Locally refined quad meshing for linear elasticity problems based on convolutional neural networks

TL;DR

This paper introduces a neural network-based method to efficiently generate refined finite element meshes for linear elasticity problems, reducing computational costs and enhancing flexibility across diverse geometries.

Contribution

It presents a convolutional neural network approach, using U-net architecture, to predict mesh refinement regions directly from domain images, bypassing traditional adaptive schemes.

Findings

The neural network accurately predicts stress maxima regions.

The method generalizes well to unseen geometries with different topologies.

It significantly reduces the computational cost of mesh refinement.

Abstract

In this paper we propose a method to generate suitably refined finite element meshes using neural networks. As a model problem we consider a linear elasticity problem on a planar domain (possibly with holes) having a polygonal boundary. We impose boundary conditions by fixing the position of a part of the boundary and applying a force on another part of the boundary. The resulting displacement and distribution of stresses depend on the geometry of the domain and on the boundary conditions. When applying a standard Galerkin discretization using quadrilateral finite elements, one usually has to perform adaptive refinement to properly resolve maxima of the stress distribution. Such an adaptive scheme requires a local error estimator and a corresponding local refinement strategy. The overall costs of such a strategy are high. We propose to reduce the costs of obtaining a suitable discretization by training a neural network whose evaluation replaces this adaptive refinement procedure. We set up a single network for a large class of possible domains and boundary conditions and not on a single domain of interest. The computational domain and boundary conditions are interpreted as images, which are suitable inputs for convolution neural networks. We use the U-net architecture and we devise training strategies by dividing the possible inputs into different categories based on their overall geometric complexity. Thus, we compare different training strategies based on varying geometric complexity. One of the advantages of the proposed approach is the interpretation of input and output as images, which do not depend on the underlying discretization scheme. Another is the generalizability and geometric flexibility. The network can be applied to previously unseen geometries, even with different topology and level of detail. Thus, training can easily be extended to other classes of geometries.