Yade

a simple question about a shear mode of concrete

Asked by zengzhihong

Hi all,
I'm a green hand of Yade users.I'm trying to build a mode of shearing of the concrete.
I consided to solve this basic on two simple examples which the one is the simpleshear and the other is uniax of concrete.
I mixed those two source, but still can not get the stimulate of the shearing concrete test
Here is part of the source:

from yade import plot
from yade.pack import *

O=Omega()

length=0.1
height=0.02
width=0.04
thickness=0.001

dtSafety=.8
sphereRadius=3.5e-3
intRadius=1.5
strainRateTension=.05
strainRateCompression=.5

concreteId=O.materials.append(CpmMat(young=24e9,frictionAngle=atan(0.8),poisson=.2,density=4800,sigmaT=3.5e6,relDuctility=30,epsCrackOnset=1e-4,isoPrestress=0))

pred=inAlignedBox((0,0,-0.02),(0.01,0.02,0.02))

leftBox = box( center=(-thickness/2.0,(height)/2.0,0), extents=(thickness/2.0,5*(height/2.0+thickness),width/2.0) ,fixed=True,wire=True)

lowBox = box( center=(length/2.0,-thickness/2.0,0), extents=(length/2.0,thickness/2.0,width/2.0) ,fixed=True,wire=True)

rightBox = box( center=(length+thickness/2.0,height/2.0,0), extents=(thickness/2.0,5*(height/2.0+thickness),width/2.0) ,fixed=True,wire=True)

upBox = box( center=(length/2.0,height+thickness/2.0,0), extents=(length/2.0,thickness/2.0,width/2.0) ,fixed=True,wire=True)

behindBox = box( center=(length/2.0,height/2.0,-width/2.0-thickness/2.0), extents=(2.5*length/2.0,height/2.0+thickness,thickness/2.0), fixed=True,wire=True)

inFrontBox = box( center=(length/2.0,height/2.0,width/2.0+thickness/2.0), extents=(2.5*length/2.0,height/2.0+thickness,thickness/2.0), fixed=True,wire=True)

O.bodies.append([leftBox,lowBox,rightBox,upBox,behindBox,inFrontBox])

spheres=pack.randomDensePack(pack.inAlignedBox((0,0,-0.02),(0.1,0.02,0.02)),spheresInCell=2000,radius=3.5e-3,material=concreteId)
O.bodies.append(spheres)

O.engines=[
 ForceResetter
 InsertionSortCollider([Bo1_Sphere_Aabb(aabbEnlargeFactor=intRadius,label='is2aabb'),],verletDist=.05*sphereRadius),
 InteractionLoop(
  [Ig2_Sphere_Sphere_ScGeom(interactionDetectionFactor=intRadius)],
  [Ip2_CpmMat_CpmMat_CpmPhys()],
  [Law2_ScGeom_CpmPhys_Cpm()],
 ),
 NewtonIntegrator(damping=.4)

 ]
O.dt=dtSafety*PWaveTimeStep()

O.engines = O.engines+[KinemCTDEngine(compSpeed=0.5,sigma_save=(),temoin_save=(),targetSigma=40000.0,LOG=False)]

ss2sc.interactionDetectionFactor=1.
is2aabb.aabbEnlargeFactor=-1.

First ,I remember that it would set UniaxialStrainer if we did a uniax test of concrete,but I don't know how to set the strain of the concrete when I wanna to do a shear test.

Second, I wonder if the KinemCTDEngine, KinemCNDEngine which make the strain by controlling walls can be work in the concrete material...?If they could ,what else parameters was needed? Otherwise, how can I put a shear-strain to the concrete material..?

Thanks for any suggestions,

Question information

Language:
English Edit question
Status:
Solved
For:
Yade Edit question
Assignee:
No assignee Edit question
Solved by:
zengzhihong
Solved:
Last query:
Last reply:
Revision history for this message
Jérôme Duriez (jduriez) said : 		#1

Hi,

For the second part of your question, the engines KinemCTD/CND/... are designed to deform "a parallelepipedic box" (according to the doc). As soon as your simulation involves such box, constituted itself of 6 "Box" bodies, it should work, regardless the material.
You will need indeed to define adequate engine parameters values, e.g. the ids of the "Box" bodies. There is an example in example folder : simple-shear/simpleShear.py

Another solution to perform simple shear simulations would be to use periodic simulations. These two methods were somewhat discussed here : https://answers.launchpad.net/yade/+question/249377

Revision history for this message
zengzhihong (919127972-9) said : 		#2

Thanks sincerely for Jérôme Duriez.
I learn a lot from the link.
Althought what I want to stimulate is the concrete sample under the earthquake effect.Actually, from my perspect, it is difficult to stimulate the earthquake waves as simple periodic loadings.Maybe because I still can not handle Yade quietly well. I will try to take some time to consider this question after this week.
Thanks again for your help .~! :)

Revision history for this message
Bruno Chareyre (bruno-chareyre) said : 		#3

>it is difficult to stimulate the earthquake waves

Stationary waves are periodic, propagating waves are not (at least in the general case). So periodic BCs may or may not apply depending on what you want to do. Defining a consistent boundary value problem seems to be the first step, before speaking of DEM and Yade. For the moment it is not very clear what you have in mind.