A Deep Learning algorithm to accelerate Algebraic Multigrid methods in Finite Element solvers of 3D elliptic PDEs

Abstract

Algebraic multigrid (AMG) methods are among the most efficient solvers for linear systems of equations and they are widely used for the solution of problems stemming from the discretization of Partial Differential Equations (PDEs). The most severe limitation of AMG methods is the dependence on parameters that require to be fine-tuned. In particular, the strong threshold parameter is the most relevant since it stands at the basis of the construction of successively coarser grids needed by the AMG methods. We introduce a novel Deep Learning algorithm that minimizes the computational cost of the AMG method when used as a finite element solver. We show that our algorithm requires minimal changes to any existing code. The proposed Artificial Neural Network (ANN) tunes the value of the strong threshold parameter by interpreting the sparse matrix of the linear system as a black-and-white image and exploiting a pooling operator to transform it into a small multi-channel image. We experimentally prove that the pooling successfully reduces the computational cost of processing a large sparse matrix and preserves the features needed for the regression task at hand. We train the proposed algorithm on a large dataset containing problems with a highly heterogeneous diffusion coefficient defined in different three-dimensional geometries and discretized with unstructured grids and linear elasticity problems with a highly heterogeneous Young's modulus. When tested on problems with coefficients or geometries not present in the training dataset, our approach reduces the computational time by up to 30%.

References in corpus (10)

• Enhancing Computational Fluid Dynamics with Machine Learning
• Prediction of Aerodynamic Flow Fields Using Convolutional Neural Networks
• The deal.II finite element library: design, features, and insights
• A Two-Stage Preconditioner for Multiphase Poromechanics in Reservoir Simulation
• A machine learning method for real-time numerical simulations of cardiac electromechanics
• Refinement of polygonal grids using Convolutional Neural Networks with applications to polygonal Discontinuous Galerkin and Virtual Element methods
• A data-driven approach for the closure of RANS models by the divergence of the Reynolds Stress Tensor
• Machine Learning based refinement strategies for polyhedral grids with applications to Virtual Element and polyhedral Discontinuous Galerkin methods
• Accelerating Algebraic Multigrid Methods via Artificial Neural Networks
• Multigrid solvers for multipoint flux approximations of the Darcy problem on rough quadrilateral grids

Cited by in corpus (3)

• Combining physics-based and data-driven models: advancing the frontiers of research with Scientific Machine Learning
• Discovering Artificial Viscosity Models for Discontinuous Galerkin Approximation of Conservation Laws using Physics-Informed Machine Learning
• R3MG: R-tree based agglomeration of polytopal grids with applications to multilevel methods