Timezone: »

 
Poster
M2N: Mesh Movement Networks for PDE Solvers
Wenbin Song · Mingrui Zhang · Joseph G Wallwork · Junpeng Gao · Zheng Tian · Fanglei Sun · Matthew Piggott · Junqing Chen · Zuoqiang Shi · Xiang Chen · Jun Wang

@

Numerical Partial Differential Equation (PDE) solvers often require discretizing the physical domain by using a mesh. Mesh movement methods provide the capability to improve the accuracy of the numerical solution without introducing extra computational burden to the PDE solver, by increasing mesh resolution where the solution is not well-resolved, whilst reducing unnecessary resolution elsewhere. However, sophisticated mesh movement methods, such as the Monge-Ampère method, generally require the solution of auxiliary equations. These solutions can be extremely expensive to compute when the mesh needs to be adapted frequently. In this paper, we propose to the best of our knowledge the first learning-based end-to-end mesh movement framework for PDE solvers. Key requirements of learning-based mesh movement methods are: alleviating mesh tangling, boundary consistency, and generalization to mesh with different resolutions. To achieve these goals, we introduce the neural spline model and the graph attention network (GAT) into our models respectively. While the Neural-Spline based model provides more flexibility for large mesh deformation, the GAT based model can handle domains with more complicated shapes and is better at performing delicate local deformation. We validate our methods on stationary and time-dependent, linear and non-linear equations, as well as regularly and irregularly shaped domains. Compared to the traditional Monge-Ampère method, our approach can greatly accelerate the mesh adaptation process by three to four orders of magnitude, whilst achieving comparable numerical error reduction.

Keywords: Deep Learning · Neural Network · Partial Differential Equation · Moving Mesh · Graph Attention Network · r-Adaptation · Mesh Adaptation · Mesh Movement · Monge–Ampere · Neural Spline
[ Paper]  [ Poster]  [ OpenReview] 

Author Information

Wenbin Song (Shanghaitech University)
Mingrui Zhang (Imperial College London)
Joseph G Wallwork (Imperial College London)
Junpeng Gao (ETHZ - ETH Zurich)
Zheng Tian (UCL)
Fanglei Sun (ShanghaiTech)
Matthew Piggott (Imperial College London)
Junqing Chen (Tsinghua University, Tsinghua University)
Zuoqiang Shi (zqshi@mail.tsinghua.edu.cn)
Xiang Chen (Huawei Technologies Ltd.)
Jun Wang (UCL)

More from the Same Authors