Yade for modelling Clay soils in shear

Hello Mehdi,

for those not so familiar with clays (like me), could you please be more
specific on how to model the forces (e.g. the formulas, how distant they
are expect to act etc.), what is the size and shape of particles
("plate-like" is too general for me)? Also how specifically (just as a user
/ you don't mind some programming) and in what time (1 month, 1 year, whole
PhD...) you want to use Yade / need some results?

Yade is open source, so you can adjust it to your needs. Any kind of
contact law can be implemented supporting also "distant" contacts. Also any
shape can be implemented, or any shape can be approximated by clumping
particles of already existing shapes.

Yade has also active group of users and developers, so it is likely you get
some support concerning your problems (see e.g. [1])

So resume, Yade "as is" can't model what you want, but it can be
implemented. Please send us more information and I am sure we can decide if
Yade is suitable for you and your purpose or not :-)
Cheers
Jan

[1] https://answers.launchpad.net/yade

2016-01-16 9:51 GMT+01:00 <email address hidden> <
<email address hidden>>:

> New question #281059 on Yade:
> https://answers.launchpad.net/yade/+question/281059
>
> Hi
>
> I am new to Yade and about to choose an open-source DEM tool for modelling
> shear strength of fibre reinforced clay soils. Is Yade a suitable tool for
> this purpose especially for modelling the behaviour of soft clay soils with
> respect to physico-chemical interactions between two clay particles (i.e.,
> Van der waals and double layer forces) and also plate-like shape of clay
> particles. Your input in highly appreciated.
>
> Kind Regards
> Mehdi
>
> --
> You received this question notification because your team yade-users is
> an answer contact for Yade.
>
> _______________________________________________
> Mailing list: https://launchpad.net/~yade-users
> Post to : <email address hidden>
> Unsubscribe : https://launchpad.net/~yade-users
> More help : https://help.launchpad.net/ListHelp
>

Hi Jan

Thanks for your insight into my query.

Particles of cohesive soilssuch as clays are platelike and very small in size (e.g., cuboids of dimensions 1.0 μm×1 μm×0.06 μm). There are not only mechanical interactions between two clay particles, but also physicochemical interactions. Properties such as the stress-strain behavior and shear strength of such a material result from a complex microscopic interactions of particles and interparticle forces.

I am happy to start programming, provided Yade is capable of doing a 3D DEM analysis on bodies of clay particles reinforced with fibres (with random distribution in the clay soil). Just feel it is a lot to start with as I am new.

So will Yade is capable of solving such applications?

Thanks
Mehdi

Hello Mehdi,

the easiest approach for the beginning is to model everything with spheres
/ clumps of (possibly overlapping) spheres. With this approach, you can
approximate various shapes (plate-like cuboids, fibres...) and it is very
easy

However, for "distant" interactions, each sphere would behave on its own.
So if two such clumps interact, in fact each sphere of one clump interact
with many spheres from the other clump, so resulting force on one point of
the first clump would be something like integral over the second clump. And
total force on the first clump something like integral over the first clump
summing contributions of each sphere. It could be very nice feature or
could be a problem :-) so please give a bit more information (see below)

What would be needed to implement is the material model. We can help you in
the beginning, creating the very basic c++ files such that you can only
concentrate on the equations. We did it before for bubble interactions and
it worked OK.

So now I have the feeling it could be done with not very big effort. Before
really confirming it (not to be the person who suggested you the completely
wrong direction :-), I have a few more questions / requests:
- do you have / is it easy to create some "patch test", simple analytical
solution to test the implementation? Something like interaction of two
particles and particle to fibre, with a few different orientations.
- what do the forces between particles depend on? Apart from material
properties, it should be only distance.. Does it fit your assumptions?
- concerning forces, could you please try to describe what you would like
to achieve? Some equations, assumptions, how far the forces are supposed to
act (relatively to the clay particle size), what they depend on etc. etc.
- have you used / tested Yade before? It needs Linux, uses Python (so you
need at least some Python scripting knowledge even for the very basic
usage) etc...
- what operating system do you use?
- anything else you feel is relevant..

thanks for the information
cheers
Jan

2016-01-19 3:42 GMT+01:00 <email address hidden> <
<email address hidden>>:

