Computational modelling of severe gradients with mass operator constructions (Q1104813)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Computational modelling of severe gradients with mass operator constructions |
scientific article |
Statements
Computational modelling of severe gradients with mass operator constructions (English)
0 references
1988
0 references
A generalized mass operator construction is developed from the Galerkin finite element method and used to compute problems with severe internal gradients as simple models for the frontal process. The generalized mass operator construction is compared with four-point upwind differencing of the advective terms. When combined with Crank-Nicolson time differencing both schemes provide optimal choices for generating accurate solutions with severe gradients, primarily by controlling dispersion errors. For very severe gradients the generalized mass operator construction produces more accurate solutions. For severe internal gradients controlled by nonlinear advection it is necessary to supplement the optimal mass operator scheme with a flux-based numerical dissipation incorporating an internal limiter.
0 references
generalized mass operator
0 references
Galerkin finite element method
0 references
four-point upwind differencing
0 references
Crank-Nicolson time differencing
0 references
flux-based numerical dissipation
0 references
internal limiter
0 references
0 references
0 references