yade-users team mailing list archive

[Fu zuoguang <question258445@xxxxxxxxxxxxxxxxxxxxx>]

Question #258445 on Yade changed:
https://answers.launchpad.net/yade/+question/258445

    Status: Answered => Open

Fu zuoguang is still having a problem:
Hi,

    All the codes above are about to model the shearing bands in PSC
with a given PSD . In this case, particle rolling should be considerd as
one of the most important mechanics in each contact for those macro
bands formation and evolution. So:

(1). I employed the 'coFrictMat' as the particle material and 'cohesiveLaw' as the contact law to consider relative rolling. The detials of engine definition in my aodes above are as follow: 
########################################################################
O.engines=[
        ForceResetter(),
        InsertionSortCollider([Bo1_Sphere_Aabb(),Bo1_Box_Aabb(),Bo1_Wall_Aabb()]),
        InteractionLoop(
            [Ig2_Sphere_Sphere_ScGeom6D(),Ig2_Box_Sphere_ScGeom()],
            [Ip2_FrictMat_FrictMat_FrictPhys(),Ip2_CohFrictMat_CohFrictMat_CohFrictPhys()],
            [Law2_ScGeom_FrictPhys_CundallStrack(),Law2_ScGeom6D_CohFrictPhys_CohesionMoment(
                useIncrementalForm=True,
                always_use_moment_law=True,
                label='cohesiveLaw')]
        ),
        GlobalStiffnessTimeStepper(active=1,timeStepUpdateInterval=100,timestepSafetyCoefficient=0.8),
        consolidation,
        NewtonIntegrator(damping=.2),
    ]
######################################################################

(2). particle generation with a given PSD. In Yade, the psd.py in examples suggested me to use sp0.makeCloud() class 
for this purpose. In this class, There are 5 keys:

1. mincorner and maxcorner to define the generation box.
2. particles number.
3. psdSizes
4. psdCumm
5. distributeMass = 0 or 1

Now the formation of  psdSizes and psdCumm in final codes are inputed
from a 'psd.txt' data, which is #4.

(3). I have tested that if I need to consider the two things above
simultaneously, there are something wrong when the running starts.

(4). In my own testing process, I choose the two different contact laws,
one is labeled 'cohesiveLaw' as mentioned above and the other is the
most common law 'Ip2_FrictMat_FrictMat_FrictPhys()' and two groups of
psd data(psd-1 and psd-2 in #4).  Other testing conditions are all the
same. The summary of results are that

1. When using 'cohesiveLaw' with psd-1, there is an error mentioned in #5 but with psd-2 there is nothing wrong.
2. When using 'Ip2_FrictMat_FrictMat_FrictPhys()' with the two psd data. All of them can run completely.

So I doubt that there may be something difficult in the combination of
the two considering tasks.

Send my whole running codes with the condition data is not easy in this website, so I copied all the testing codes in this page.
I am sorry for bringing you some troubles in your reading and  need your help to check the reason.

Cheers

-- 
You received this question notification because you are a member of
yade-users, which is an answer contact for Yade.