> Question #281059 on Yade changed:
> https://answers.launchpad.net/yade/+question/281059
>
> Status: Answered => Open
>
> <email address hidden> is still having a problem:
> Hi Jan
>
> Thanks for your insight into my query.
>
> Particles of cohesive soilssuch as clays are platelike and very small in
> size (e.g., cuboids of dimensions 1.0 μm×1 μm×0.06 μm). There are not
> only mechanical interactions between two clay particles, but also
> physicochemical interactions. Properties such as the stress-strain
> behavior and shear strength of such a material result from a complex
> microscopic interactions of particles and interparticle forces.
>
> I am happy to start programming, provided Yade is capable of doing a 3D
> DEM analysis on bodies of clay particles reinforced with fibres (with
> random distribution in the clay soil). Just feel it is a lot to start
> with as I am new.
>
> So will Yade is capable of solving such applications?
>
> Thanks
> Mehdi
>
> --
> You received this question notification because your team yade-users is
> an answer contact for Yade.
>
> _______________________________________________
> Mailing list: https://launchpad.net/~yade-users
> Post to : <email address hidden>
> Unsubscribe : https://launchpad.net/~yade-users
> More help : https://help.launchpad.net/ListHelp
>

Hi Dear Jan

Thanks again for your support.
Q. Do you have / is it easy to create some "patch test", simple analytical
solution to test the implementation? Something like interaction of two
particles and particle to fibre, with a few different orientations.

A. Yes we can do this. However, I think it is better to initially solve easier form of our problem (i.e., modelling dry soil without fibres' contribution) and once it is verified with Yade platform then we can add to it to include the effect of fibres. What is the maximum number of particles that can be used in Yade? Assuming that clay particles are very small and a small sample of 50mm (dia.) and 100mm(height) will include millions of particles. Perhaps we can bond some particles together to create agglomerates. However, the other issue is the relative size of the fibres compared to clay particles as they can connect two particles or hundreds of particles.

Q. What do the forces between particles depend on? Apart from material
properties, it should be only distance.. Does it fit your assumptions?

A. Yes distance as well as gravity

Q. Concerning forces, could you please try to describe what you would like
to achieve? Some equations, assumptions, how far the forces are supposed to
act (relatively to the clay particle size), what they depend on etc. etc.

A. Our main problem is to modeling the fibres as a particles with random distribution inside the soil particles and consider their interaction (i.e., non-homogeneous, anisotropic media).

Q. Have you used / tested Yade before? It needs Linux, uses Python (so you
need at least some Python scripting knowledge even for the very basic
usage) etc...
- what operating system do you use?
- anything else you feel is relevant..

We have Yade installed on UBUNTU and we are familiar with Python scripting.

Here is a picture of the soil sample we use (this sample has been sheared and it's been dried)
https://www.dropbox.com/s/88631xwep5laivc/IMG_9585.JPG?dl=0

Hello,

However, I think it is better to initially solve
> easier form of our problem (i.e., modelling dry soil without fibres'
> contribution) and once it is verified with Yade platform then we can add
> to it to include the effect of fibres.

good idea :-)

> What is the maximum number of
> particles that can be used in Yade? Assuming that clay particles are
> very small and a small sample of 50mm (dia.) and 100mm(height) will
> include millions of particles. Perhaps we can bond some particles
> together to create agglomerates. However, the other issue is the
> relative size of the fibres compared to clay particles as they can
> connect two particles or hundreds of particles.
>

The number of particles is theoretically unlimited. What limits maximum
number is:
- available memory of your computer
- available simulation time (the more particles you uses, the more time you
will have to wait for the results)

I personally used maximally 500k particles, approaching the RAM limit of my
computer. Usually I use between 20k and 150k particles without problem

Looking for the picture of the experiment, I am a bit skeptic about the
feasibility to simulate the whole experiment with very small particles.. I
assume it is quasi-static loading, so using explicit dynamic code, which
needs very short time step, would make the simulation time very long..

As an alternative, modeling just part of the specimen should be ok, some
multiscale approach could also be possible

you can try some of the examples [1] modifying the particle size and amount
to get some feeling what is possible. The examples are with spheres,
simulation with any other shape would be slower..

cheers
Jan

[1] http://yade-dem.org/doc/tutorial-examples.html