next up previous contents
Next: The wall function method Up: Implementation of the boundary Previous: Edges

Boundary conditions for the transport equation

  The boundary conditions for the transport equation are much easier to implement than the boundary conditions for the velocity components.
In the case of Dirichlet boundary conditions it is sufficient to use linear extrapolation to eliminate virtual scalars. So for example in Figure 7.16 we use the following formulae:

 

for normal boundary points and

 

for the corner points.

  
Figure 7.16: Cells for scalar quantities and corresponding virtual points.

In the case of a Robbins boundary condition we follow Van Kan et al. (1991). This means that the Robbins boundary condition

 

is substituted in the diffusive term:

 

where is the boundary at which the robbins boundary condition is given. The approximation of the first integral in the right-hand side of (7.76) is the same as for the inner region. The evaluation of the last term of the right-hand side of (7.76) on, for example, the left boundary of the domain is as follows:

In the case of non-smooth grids, the following approximation could be better used:

Virtual scalars are eliminated in the usual way. Similar expressions may be found for the three-dimensional case.


Tatiana Tijanova
Wed Mar 26 10:36:42 MET 1997