Recommendations on Enhancing the Efficiency of Algebraic Multigrid Preconditioned GMRES in Solving Coupled Fluid Flow Equations

Abstract

The algebraic multigrid (AMG) algorithm as a preconditioner to the Krylov subspace methods has drawn the attention of many researchers in solving fluid flow and heat transfer problems. However, the efficient employment of this solver needs experience, because users have to quantify several important parameters. In this work, we choose a hybrid finite-volume element method and quantify the optimum magnitudes for those parameters. To generalize our results, two sets of fluid flow governing equations, the thermobuoyant flow and confined diffusion flame, are studied and the optimum values are determined. The results indicate that the AMG can be very effective if a proper storage method is chosen.

Notes

F, Falgout; RS, Ruge-Stüben; MRS, Modified Ruge-Stüben.

F, Falgout; RS, Ruge-Stüben; MRS, Modified Ruge-Stüben.