yade-users team mailing list archive

[Bruno Chareyre <question251712@xxxxxxxxxxxxxxxxxxxxx>]

Question #251712 on Yade changed:
https://answers.launchpad.net/yade/+question/251712

Bruno Chareyre posted a new comment:
I think 1/ is a postprocessing quantity  reflecting how contacts behave
in average. Not wrong in itself but not linked to some average stress in
the particles or in the bulk (afaik it was used by Vaclav in combination
with the CPM-type laws). If you want per-particle stress
bodyStressTensors() should be ok.

2/ I agree with Jerome: getting p and q is straighforward and it doesn't
need eigen values. No need to solve a system of equations either. All
this is independent of tesselating or not. Please be more specific if
something has to be clarified.

@Jerome
Not sure I understand what you want to commit. Getting principal stress could be an interesting option but keep it optional because in most cases it would be a pain to get results in this form: if one needs the average stress in all particles it can't be the average of the principal stresses which are all relative to a different local coordinate systems. That is why we need micro-stresses in a single fixed frame in the first place.

Getting eigenvalues from the (x,y,z) matrix for a vtk export is also
straightforward:

Yade [3]: numpy.linalg.eig(Matrix3(1,2,3,2,1,0,3,0,2))
 ->  [3]: 
(array([ 5.        , -2.30277564,  1.30277564]),
 array([[-0.66666667, -0.73465607, -0.12584124],
       [-0.33333333,  0.44487192, -0.83125078],
       [-0.66666667,  0.51222011,  0.54146663]]))

It is also possible to compute eigen values in paraview but it would
need to type the equations it seems.

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