Agglomeration of Polygonal Grids using Graph Neural Networks with applications to Multigrid solvers

by   P. F. Antonietti, et al.

Agglomeration-based strategies are important both within adaptive refinement algorithms and to construct scalable multilevel algebraic solvers. In order to automatically perform agglomeration of polygonal grids, we propose the use of Graph Neural Networks (GNNs) to partition the connectivity graph of a computational mesh. GNNs have the advantage to process naturally and simultaneously both the graph structure of mesh and the geometrical information, such as the areas of the elements or their barycentric coordinates. This is not the case with other approaches such as METIS, a standard algorithm for graph partitioning which is meant to process only the graph information, or the k-means clustering algorithm, which can process only the geometrical information. Performance in terms of quality metrics is enhanced for Machine Learning (ML) strategies, with GNNs featuring a lower computational cost online. Such models also show a good degree of generalization when applied to more complex geometries, such as brain MRI scans, and the capability of preserving the quality of the grid. The effectiveness of these strategies is demonstrated also when applied to MultiGrid (MG) solvers in a Polygonal Discontinuous Galerkin (PolyDG) framework.


page 1

page 2

page 3

page 4


Power to the Relational Inductive Bias: Graph Neural Networks in Electrical Power Grids

The application of graph neural networks (GNNs) to the domain of electri...

Bi-Stride Multi-Scale Graph Neural Network for Mesh-Based Physical Simulation

Learning physical systems on unstructured meshes by flat Graph neural ne...

Graph-Based Deep Learning for Sea Surface Temperature Forecasts

Sea surface temperature (SST) forecasts help with managing the marine ec...

Performance Analysis of FEM Solvers on Practical Electromagnetic Problems

The paper presents a comparative analysis of different commercial and ac...

Please sign up or login with your details

Forgot password? Click here to reset