hull resistance need help
 

hull resistance

I am trying to set up a problem in fluent. I want mesh a ship in gambit. I want to use multiphase model.

Would you please help me. This is my turn project and I need a detailed help as a university student.

I would appreciate if you help me.

Thanks for advance.

Regards Hakan

Re: hull resistance need help
 

Let us assume you can export your geometry as IGES or STEP file, import that file in GAMBIT, roll up your sleevs and form one volume form the imported faces. Construct a big brick around that ship, subtract the two using boolean operations, and pay attention not to keep the ship volume. Mesh the resulting volume, enjoy your project.

Re: hull resistance need help
 

Hakan

I received your email, I guess it is better,for the time being, to use the forum, the discussion might call the attention of other readers and you might get useful feedback.

It is a floating moving ship, that is ok, but for a fluid dynamicist, it is just a moving body in a fluid. From this perspective, you have to follow the same guide lines as any other moving boundary problem.

1- Define the geometry

2- Define the model eqations that will be solved and the methodology to follow in order to meet the particulars of the problem you are solving.

3- Define the boundary conditions

Here is the plan detailed:-

1- Geometry: You already have done that (you have to read point 3, there is some modifications to be introduced).

2- We are going to solve the continuity and momentum equations, there is no need at this stage to solve the energy equation. Neither we need to consider the air resistance (It can be added latter on, just let us solve a simplified model and start). Because the water surface just reaches the water line of the hull, we need to use the VOF method (There are two good tutorials included in the help system of FLUENT and associated user center, down load and practice with them untill you are comfortable using the VOF method.

3- Boundary conditions

Ship hull, this is a moving wall

Water surface, this is an interface between the water and air Water domain, the outer limits are far field condition.

If you decide at this stage to include the drag caused by the air in your model, then the outer limits of this fluid region would have the same far field condition.

Important: From this last point you see that the Brick volume you have created in Gambit has to be split into two so one of them would represent the water fluid region and the other would represent the air fluid region. Although the hull velocity is small, but the boundary layer would be turblent half way of the hull (you can calculate the Re number based on the hull length to check this assumption), so when you solve the momentum equation you have to activate a suitable turbulence model, the k-e model would be fine.

There are two sources of information availble for you

1- The FLUENT Support engineer responsible for your university and the FLUENT Help system available on your computer

2- Use the internet, google for "ship drag" you will find more than you need. Let me know if I can be of further help to you. Good Luck

Re: hull resistance need help
 

I see there are many people interested to hull resistance. I want to post you, Ahmed, a problem I can't solve. I did many analysis on a tanker hull, I choose the 1:1 scale and VOF model. In this way, the convergence is more difficult, because I have Re = 10E7 instead 10E5, and you have to pay attention to mesg quality, discretization order and turbulence models. However, the results are very good, compare to tanks data. My objective was also to capture the wake at propeller plane, for this I choose a 1:1 scale, to avoid unknown scale effects. In all my analysis w (Taylor notation) is 0.20-0.23, the expected value is 0.29-0.32. This means, velocity inflow at propeller is greater then expected. I tried k-w SST, k-e Realizable, different type of mesh... I have always a lower w. Yes, there are a separation zone behind hull, but it is well predicted. From other parameters, i suspect my model is very good for "global" values (overall lift and drag, wave pattern..) but not so for "local" variables, like velocity in a little surface behind hull. Have you any experiences or suggestion about? Thanks and regards Nando