Mesh-based simulations play a key role when modeling complex physical systems that, in many disciplines across science and engineering, require the solution to parametrized time-dependent nonlinear partial differential equations (PDEs). In this context, full order models (FOMs), such as those relying on the finite element method, can reach high levels of accuracy, however often yielding intensive simulations to run. For this reason, surrogate models are developed to replace computationally expensive solvers with more efficient ones, which can strike favorable trade-offs between accuracy and efficiency. This work explores the potential usage of graph neural networks (GNNs) for the simulation of time-dependent PDEs in the presence of geometrical variability. In particular, we propose a systematic strategy to build surrogate models based on a data-driven time-stepping scheme where a GNN architecture is used to efficiently evolve the system. With respect to the majority of surrogate models, the proposed approach stands out for its ability of tackling problems with parameter-dependent spatial domains, while simultaneously generalizing to different geometries and mesh resolutions. We assess the effectiveness of the proposed approach through a series of numerical experiments, involving both two- and three-dimensional problems, showing that GNNs can provide a valid alternative to traditional surrogate models in terms of computational efficiency and generalization to new scenarios.

1
F.
Negri
,
G.
Rozza
,
A.
Manzoni
, and
A.
Quarteroni
, “
Reduced basis method for parametrized elliptic optimal control problems
,”
SIAM J. Sci. Comput.
35
,
A2316
A2340
(
2013
).
2
A.
Quarteroni
,
A.
Manzoni
, and
F.
Negri
,
Reduced Basis Methods for Partial Differential Equations: An Introduction
(
Springer
,
2015
), Vol. 92.
3
J. S.
Hesthaven
,
G.
Rozza
, and
B.
Stamm
,
Certified Reduced Basis Methods for Parametrized Partial Differential Equations
(
Springer
,
2016
), Vol. 590.
4
J. S.
Hesthaven
and
S.
Ubbiali
, “
Non-intrusive reduced order modeling of nonlinear problems using neural networks
,”
J. Comput. Phys.
363
,
55
78
(
2018
).
5
M.
Guo
and
J. S.
Hesthaven
, “
Data-driven reduced order modeling for time-dependent problems
,”
Comput. Methods Appl. Mech. Eng.
345
,
75
99
(
2019
).
6
M.
Guo
,
S. A.
McQuarrie
, and
K. E.
Willcox
, “
Bayesian operator inference for data-driven reduced-order modeling
,”
Comput. Methods Appl. Mech. Eng.
402
,
115336
(
2022
).
7
K.
Carlberg
, “
Adaptive h-refinement for reduced-order models
,”
Int. J. Numer. Methods Eng.
102
,
1192
1210
(
2015
).
8
J. S.
Hesthaven
,
C.
Pagliantini
, and
N.
Ripamonti
, “
Rank-adaptive structure-preserving model order reduction of Hamiltonian systems
,”
ESAIM: Math. Modell. Numer. Anal.
56
,
617
650
(
2022
).
9
M.
Hess
,
A.
Alla
,
A.
Quaini
,
G.
Rozza
, and
M.
Gunzburger
, “
A localized reduced-order modeling approach for PDEs with bifurcating solutions
,”
Comput. Methods Appl. Mech. Eng.
351
,
379
403
(
2019
).
10
D.
Amsallem
and
C.
Farhat
, “
An online method for interpolating linear parametric reduced-order models
,”
SIAM J. Sci. Comput.
33
,
2169
2198
(
2011
).
11
C.
Pagliantini
, “
Dynamical reduced basis methods for Hamiltonian systems
,”
Numer. Math.
148
,
409
448
(
2021
).
12
G.
Buza
,
S.
Jain
, and
G.
Haller
, “
Using spectral submanifolds for optimal mode selection in nonlinear model reduction
,”
Proc. R. Soc. A
477
,
20200725
(
2021
).
13
M.
Li
,
S.
Jain
, and
G.
Haller
, “
Model reduction for constrained mechanical systems via spectral submanifolds
,”
Nonlinear Dyn.
111
,
8881
8911
(
2023
).
14
A.
Bonito
,
A.
Cohen
,
R.
DeVore
,
D.
Guignard
,
P.
Jantsch
, and
G.
Petrova
, “
Nonlinear methods for model reduction
,”
ESAIM: Math. Modell. Numer. Anal.
55
,
507
531
(
2021
).
15
S.
Fresca
,
L.
Dede’
, and
A.
Manzoni
, “
A comprehensive deep learning-based approach to reduced order modeling of nonlinear time-dependent parametrized PDEs
,”
J. Sci. Comput.
87
,
1
36
(
2021
).
16
N.
Franco
,
A.
Manzoni
, and
P.
Zunino
, “
A deep learning approach to reduced order modelling of parameter dependent partial differential equations
,”
Math. Comput.
92
,
483
524
(
2023
).
17
S.
Fresca
and
A.
Manzoni
, “
POD-DL-ROM: Enhancing deep learning-based reduced order models for nonlinear parametrized PDEs by proper orthogonal decomposition
,”
Comput. Methods Appl. Mech. Eng.
388
,
114181
(
2022
).
18
S.
