BCs in a Tunnel Modeling
 

Hi everyone,

I’m trying to study the velocity distribution next to the uppermost part of a car tunnel. The problem has been described below (I know this not a precise modeling, but all I need is a rough estimation). Here is the BCs I applied:



AA:

1- Moving wall (L=50 m, V=17.6 m/s)

2- Pressure-Inlet

3- Fixed wall

4- Pressure-Outlet


BB:

1- Moving wall (L=100 m, V=17.6 m/s)

2- Pressure-Inlet

3- Fixed wall

4- Pressure-Outlet


CC:

1- Moving wall (L=50 m, V=17.6 m/s)

2- Pressure-Inlet

3- Fixed wall

4- Outlet-vent


DD:

1- Moving wall (L=50 m, V=17.6 m/s)

2- Pressure-Inlet

3- Fixed wall

4- Outflow


The results are not similar, neither make sense! BTW, I solved both steady and unsteady. Also Turbulence (k-epsilon) modeling has been used. I’m not sure about 2 and 4 boundaries.

Does any one know what the best BCs are?


http://mehdifamouri.persiangig.com/01.JPG

and here is the Computational Domain with the BCs:
http://mehdifamouri.persiangig.com/02.JPG
Thanks in Advance,

Correct boundary conditions are
1 - ground - moving wall at 17.6m/s
2 - outlet - pressure outlet 0Pa gauge
3 - roof - moving wall at 17.6m/s
4 - velocity inlet at 17.6m/s

assumes:
>you are modelling car in an actual tunnel and not wind tunnel which has fixed relative walls
>domain is large enough to allow turbulence profile to establish before impacting vehicle
>upper boundary is the actual roof and not domain cut off in which case it should be symmetry
>Note that wheels need MRF zones
>Ensure you use prisms on car if possible and have high density wake region captured behind

Hope that helps!