yade-users team mailing list archive

[Xu <question703169@xxxxxxxxxxxxxxxxxxxxx>]

New question #703169 on Yade:
https://answers.launchpad.net/yade/+question/703169

Hi all,
I have simulated 2D triaxial undrained tests  using sphere and clump particles respectly. However,  there is a problem in terms of energy conaervation. I compare the input work done by external foece and the total of energy components calculated from microscale. These two are worth the same when using the sphere, a claer difference is observed for clump (two spheres or more).
The amount of input work can be written in an incremental form as
dW=sijdeij
The total energy calculated from the microscale
Etotal=Estrain+Efriction+Edamp-(Estrain_ini+Efriction_ini+Edamp_ini)

So I check the script [1]. The cause of the difference may lie in line 29,30. I found that the Ra and Rb reperent the radius of the basal sphere in clump, but not the radius of the clump. I know the clump do not have the radius, so I replace Ra, Rb with the radius of the sphere having the same volume as the clump. Then the microscale energy calculated by the new Kn,Ks (calculate by the line 35,37 in [1]) has no difference with input work both in the simulation of sphere and clump （the error<1%). I am not sure if  i am right, or if you have another explanation for this phenomenon.

And i have another question. If my opinion is right, the calculation of Kn,Ks is wrong for the clump, the calculation of the normal stress and shear stress in the interaction are also wrong.  And i don't know how to fix this problem.                     

I really appreciate if you can confirm the issue or help me to solve the problem.
Thanks a lot,
Xu

[1] https://gitlab.com/yade-dev/trunk/-/blob/master/pkg/dem/FrictPhys.cpp

-- 
You received this question notification because your team yade-users is
an answer contact for Yade.