Brivio
,
S.
Fresca
,
N. R.
Franco
, and
A.
Manzoni
, “Error estimates for POD-DL-ROMs: A deep learning framework for reduced order modeling of nonlinear parametrized PDEs enhanced by proper orthogonal decomposition,” arXiv:2305.04680 (2023).
19
L.
Cicci
,
S.
Fresca
, and
A.
Manzoni
, “
Deep-HyROMnet: A deep learning-based operator approximation for hyper-reduction of nonlinear parametrized PDEs
,”
J. Sci. Comput.
93
,
57
(
2022
).
20
F.
Fatone
,
S.
Fresca
, and
A.
Manzoni
, “Long-time prediction of nonlinear parametrized dynamical systems by deep learning-based reduced order models,” arXiv:2201.10215 (2022).
21
F.
Romor
,
G.
Stabile
, and
G.
Rozza
, “
Non-linear manifold reduced-order models with convolutional autoencoders and reduced over-collocation method
,”
J. Sci. Comput.
94
,
74
(
2023
).
22
J.
Shlomi
,
P.
Battaglia
, and
J.-R.
Vlimant
, “
Graph neural networks in particle physics
,”
Mach. Learn.: Sci. Technol.
2
,
021001
(
2020
).
23
K.
Shukla
,
M.
Xu
,
N.
Trask
, and
G. E.
Karniadakis
, “
Scalable algorithms for physics-informed neural and graph networks
,”
Data-Centric Eng.
3
,
e24
(
2022
).
24
M.
Horie
and
N.
Mitsume
, “
Physics-embedded neural networks: Graph neural PDE solvers with mixed boundary conditions
,”
Adv. Neural Inf. Process. Syst.
35
,
23218
23229
(
2022
).
25
L.
Lu
,
P.
Jin
,
G.
Pang
,
Z.
Zhang
, and
G. E.
Karniadakis
, “
Learning nonlinear operators via deeponet based on the universal approximation theorem of operators
,”
Nat. Mach. Intell.
3
,
218
229
(
2021
).
26
Z.
Li
,
N.
Kovachki
,
K.
Azizzadenesheli
,
B.
Liu
,
K.
Bhattacharya
,
A.
Stuart
, and
A.
Anandkumar
, “Fourier neural operator for parametric partial differential equations,” arXiv:2010.08895 (2020).
27
S.
Pan
,
S. L.
Brunton
, and
J. N.
Kutz
, “Neural implicit flow: A mesh-agnostic dimensionality reduction paradigm of spatio-temporal data,” arXiv:2204.03216 (2022).
28
F.
Scarselli
,
M.
Gori
,
A. C.
Tsoi
,
M.
Hagenbuchner
, and
G.
Monfardini
, “
The graph neural network model
,”
IEEE Trans. Neural Netw.
20
,
61
80
(
2008
).
29
P. W.
Battaglia
,
J. B.
Hamrick
,
V.
Bapst
,
A.
Sanchez-Gonzalez
,
V.
Zambaldi
,
M.
Malinowski
,
A.
Tacchetti
,
D.
Raposo
,
A.
Santoro
,
R.
Faulkner
, and
C.
Gulcehre
, “Relational inductive biases, deep learning, and graph networks,” arXiv:1806.01261 (2018).
30
T.
Pfaff
,
M.
Fortunato
,
A.
Sanchez-Gonzalez
, and
P. W.
Battaglia
, “Learning mesh-based simulation with graph networks,” arXiv:2010.03409 (2020).
31
L.
Pegolotti
,
M. R.
Pfaller
,
N. L.
Rubio
,
K.
Ding
,
R. B.
Brufau
,
E.
Darve
, and
A. L.
Marsden
, “Learning reduced-order models for cardiovascular simulations with graph neural networks,” arXiv:2303.07310 (2023).
32
R. J.
Gladstone
,
H.
Rahmani
,
V.
Suryakumar
,
H.
Meidani
,
M.
D’Elia
, and
A.
Zareei
, “GNN-based physics solver for time-independent PDEs,” arXiv:2303.15681 (2023).
33
F.
Pichi
,
B.
Moya
, and
J. S.
Hesthaven
, “A graph convolutional autoencoder approach to model order reduction for parametrized PDEs,” arXiv:2305.08573 (2023).
34
A.
Quarteroni
and
A.
Valli
,
Numerical Approximation of Partial Differential Equations
(
Springer
,
1994
), Vol. 23.
35
W. L.
Hamilton
,
R.
Ying
, and
J.
Leskovec
, “Inductive representation learning on large graphs,” Advances in Neural Information Processing Systems (Curran Associates, Inc., 2017), p. 1024.
36
A. P.
Toshev
,
L.
Paehler
,
A.
Panizza
, and
N. A.
Adams
, “On the relationships between graph neural networks for the simulation of physical systems and classical numerical methods,” arXiv:2304.00146 (2023).
37
M.
Niepert
,
M.
Ahmed
, and
K.
Kutzkov
, “Learning convolutional neural networks for graphs,” in Proceedings of the 33rd International Conference on Machine Learning, Proceedings of Machine Learning Research, edited by M. F. Balcan and K. Q. Weinberger (PMLR, New York, 2016), Vol. 48, pp. 2014–2023.
