Automatic stabilization of finite-element simulations using neural networks and hierarchical matrices
From MaRDI portal
Publication:6097614
DOI10.1016/J.CMA.2023.116073arXiv2212.12695OpenAlexW4367856946MaRDI QIDQ6097614FDOQ6097614
Authors: Tomasz Służalec, Mateusz Dobija, Anna Paszyńska, Ignacio Muga, Marcin M. Łoś, Maciej Paszynski 
Publication date: 6 June 2023
Published in: Computer Methods in Applied Mechanics and Engineering (Search for Journal in Brave)
Abstract: Petrov-Galerkin formulations with optimal test functions allow for the stabilization of finite element simulations. In particular, given a discrete trial space, the optimal test space induces a numerical scheme delivering the best approximation in terms of a problem-dependent energy norm. This ideal approach has two shortcomings: first, we need to explicitly know the set of optimal test functions; and second, the optimal test functions may have large supports inducing expensive dense linear systems. Nevertheless, parametric families of PDEs are an example where it is worth investing some (offline) computational effort to obtain stabilized linear systems that can be solved efficiently, for a given set of parameters, in an online stage. Therefore, as a remedy for the first shortcoming, we explicitly compute (offline) a function mapping any PDE-parameter, to the matrix of coefficients of optimal test functions (in a basis expansion) associated with that PDE-parameter. Next, as a remedy for the second shortcoming, we use the low-rank approximation to hierarchically compress the (non-square) matrix of coefficients of optimal test functions. In order to accelerate this process, we train a neural network to learn a critical bottleneck of the compression algorithm (for a given set of PDE-parameters). When solving online the resulting (compressed) Petrov-Galerkin formulation, we employ a GMRES iterative solver with inexpensive matrix-vector multiplications thanks to the low-rank features of the compressed matrix. We perform experiments showing that the full online procedure as fast as the original (unstable) Galerkin approach. In other words, we get the stabilization with hierarchical matrices and neural networks practically for free. We illustrate our findings by means of 2D Eriksson-Johnson and Hemholtz model problems.
Full work available at URL: https://arxiv.org/abs/2212.12695
Recommendations
- A class of discontinuous Petrov-Galerkin methods. II. Optimal test functions
- Neural control of discrete weak formulations: Galerkin, least squares \& minimal-residual methods with quasi-optimal weights
- Finite element stabilization parameters computed from element matrices and vectors
- A finite element based deep learning solver for parametric PDEs
- A discontinuous Petrov-Galerkin methodology for adaptive solutions to the incompressible Navier-Stokes equations
neural networksparametric PDEshierarchical matricesPetrov-Galerkin methodoptimal test functionsautomatic stabilization
Cites Work
- Multifrontal parallel distributed symmetric and unsymmetric solvers
- A fully asynchronous multifrontal solver using distributed dynamic scheduling
- A sparse matrix arithmetic based on \({\mathfrak H}\)-matrices. I: Introduction to \({\mathfrak H}\)-matrices
- Parallel iterative methods for sparse linear systems
- A dual Petrov-Galerkin finite element method for the convection-diffusion equation
- Hierarchical matrices: algorithms and analysis
- A class of discontinuous Petrov-Galerkin methods. II. Optimal test functions
- Why it is difficult to solve Helmholtz problems with classical iterative methods
- Wavenumber explicit analysis of a DPG method for the multidimensional Helmholtz equation
- An Overview of the Discontinuous Petrov Galerkin Method
- Robust DPG method for convection-dominated diffusion problems
- A new finite element formulation for computational fluid dynamics. VI. Convergence analysis of the generalized SUPG formulation for linear time- dependent multidimensional advective-diffusive systems
- The value of continuity: refined isogeometric analysis and fast direct solvers
- A direct elliptic solver based on hierarchically low-rank Schur complements
- Isogeometric residual minimization method (iGRM) with direction splitting for non-stationary advection-diffusion problems
- Hierarchical preconditioners for high-order FEM
- Refined isogeometric analysis for a preconditioned conjugate gradient solver
- Isogeometric residual minimization method (iGRM) with direction splitting preconditioner for stationary advection-dominated diffusion problems
- The surrogate matrix methodology: accelerating isogeometric analysis of waves
- Towards accuracy and scalability: combining isogeometric analysis with deflation to obtain scalable convergence for the Helmholtz equation
- The surrogate matrix methodology: low-cost assembly for isogeometric analysis
- Iterative solution of Helmholtz problem with high-order isogeometric analysis and finite element method at mid-range frequencies
- A two-scale approach for efficient on-the-fly operator assembly in massively parallel high performance multigrid codes
- The Surrogate Matrix Methodology: A Priori Error Estimation
- Isogeometric residual minimization (iGRM) for non-stationary Stokes and Navier-Stokes problems
- Discretization of Linear Problems in Banach Spaces: Residual Minimization, Nonlinear Petrov--Galerkin, and Monotone Mixed Methods
- Minimal residual space-time discretizations of parabolic equations: asymmetric spatial operators
Cited In (2)
This page was built for publication: Automatic stabilization of finite-element simulations using neural networks and hierarchical matrices
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6097614)
