CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   ANSYS (https://www.cfd-online.com/Forums/ansys/)
-   -   Numerical investigation of wind behaviour around a marine platform (https://www.cfd-online.com/Forums/ansys/98060-numerical-investigation-wind-behaviour-around-marine-platform.html)

 MarTar March 2, 2012 03:51

Numerical investigation of wind behaviour around a marine platform

Hi all,
i have to perform a numerical investigation of the wind/air flowfield around a marine platform, by means of a commercial computational fluid dynamic code (probably Ansys CFX).

I have already realized CFD simulations for other applications (wind turbine, turbomachinery and heat exchanger), but the task of simulating the air flowfield of a marine platform is new for me and, perhaps, partially out of my personal skills and knowledges.

1) the possible solutions for modeling the interface between air and water;
2) the issues potentially related to modelling the interface air-water (e.g. numerical
considerations, computational effort involved in such a modeling or other aspects);
3) guidelines in modeling air turbulence;
4) the existence of a way of importing meteorological variables (wind values and
directions, obtained from previous meteorological simulations) into the CFD model of the
platform (e.g. i want to use the meteorological fields around the platform as a sort of
boundary/initial condition).

Also the answer to one of the previous questions is well appreciated.

 MarTar March 9, 2012 04:20

Anyone can help me?

Even general guidelines can be useful for me, in order to better focalize some issues of this kind of analysis.

 scipy March 9, 2012 05:45

What is the objective of your simulation? Does the water need to be taken into consideration? Are you looking to model mechanical stresses on the pylons of the platform due to waves and other phenomena? Does it have to be multiphase? Can your simulation be done as a steady state or does it have to be transient? Any exhausts from the platform (like CH4 species maybe) or other dissipation?

If you are only after airflow around the platform, then the water could be taken into account as an either regular (stationary) or moving/translatory wall with an average surface roughness assigned to it. The domain around the platform would have to be large enough, so probably a half-sphere that would extend a few hundred meters radius away from the platform.. Then farfield velocity components (sin x, cosx etc) can either be assigned as fixed, or as a parameter based on some sort of meteorological data if you really need to run a transient calculation..

Nothing else comes to mind right now :)

 MarTar March 9, 2012 07:14

The main target of my analysis is the flowfield characterization around and above a marine platform, with particular regard to the air-flow. In fact, i have to understand what kind of influence an offshore platform exert on the local wind, mainly from a "meteorological" point of view. In so doing, i prefer an unsteady simulation, modeling the air-water interface but, in this moment, i can neglect the winds loads on the emerged structures of the platform and the wave loads on the summerged ones.
These are the principal areas about which i would need clarification:
- modeling the air-water interface: what kind of approach are usable with a commercial CFD
code? Is it necessary air-water modeling for the targhet of my analysis?
- the way of importing local wind, calculated with a numerical meteorological code, into the
CFD calculation, as a sort of boundary condition.

Again, thanks for support!

 scipy March 9, 2012 09:20

I think you should forget about the water-air interaction. It's not going to yield anything useful for an airflow analysis over a platform. I also don't know in what data format is the meteorological information available, but if you can get wind speeds and direction, that can probably easily be converted into a x-y format and then imported as a data set into Fluent.. there is probably documentation and help files available on this topic.

Also, I don't know how complex your geometry is, especially if you are trying to judge the obstruction that the platform is creating to local airflow.. If the platform is located far enough off shore then it prolly won't have much effect on the airflow in general, but if it's located in some sort of a cove and you're looking to simulate the influence on the local airflow then the geometry and the mesh size could be quite sizable.

Do you plan on including things such as stairs, handrails, other details of the geometry? Or is the platform going to be just 4-8 cylinders sticking out of the water and then a block on top?

 eslamkashi March 29, 2012 08:33

Hi
I think geometry creation and meshing it, is your essential problems. If you want import from pdms I can help you, but in meshing I need help too
I before import wind data to cfx, it possible

 MarTar July 27, 2012 05:23

Quote:
 Originally Posted by eslamkashi (Post 352171) Hi I think geometry creation and meshing it, is your essential problems. If you want import from pdms I can help you, but in meshing I need help too I before import wind data to cfx, it possible

Can you explain me what is "pdms"?
How do you import wind data? The is, which format you use? In which way you use wind data as a sort of boundary conditions?

Thanks for help!

 eslamkashi July 28, 2012 04:34

Hi, MarTar

PDMS is a software for 3D plant design http://www.aveva.com/en/Products_and_Services/Product_Finder.aspx#open:34CB3956-1861-4715-AB4C-F62E8D68AF64

 tsameemservers July 28, 2012 08:36

I have a question ,

Can you tell me your CPU and RAM for smoothly performing this ?

 eslamkashi July 28, 2012 11:06

Quote:
 Originally Posted by tsameemservers (Post 374111) I have a question , Can you tell me your CPU and RAM for smoothly performing this ?
Core I7 950 +12 gig ram

 tsameemservers July 28, 2012 13:51

is VGA Important?

 eslamkashi July 29, 2012 10:22

I dont know, no!
Do you have harpoon? It seems very usefull for this purpose, can anyone tell about thats license ?

 All times are GMT -4. The time now is 01:08.