yade-users team mailing list archive

[steve <question696047@xxxxxxxxxxxxxxxxxxxxx>]

Question #696047 on Yade changed:
https://answers.launchpad.net/yade/+question/696047

    Status: Answered => Open

steve is still having a problem:
Thanks Karol,
I am sorry if I my statement was confusing, let me just explain it again. So I've already obtained the flexural stress & strain form the experiments with a known dimension b=100mm. And then I construct a 2D model in yade in order to simulate this experiments and see the crack pattern. However, during the calibration process, I have some problem in matching the 2D simulation results to the experiment data. Actually, I am not applying force in both experiment and simulation, but control the loading rate of the loading plate and obtain the force that acting on the plates. So in this case, is the force obtained in the simulation supposed to multiply a scale 100/b' (100 is the thickness in the experiment, b' is the thickness of the model) to restore the force to the 3D condition and then use the formula to calculate the stress? If so, then what would be the thickness of the model? is it the average diameter of the particle? or is there no thickness at all because it is 2D simulation and both force and displacement in the out-of -plane direction has been restricted?
And I test both methods in simulation which are by formula sigma=fL/d^2 and by directly get from particles:
###
sigma=bodyStressTensors()
tensile_stress=sum(sigma[i][0,0] for i in p) ## p is the particle ids located in the bottom of the beam between two loading point
###
Although the shape of curves are similar, but the peak flexural stress obtained directly from particles are over two times larger than that of formula. I am really confused where this difference come from if both methods are good enough.
Best regards,
Steve

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