Problem with moving boundary condition

pilou

Dear all,

Thanks a lot for this amazing software.
I'm currently running a case involving moving boundary conditions, basically a flap in box.
It happens, in a lot of simulations, that I get an error message, which is the following "particle boundary out> type : Moving cause"

In fact the bottom of the flap seems to exit the main tank because it's not perfectly on the same line as the axis defined.

It is a 3D case and the flap is a simple rectangle, I noticed that declaring the flap with a small gap at its bottom somehow solves the problem, but I think there is a better way to do so.

I guess there is a problem with the "real" location of the boundary particles that should be slightly different from what is defined in the definition file.
Does anyone knows the exact procedure of boundary particles location and how to overpass this problem ?

Best regards

Pilou

Alex

http://www.dual.sphysics.org/index.php/forums/#10

Alex

Read the FAQ section first.

And if you still have problems send to us an email (dualsphysics@gmail.com)

Regards

ric_arr

I Also have a issue with Boundary condition. Sometimes my simulation are aborted with this problem "boundary particle was exluded" That appends when I improve the movement of the wavemaker (amplitude) or when I choose a particular rapport between smoothing length and dp. I really don't understand why! I try to change (improve) the "big" domain but it doesn't work! I hope you can explain me.

Thanks

Ric

Alex

(i) defining boundary points at the minimum and maximum positions that the particles are expected to reach during the simulation. The option will create a particle at the given location x,y,z.

(ii) using the parameters of DualSPHysics execution mentioned in the answer to question 8:
- domain_particles[:xmin,ymin,zmin,xmax,ymax,zmax]

- domain_particles_prc:xmin,ymin,zmin,xmax,ymax,zmax

- domain_fixed:xmin,ymin,zmin,xmax,ymax,zmax

so that the limits can be specified instead of using initial particle positions

Alex

http://www.dual.sphysics.org/index.php/forums/#10

ric_arr

Thank Alex. is possible to write one negative number in the x min? i try to write -10 and the simulation don't start. I can do something?

Ric

Alex

Negative numbers can be defined as position for sure!
Send to us your case and we will help you.

Remember that:
In the input XML file, the parameters pointmin and pointmax only define the domain to create particles so out of these limits fluid or boundary particles will not be created. The limits of the computational domain are computed at the beginning of the DualSPHysics simulation and use the initial minimum and maximum positions of the particles that were already created with GenCase

ehsan

Dear all
I try to run a case in which the movement of piston wavemaker is defined using data from an external file. But when I do this, the model doesn't progress in time steps during run time (although I get no error messages).
I read the FAQ section (9-How can I define the movement of boundaries? ) where it is mentioned the external file of prescribed movement should include info of time, X-position, Y-position, Z-position and "velocities"; But the "motion08mov_f3.out" file (describes the prescribed motion used in the eighth example of " MOTION" directory) just contains info about time and x,y,z positions and there is no column for velocities! Moreover, there isn't any guide in "GenCase_XML_GUIDE.pdf" file about velocities definition for predefined movement (See page 57). So, the external file I used for wavemaker movement doesn't contains velocities info (like "motion08mov_f3.out" file). I guess model doesn't work maybe because it needs wavemaker velocities info.

Does anyone has ran the case of piston wavemaker with prescribed motion and know how to overpass this problem ?

Best regards
Ehsan

Alex

You do not need to specify velocities, that is extra information that can be useful. Only time and posicion(for example X-position) are required. You load the file and you specify where data is (what column):



or

uisceng

As Alex already said, you don't need to prescribe velocities, only the position versus time needs to be prescribed.....but I also had to do the following:

1) Make sure the file was saved as a .dat file
2) Call that file in the xml file
3) Copy the motion file into the output file (for example, use batch command 'xcopy *.dat %dirout%'

Number three took me a while to realize.

Good luck.

ehsan

Thank you Alex for your reply.
The key point was that mentioned by "uisceng". I followed that three steps and the problem was solved. Thank You very much "uisceng" for your useful advices.

All the best
Ehsan