yade-users team mailing list archive

[Fu zuoguang <question295521@xxxxxxxxxxxxxxxxxxxxx>]

New question #295521 on Yade:
https://answers.launchpad.net/yade/+question/295521

Dear Prof. Chareyre and all users,

Fabric analysis is one of the most common methods for detecting those micro-level natures of such granular materials. Recently I have tried to do it in 2D-simulation task in Yade-1.14-daily version with the simplest isotropic condition. But unfortunately, I’m lost in getting the same results as that in Ref.[1]. All the details of my simulation are in Fig.1

Fig dir: http://tinypic.com/view.php?pic=2dwhukl&s=9#.V2kZvbEs-VU

After the consolidation process finished, I listed all the directional data I want to use as the following mode:

#################################################################
Body_Numbers,  Body_Cur_P_x,  Body_Cur_P_y,  Body_Cur_P_z,
           1,                  -1.28536e-01,   0.00000e+00,   0.00000e+00,
           2,                   1.28528e-01,   0.00000e+00,   0.00000e+00,
           3,                   0.00000e+00,  -1.28642e-01,   0.00000e+00,
           4,                   0.00000e+00,   1.28632e-01,   0.00000e+00,
           5,                   0.00000e+00,   0.00000e+00,  -5.00000e-01,
           6,                   0.00000e+00,   0.00000e+00,   5.00000e-01,
           7,                   1.20924e-01,   6.60472e-02,   0.00000e+00,
           8,                  -1.22713e-01,   1.02525e-01,   0.00000e+00,
           9,                   4.91498e-02,   1.84187e-02,   0.00000e+00,
          10,                -4.19541e-02,  -3.18193e-02,   0.00000e+00,
           .....
###########################################

So，I firstly summarized the contact-normal as the vector variable and transformed it from Cartesian to polar axis (using format 1-1) but got a strange result as shown in Fig.2. All the contact-normals are not symmetrically distributed around the origin of principle axes, but most of them are located in the area of (math.pi to 3/2math.pi).

Then, I respectively used the particle position and the contact-point position to execute the same process as above for ensuring that I got a correct output of those data, and plotted them in the same style as in Fig.3 and Fig.4. Now, the statistical results of the two group data show me the same trend. They are distributed axisymmetrically around the principle axes. And in the same quadrant, the number of data located in diagonal area is more than that in others, which can meet the deviatoric-shear condition.

Why is only the contact normal showing the non-symmetric distribution here? 
[1] Are there anything wrong in my understanding of contact normal distribution?
[2] Or Yade has its own pattern in defining the contact normal?

Seeking your help!

Ref.
[1] Distribution of directional data and fabric tensors. 1984_Int.J.Eng.Sci


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