• Corpus ID: 233024982

Semi matrix-free twogrid shifted Laplacian preconditioner for the Helmholtz equation with near optimal shifts

  title={Semi matrix-free twogrid shifted Laplacian preconditioner for the Helmholtz equation with near optimal shifts},
  author={Daniel Drzisga and Tobias K{\"o}ppl and Barbara I. Wohlmuth},
Due to its significance in terms of wave phenomena a considerable effort has been put into the design of preconditioners for the Helmholtz equation. One option to derive a preconditioner is to apply a multigrid method on a shifted operator. In such an approach, the wavenumber is shifted by some imaginary value. This step is motivated by the observation that the shifted problem can be more efficiently handled by iterative solvers when compared to the standard Helmholtz equation. However, up to… 

Local Fourier analysis of the complex shifted Laplacian preconditioner for Helmholtz problems

A wavenumber‐dependent minimal complex shift parameter is proposed, which is predicted by a rigorous k‐grid local Fourier analysis of the multigrid scheme and claimed to provide the reader with a parameter choice that leads to efficient Krylov convergence.


  • Elsevier
  • 2000

MFEM: A modular finite element library

  • Computers & Mathematics with Applications, 81:42–74
  • 2021

Closed Form Dispersion Corrections Including a Real Shifted WaveNumber for Finite Difference Discretizations of 2D Constant Coefficient Helmholtz Problems

All grid-based discretizations of the Helmholtz equation suffer from the so-called pollution effect, which is caused by numerical dispersion: plane waves propagate at the discrete level at speeds w...

A two-level shifted Laplace preconditioner for Helmholtz problems: Field-of-values analysis and wavenumber-independent convergence

This work revisits the idea of combining the shifted Laplace preconditioner with two-level deflation and applies it to Helmholtz problems discretized with linear finite elements and presents numerical evidence that wavenumber-independent convergence of GMRES also holds for pollution-free meshes, where the coarse mesh size satisfies Hk^{3/2} \leq C $, and inexact coarse grid solves.

Learning Algebraic Multigrid Using Graph Neural Networks

A framework for learning AMG prolongation operators for linear systems with sparse symmetric positive (semi-) definite matrices is proposed and a single graph neural network is trained to learn a mapping from an entire class of such matrices to prolongation Operators, using an efficient unsupervised loss function.

PyTorch: An Imperative Style, High-Performance Deep Learning Library

This paper details the principles that drove the implementation of PyTorch and how they are reflected in its architecture, and explains how the careful and pragmatic implementation of the key components of its runtime enables them to work together to achieve compelling performance.

Stencil Scaling for Vector-Valued PDEs on Hybrid Grids With Applications to Generalized Newtonian Fluids

Matrix-free finite element implementations for large applications provide an attractive alternative to standard sparse matrix data formats due to the significantly reduced memory consumption. Here,


I” ln”“ntainO”S meas, such ai tile foothills of tile Canadian Rwkira. careful processing of near~surface seismic dam is essential. for withour an accurate interpretation of the near surface. a