CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Elliptic meshing generation (http://www.cfd-online.com/Forums/main/4074-elliptic-meshing-generation.html)

Meggie November 3, 2001 04:40

Elliptic meshing generation
 
please recommend some simple code. I am entry-learner.

Axel Rohde November 6, 2001 00:27

Re: Elliptic meshing generation
 
I wrote an elliptic mesh generator many years ago, based on four different numerical methods. Unfortunately, I no longer have the report, otherwise I would send it to you (the source code would not help you; I don't even understand it anymore).

Elliptic grid generators are based on Laplace's or Poisson's equation (P = L + source term). In essence, you are solving a (2-D) heat conduction problem, and your converged mesh consists of lines of constant temperature (isotherms) intersected normally by flux lines. There are several ways to solve the Laplace or Poisson equation numerically. In increasing complexity, these are,

SG-SOR

SG-ADI

MG-SOR

MG-ADI

where,

SG = Single-Grid

MG = Multi-Grid

SOR = Successive Over-Relaxation

ADI = Alternating Direction Implicit

You should be able to find references to these methods in the literature. A good starting point for elliptic mesh generation is Chapter 5, "Computational Fluid Dynamics - The Basics with Applications", by John D. Anderson, Jr.


S.P.Asok November 7, 2001 03:08

Re: Elliptic meshing generation
 
Mr.Axel,is it that the generated mesh can be used for all kind of discretisations.I am a finite element follower.Regards,asoksp

Axel Rohde November 8, 2001 08:48

Re: Elliptic meshing generation
 
Certainly! The elliptically generated mesh has quadrilateral cells. Most FE analysts prefer triangular cells, which can be obtained by simply cutting the quadrilaterals at opposing corner points.


All times are GMT -4. The time now is 13:08.