Yade

soil-tire contact by FEM-DEM coupling

Asked by Ryota Nakanishi

Hi all,

I would like to analyze contact problems between soil and tire by applying the coupling between DEM, i.e. YADE, and FEM like Ref[1]. I think what I need is surface coupling and volume coupling. I read the example scripts in Ref[1], but I couldn’t understand how to execute them.

Judging from the response in Q&A site of YADE, I think new documentation is not yet available on the coupling of YADE with FEM, except for the recent example using escrypt [2]. Does anyone know the recent status of development on coupling of DEM and FEM by adding simple python script?

Thanks,

Ryota

Ref:
[1] J. Stránský & M.Jirásek. OPEN SOURCE FEM-DEM COUPLING. Proc. 18th International Conference ENGINEERING MECHANICS 2012, pp. 1237-1251.
[2] https://yade-dem.org/doc/FEMxDEM.html

Question information

Language:
English Edit question
Status:
Solved
For:
Yade Edit question
Assignee:
No assignee Edit question
Solved by:
Jan Stránský
Solved:
Last query:
Last reply:
Revision history for this message
Hongyang Cheng (alex-cheng) said : 		#1

Dear Ryota,

I had worked on soil-geosynthetic interaction based on Ning's approach.
Please see [1] for the surface coupling scheme.

For further details, please contact me via <email address hidden>.

[1] https://link.springer.com/chapter/10.1007/978-981-10-1926-5_47

Best regards,

Hongyang

Revision history for this message
Best Jan Stránský (honzik) said : 		#2

Hi Ryota,
I will put my codes to github soon
cheers
Jan

Revision history for this message
Hongyang Cheng (alex-cheng) said : 		#3

Hi Ryota,

I uploaded the scripts on github some time ago.
Not sure it will work for others, but worth mentioning if one would like to have a try.

Note that the script actually comes from Jan's contribution years ago...

https://github.com/chyalexcheng/multiscale

Cheers

Revision history for this message
Ryota Nakanishi (matr7) said : 		#4

Hi Hongyang and Jan,

Thank you for your help, Hongyan! I will try your scripts.

And Thank you for your reply, Jan! I'll look forward to see your codes on github.

Thanks a lot.
Ryota

Revision history for this message
Klaus Thoeni (klaus.thoeni) said : 		#5

Hi Ryota,

not sure if it helps but yade has deformable elements the tire could be represented with. If you are interested in the soil-tire interaction I would consider this an excellent approach. Nevertheless, if you are interested in stresses in the tire it would not be suitable. Have a look here [1] for examples of using such pFacet elements and here [2] for the theory.

HTH
Klaus

[1] https://github.com/yade/trunk/tree/master/examples/pfacet
[2] http://www.sciencedirect.com/science/article/pii/S0266114415001235

Revision history for this message
Ryota Nakanishi (matr7) said : 		#6

Hi Klaus,

yeah, I know this paper and I'm thinking of using the method as an alternative plan.
But why is it unsuitable for watching stresses in the tire by this method? I think representing the tire including the internal parts of it enable to see stresses in it. But of course, this approach needs a lot of DEM elements and then require a lot of computation time. This is one of the reasons why I want to use FEM-DEM coupling.

Thanks a lot.
Ryota

Revision history for this message
Klaus Thoeni (klaus.thoeni) said : 		#7

The current implementation of the pFacet is based on gridConnections, i.e., it is a discrete element with "internal contact forces/moments" in its edges. The formulation is not continuum based as the FEM. You could calculate the stresses from the "internal contact forces" but not sure how accurate this is going to be. In terms of computational effort, I think it would be the same. However, the pFacet should be more robust for contact tracking, especially for sliding contacts.

HTH
Klaus

Revision history for this message
Ryota Nakanishi (matr7) said : 		#8

Thank you Klaus, I understand why that is not suitable for calculating stresses in the tire.

Revision history for this message
Ryota Nakanishi (matr7) said : 		#9

Thanks Jan Stránský, that solved my question.