fluent for ship hydrodynamics

[pavel915]

I want to measure the wake (i.e. velocity losses of water in the propeller plane)behind the stern of ship.
For this i am trying fluent. I have made the model in gambit , by importing the ship from Rhino and made a flow volume like the image i attached. I specified inlet as velocity inlet and outlet as "outflow". The dimension of flow volume is 6L behind the stern and 3L infront of the bow. all other boundaries are specified as wall.

I want to know ,whether my approach and boundray conditions are right or wrong.

Thank you;

[fastwave]

Depends what you are trying to do. If it is a research project and you will compare to wind tunnel or towing tank then it might be best to model the walls of the tank.

Otherwise a far field type boundary might be better.

[fastwave]

I would first run without vof, to investigate the turbulence model and grid density. Once i was happy with those then I night consider the vof. For vof it is best to use a structured grid, and the grid size would dramatically increase as well as the calculation time and instability of the solution.

Try the simple case first.

[pavel915]

The resulf i have found without vof; is closer to the real one for a ship i have varified.
But i am really confused; wheather ; vof will give a better result ; or i can continue without vof taking a appropiate turbulence model.
I want to use Reynolds stress model(RSM) as so far i know it is the best model for this kind of flow.

The details of the bondary coditions i have taken is like the image.

[fastwave]

If you are happay with the agreement then it's your choice. Going vof will open up a can of worms and it seems to me that you are not that experienced with CFD. So start simple. Also if the ship you are talking about is the one in the pic then it looks lika a planning hull. So high speed and dry transom.

That shouls mean that your current model should not agree that well, unsless you are measuring Fwd of the transom.

Might be more appropriate to extend the hull surface from the experimetal data you have.

Vof should give improvement especially in that area if done pcorrectly

[pavel915]

@fastwave,
Thanks for your reply;
Yes i am a student and not much experienced in cfd.
The ship i am investigating is not a plaing hull and the speed is not much; close to 10 knts; the image is attached is just for a sample, i collected the image from a website and my work is just similar to that image except that my ship is displacement ship.
I want to investigate displacement hull with/without immersed transom.

[fastwave]

In that case i would concentrate on getting the grid y+ and turbulence model right and forget about the vof.