The OpenSees Community

Hi, I am trying to model a retaining wall. The wall is dispBeamCol element. I use beamcontact2D for soil wall interaction. when I use UmfPack for the analysis process, the model only is able to excavate one layer of soil and then it will give me an error for the other steps as below:

WARNING: symbolic analysis returns -8 -- Umfpackgenlinsolver::setsize
WARNING:UmfpackGenLinSOE::setSize : solver failed setSize()
StaticAnalysis::handle() - LinearSOE::setSize() failedStaticAnalysis::analyze() - domainChanged failed at step 0 of 1
OpenSees > analyze failed, returned: -1 error flag

Could anyone help with that?
Thanks,

Hi everyone!
sp command is inducing disp or strain to a node?

Thanks!

Ziad wrote: ↑Mon Oct 23, 2023 3:48 am I am building a structure on openseespy. Everything is in order, I have been able to apply a distributed load:
op.eleLoad('-ele', 1, '-type', '-beamUniform', -Wx, 0.0, 0.0).

I wanted to know how can I apply a Triangular distributed load. For example, Load experienced by soil.

Hello, Are you applying the load on quad elements (soil)? I want to apply a distributed load on soil surface. Does this approach work?

Thanks for your help in advance!

Hello all, Could you find any recommended approach to apply distributed load on quad elements? Did the mentioned approach work well in your analysis?

Thank you very much for your response.
Based on your experience, which interface elements available in OpenSees works better for soil-retaining wall interface?

Hi all,
Is it possible to model soil-retaining wall interface using Py-Tz-Qz contacts ? Is there an example for that? Currently I'm using BeamContact2D, but I guess It doesn't work properly at the toe of the wall.

Thank you for your response.

I am using plain strain and PDMY02 constitutive model.

When I compress a single element horizontally, the horizontal and vertical strain values are exactly the same values (one is negative and the other is positive). It is like the element is in constant volume condition. I would like to know why it behaves in this manner and if there is a way to stop it?

Hi all,

In a 2d element test with a quad element, does the quad elemnts have constant value behavior? Is there a way to not behave like that?

Hello All, I am trying to model a retaining wall. I construct the model according the cantilever example in the wiki. However, I like to model the base of the wall in a way to consider the friction in x direction(not a roller at the base). A simple example of what I wrote can be seen below.

I fixed a base node and used a zeroLengthContact2D to attach the fixed base to the base of the wall. The results are still show that there is no friction in the model since changing mu doesn't have any effect on the results. Can anyone help me with this issue?

Thank you so much for your help!

################################################################################

ops.model('BasicBuilder','-ndm',2,'-ndf',3)
ops.node(1,0,0)
ops.node(10,0,1)

ops.element('dispBeamColumn', 1 , *[1,10], transfTag, integTag )

ops.model('BasicBuilder','-ndm',2,'-ndf',3)
ops.node(2,0,0)
ops.node(3,0,0)

ops.fix(3,1,1) #base node

ops.equalDOF(1,2,1,2) #coneccting 2DOF node to Wall 3DOF node

ops.element('zeroLengthContact2D', 10, *[2,3], 1e8, 1e8, 0.4, '-normal', 0, 1)

Thank you! You mean zeroLengthContact2D ? or simple zeroLength with frictional material assigned to that?

Thank you for the response.
I changed the code and it worked for turning on and off the friction. However I still have this problem with updateMaterialStage for PDMY. When I use this approach for that I face an error saying:

WARNING - unable to assign parameter to object of type -material

Thank you very much for your help.

Hi,
Is there a way to define a frictional fixity in opensees? I want to model a roller that is not completely free and follow coulomb friction law.

Thanks

Hello,
I am trying to model a retaining wall and I used BeamContact2D for wall-soil interaction. I set the friction to be zero for generating in-situ stress state using initialstateanalysis. But I face a notice which say "Domain::addParameter - parameter with tag 0 already exists in model" . However, this error does not stop analysis.
Could anyone help me that what is the meaning of it? and how can I solve this? Does it affect the results?

You can find how I wrote these part of the script below, the error is related to this section.

Thank you very much!



ops.InitialStateAnalysis('on')


ops.updateMaterialStage('-material', 228, '-stage', 0)

ops.setParameter('-val', 0, '-eleRange', 20000, 20020 , 'friction' )
ops.setParameter('-val', 0, '-eleRange', 30000, 30020 , 'friction' )

ops.analyze(4)

ops.updateMaterialStage('-material', 228, '-stage', 1)
ops.analyze(4)

ops.InitialStateAnalysis('off')

ops.setParameter('-val', 1, '-eleRange', 20000, 20020 , 'friction' )
ops.setParameter('-val', 1, '-eleRange', 30000, 30020 , 'friction' )

Thank you for your response!
Since the Nine-Four quad up element has some pore pressure nodes, how can we define those nodes to be dry soil ? The only thing we can change is the fixity for pwp which means opening and closing the drainage. How can we specify that there is no water?

Hello, Is Nine Four Quad element only useful for saturated soil? Can I use it for Dry sand? I used Quad element so far, but I want to use Nine-noded element for better results and more precise shape function.

Thank you for your help!