CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   STAR-CCM+ (http://www.cfd-online.com/Forums/star-ccm/)
-   -   PMM simulation (http://www.cfd-online.com/Forums/star-ccm/113820-pmm-simulation.html)

Ankush K February 27, 2013 07:48

PMM simulation
 
Hi,

I am trying to simulate Planar motion mechanism(PMM) test on a ship using CFD.
I want to generate sway and yaw motion on the hull using STAR-CCM+.
I want to know the steps involved and the morphing technique used. :)

ryancoe February 27, 2013 09:22

I've done this and documented some of the details in a conference paper:

http://ieeexplore.ieee.org/xpls/abs_...number=6405027

Within STAR, this test can be accomplished by using a time-dependent Rotation and Translation motion (no morphing necessary).

grad@itu February 27, 2013 12:24

you can use 6DOF motion and Volume of Fluids (VOF) options for your need. With VOF model you defines fluid volume and with 6DOF motion you can define most of the rigid body motion and different types of waves (sinuoidal,random ot something like that)

I give an example application link that i did, can you look that? is this your problem?

http://www.youtube.com/watch?v=7KKNF3_I_7A

this account have 4 application for now and all of these are related to same phenomena.

Ankush K February 28, 2013 02:11

How should I perform grid independency test to finalize my mesh(cell size, etc.)?

siara817 February 28, 2013 05:53

You need to run your simulation and after convergence check a physical property. Then refine your mesh and run again. After convergence, check that physical property to see, how much is the change. There is a literature source, the author is Roache.
Roache, P. (1994) Perspective - a Method for Uniform Reporting of Grid Refinement Studies, Journal of Fluids Engineering-Transactions of the Asme, 116, 405-413.

grad@itu February 28, 2013 08:38

I totally agree to Siara817 excellent and clear answer..
Briefly,

* The solution must be converged with relatively coarse mesh
*You must check the physical properties, force,heat flux, mass flux for example.. What are the value of these datas and note it.

Again,

* The solution must be converged with relatively finer mesh
*You must check the physical properties, force,heat flux, mass flux for example.. What are the value of these datas and note it.

Every step, you must note it physical results and you must chech the changes between coarse and finer mesh (coarse - less coarse - finer - more fine .....).

After look the changes, if the changes reduce between any step - or acceptable changes like 0.00348 for ex. between third try and fourth try- if the change is so little the its mesh dependence solution of your problem and if you try more finer mesh you ll give unnecessary little change but you have more cell count and more computation and more computational cost except round off and truncation errors!

ryancoe March 1, 2013 14:59

To analyze the convergence of an unsteady simulation such as this, you need to consider both spatial and temporal discretization independence (i.e., mesh size and time-step size). This is made more challenging by unsteady nature of the flow. See this article for a good method (also applied to unsteady surface ship simulation): http://web.mit.edu/weymouth/Public/P...mouth_2005.pdf

Hope this helps.

Ankush K April 2, 2014 02:34

Quote:

Originally Posted by ryancoe (Post 410429)
I've done this and documented some of the details in a conference paper:

http://ieeexplore.ieee.org/xpls/abs_...number=6405027

Within STAR, this test can be accomplished by using a time-dependent Rotation and Translation motion (no morphing necessary).

Thanks Ryan.
I was using Morphing to simulate HPMM for my ship but the results were not very encouraging.
How do you suggest about using DFBI-Rotation and Translation for the same analysis?

Ankush K April 2, 2014 02:37

Quote:

Originally Posted by grad@itu (Post 410470)
you can use 6DOF motion and Volume of Fluids (VOF) options for your need. With VOF model you defines fluid volume and with 6DOF motion you can define most of the rigid body motion and different types of waves (sinuoidal,random ot something like that)

I give an example application link that i did, can you look that? is this your problem?

http://www.youtube.com/watch?v=7KKNF3_I_7A

this account have 4 application for now and all of these are related to same phenomena.

Are you suggesting DFBI-Rotation & Translation for HPMM simulation??


All times are GMT -4. The time now is 08:47.