38
S.
Zhang
,
H.
Tong
,
J.
Xu
, and
R.
Maciejewski
, “
Graph convolutional networks: A comprehensive review
,”
Comput. Soc. Netw.
6
,
1
23
(
2019
).
39
T.
Kipf
and
M.
Welling
, “Semi-supervised classification with graph convolutional networks,” in 5th International Conference on Learning Representations (ICLR-17) (OpenReview, 2016).
40
P.
Veličković
,
G.
Cucurull
,
A.
Casanova
,
A.
Romero
,
P.
Liò
, and
Y.
Bengio
, “Graph attention networks,” in 6th International Conference on Learning Representation (OpenReview, 2017).
41
Y.
Shi
,
Z.
Huang
,
S.
Feng
,
H.
Zhong
,
W.
Wang
, and
Y.
Sun
, “Masked label prediction: Unified message passing model for semi-supervised classification,” in Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21, edited by Z.-H. Zhou (International Joint Conferences on Artificial Intelligence Organization, 2021), pp. 1548–1554, main Track.
42
P.
Battaglia
,
R.
Pascanu
,
M.
Lai
,
D. J.
Rezende
, and
K.
kavukcuoglu
, “Interaction networks for learning about objects, relations and physics,” in Proceedings of the 30th International Conference on Neural Information Processing Systems, NIPS’16 (Curran Associates Inc., Red Hook, NY, 2016), pp. 4509–4517.
43
F.
Murtagh
, “
Multilayer perceptrons for classification and regression
,”
Neurocomputing
2
,
183
197
(
1991
).
44
A.
Sanchez-Gonzalez
,
J.
Godwin
,
T.
Pfaff
,
R.
Ying
,
J.
Leskovec
, and
P.
Battaglia
, “Learning to simulate complex physics with graph networks,” in International Conference on Machine Learning (PMLR, 2020), pp. 8459–8468.
45
T. K.
Rusch
,
M. M.
Bronstein
, and
S.
Mishra
, “A survey on oversmoothing in graph neural networks,” arXiv:2303.10993 (2023).
46
S.
Barwey
,
V.
Shankar
, and
R.
Maulik
, “Multiscale graph neural network autoencoders for interpretable scientific machine learning,” arXiv:2302.06186 (2023).
47
N. R.
Franco
,
A.
Manzoni
, and
P.
Zunino
, “Learning operators with mesh-informed neural networks,” arXiv:2203.11648 (2022).
48
D.
Rumelhart
,
G.
Hinton
, and
R.
Williams
, “
Learning representations by back-propagating errors
,”
Nature
323
,
533
536
(
1986
).
49
D. P.
Kingma
and
J.
Ba
,
Adam: A Method for Stochastic Optimization
(
International Conference on Learning Representations (ICLR)
,
2015
).
50
Q.
Hernández
,
A.
Badías
,
F.
Chinesta
, and
E.
Cueto
, “Thermodynamics-informed graph neural networks,” arXiv:2203.01874 (2022).
51
N. T.
Mücke
,
S. M.
Bohté
, and
C. W.
Oosterlee
, “
Reduced order modeling for parameterized time-dependent PDEs using spatially and memory aware deep learning
,”
J. Comput. Sci.
53
,
101408
(
2021
).
52
S. T.
Kelly
and
B. I.
Epureanu
, “Data-driven reduced-order model for turbomachinery blisks with friction nonlinearity,” in Nonlinear Structures & Systems, Volume 1: Proceedings of the 40th IMAC, A Conference and Exposition on Structural Dynamics 2022 (Springer, 2022), pp. 97–100.
53
Q.
Wang
,
N.
Ripamonti
, and
J. S.
Hesthaven
, “
Recurrent neural network closure of parametric POD-Galerkin reduced-order models based on the Mori-Zwanzig formalism
,”
J. Comput. Phys.
410
,
109402
(
2020
).
54
Q.
Cao
,
S.
Goswami
, and
G. E.
Karniadakis
, “LNO: Laplace neural operator for solving differential equations,” arXiv:2303.10528 (2023).
55
L.
Yan
and
T.
Zhou
, “An adaptive surrogate modeling based on deep neural networks for large-scale Bayesian inverse problems,” arXiv:1911.08926 (2019).
56
J.
Gilmer
,
S. S.
Schoenholz
,
P. F.
Riley
,
O.
Vinyals
, and
G. E.
Dahl
, “Neural message passing for quantum chemistry,” in International Conference on Machine Learning (PMLR, 2017), pp. 1263–1272.
57
Y.
Liu
,
C.
Ponce
,
S. L.
Brunton
, and
J. N.
Kutz
, “
Multiresolution convolutional autoencoders
,”
J. Comput. Phys.
474
,
111801
(
2023
).
58
S.
Hijazi
,
G.
Stabile
,
A.
Mola
, and
G.
Rozza
, “
Data-driven POD-Galerkin reduced order model for turbulent flows
,”
J. Comput. Phys.
416
,
109513
(
2020
).
You do not currently have access to this content.