Towards Universal Mesh Movement Networks

TL;DR

The paper introduces UM2N, a universal, pre-trained mesh movement network that efficiently adapts to various PDEs and geometries without re-training, outperforming existing methods in speed and robustness.

Contribution

We propose UM2N, a novel universal mesh movement network that operates in a zero-shot manner across different PDEs and geometries, addressing limitations of prior learning-based approaches.

Findings

Outperforms existing learning-based mesh movement methods.

Significantly accelerates mesh movement compared to traditional PDE solvers.

Effective in complex real-world scenarios like tsunami simulation.

Abstract

Solving complex Partial Differential Equations (PDEs) accurately and efficiently is an essential and challenging problem in all scientific and engineering disciplines. Mesh movement methods provide the capability to improve the accuracy of the numerical solution without increasing the overall mesh degree of freedom count. Conventional sophisticated mesh movement methods are extremely expensive and struggle to handle scenarios with complex boundary geometries. However, existing learning-based methods require re-training from scratch given a different PDE type or boundary geometry, which limits their applicability, and also often suffer from robustness issues in the form of inverted elements. In this paper, we introduce the Universal Mesh Movement Network (UM2N), which -- once trained -- can be applied in a non-intrusive, zero-shot manner to move meshes with different size distributions and structures, for solvers applicable to different PDE types and boundary geometries. UM2N consists of a Graph Transformer (GT) encoder for extracting features and a Graph Attention Network (GAT) based decoder for moving the mesh. We evaluate our method on advection and Navier-Stokes based examples, as well as a real-world tsunami simulation case. Our method outperforms existing learning-based mesh movement methods in terms of the benchmarks described above. In comparison to the conventional sophisticated Monge-Amp\`ere PDE-solver based method, our approach not only significantly accelerates mesh movement, but also proves effective in scenarios where the conventional method fails. Our project page is at https://erizmr.github.io/UM2N/.