (TU Delft - Electrical Engineering, Mathematics and Computer Science)

R.P. Dwight – Mentor (TU Delft - Aerospace Engineering)

M. Verlaan – Graduation committee member (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Faculty
Electrical Engineering, Mathematics and Computer Science
More Info
expand_more
Publication Year
2026
Language
English
Graduation Date
11-06-2026
Awarding Institution
Delft University of Technology
Programme
Applied Mathematics
Faculty
Electrical Engineering, Mathematics and Computer Science
Downloads counter
16
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Machine-learned surrogate time integrators promise large speed-ups over classical solvers, yet their performance is usually reported as a single aggregate error, leaving open the question of when they remain stable. This thesis determines the empirical stability limits of graph neural network (GNN) surrogates for the two-dimensional advection–diffusion equation on unstructured meshes with periodic boundary conditions, expressed directly in the dimensionless CFL and Fourier numbers.

Surrogates are trained to minimise the one-step error on fixed velocity and diffusion fields and evaluated autoregressively on unseen fields over horizons eight times the training window. Two families are compared at a fixed message passing budget: single-scale models, and multiscale models organised as V-cycles over predetermined coarsened graphs. For each model, a piecewise-linear fit of the final rollout error against the CFL and Fourier numbers yields empirical stability limits, defined by a blow-up threshold.

Within these limits the surrogates reproduce the finite element reference accurately on both seen and unseen fields and show no abrupt change beyond the training horizon, although the diffusion-dominated regime is consistently harder than the advection-dominated one. The single-scale CFL limit tracks the number of message-passing blocks and lies slightly above it. Adding coarse levels at a fixed total message passing layer budget broadens the advective stability range, decisively at the largest stride, but a two-level hierarchy trades diffusive stability and in-region accuracy for this gain. Only a three-level V-cycle removes the penalty, attaining zero blow-ups on both axes, and deeper models show no oversmoothing.

The diffusion-side limits carry large variance, traced partly to the dissipative backward-Euler reference, and should be read as indicative. The work delivers a concrete operational range for GNN surrogates and identifies how multiscale models can extend it.

Files

MEP_MMVDB.pdf
(pdf | 33.4 Mb)
License info not available