CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Advection on 2D nonuniform triangular mesh (http://www.cfd-online.com/Forums/main/92640-advection-2d-nonuniform-triangular-mesh.html)

Nereus September 20, 2011 13:34

Advection on 2D nonuniform triangular mesh
 
Hello all:

I have a basic question that I hoped someone could answer. I'm very much a newbie, so apologies if the question is poorly posed.

I have been given files that define a 2D nonuniform triangular mesh and the velocity field over that domain (i.e., at the nodes). I simply want to simulate the introduction of an inert tracer into that the domain and its subsequent transport. So, if the concentration of the tracer is C(x,y,t), then the governing PDE I'd have would be something along the lines of

\frac{\partial C}{\partial t}=-v\frac{\partial C}{\partial x}-w\frac{\partial C}{\partial y}.

where v and w are defined by my velocity field. My question then is, what is the best approach to adopt in solving the PDE given the nonuniform nature of the grid? FEM, FVM, or FDM?

Any advice would be greatly appreciated!

DoHander September 20, 2011 14:46

Both FVM and FEM will work for unstructured grids. I would go with FVM for this problem, but this is up to you.

seyedashraf September 21, 2011 02:06

I was going to say that both FVM and FEM are suitable but this quote
Quote:

given the nonuniform nature of the grid
shows that you better try FEM cause its designed to use node coordinates and its very useful in complex domains and grids


All times are GMT -4. The time now is 01:50.