On multigrid methods for the compressible navier-stokes equations

Dimitris Drikakis, Oleg Iliev, Daniela Vassileva

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


An investigation of the efficiency of multigrid algorithms for the compressible Navier-Stokes equations is presented. The computational code that forms the basis for this investigation utilises a hybrid Godunov-type method and central differences for discretising the inviscid and viscous fluxes, respectively, as well as implicit-unfactored and explicit solvers. To accelerate the numerical convergence towards a steady state solution we have employed a non-linear multigrid method. Further, we have also implemented a dynamically adaptive multigrid algorithm in conjunction with the explicit solver. Computations have been conducted for low Reynolds number compressible flows around an aerofoil bothat subsonic and supersonic flow conditions. Results from several numerical experiments are presented in order to examine the performance of the multigrid algorithms in conjunction with the explicit and implicit solvers.

Original languageEnglish
Title of host publicationLarge-Scale Scientific Computing - 3rd International Conference, LSSC 2001, Revised Papers
EditorsSvetozar Margenov, Jerzy Wasniewski, Plamen Yalamov
PublisherSpringer Verlag
Number of pages9
ISBN (Print)3540430431
Publication statusPublished - 2001
Externally publishedYes
Event3rd International Conference on Large-Scale Scientific Computing, LSSC 2001 - Sozopol, Bulgaria
Duration: 6 Jun 200110 Jun 2001

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference3rd International Conference on Large-Scale Scientific Computing, LSSC 2001


  • Adaptivity
  • Compressible flows
  • Navier-Stokes equations
  • Nonlinear multigrid


Dive into the research topics of 'On multigrid methods for the compressible navier-stokes equations'. Together they form a unique fingerprint.

Cite this