AMG preconditioning for nonlinear degenerate parabolic equations on nonuniform grids with application to monument degradation (Q1946153)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | AMG preconditioning for nonlinear degenerate parabolic equations on nonuniform grids with application to monument degradation |
scientific article |
Statements
AMG preconditioning for nonlinear degenerate parabolic equations on nonuniform grids with application to monument degradation (English)
0 references
18 April 2013
0 references
A numerical method for a nonlinear parabolic diffusion equation is treated. The nonlinearity of the diffusion coefficient is a nonnegative function that can degenerate whenever the coefficient vanishes. The model is motivated by the modelling of marble degradation by chemical pollutants. The time discretization is done using the so-called \(\theta\) method and the space discretization is based on nonuniform finite differences for one and two space domains. The resulting algebraic systems are nonlinear and they are treated using Newton's methods, and the locally Toeplitz linear systems from the Jacobian are solved efficiently using AMG (algebraic multigrid) preconditioners. The analysis of the spectrum of the preconditioned matrix that proves the effectiveness of the method is studied. Then, the discretization of the model for the chemical aggression on marble stones consisting of a system of two partial differential equations is proposed. Finally, numerical experiments conclude the paper. The first one is the well-known Barneblatt solution on which the confirmation of the theoretical results for the proposed method are presented, and the second one investigates the marble sulfation model.
0 references
nonlinear and degenerate parabolic equations
0 references
finite differences
0 references
Newton-Krylov method
0 references
multigrid preconditioning
0 references
flow in porous media
0 references
marble sulfation
0 references
theta method
0 references
numerical experiments
0 references