yade-users team mailing list archive

[Jan Stránský <question690942@xxxxxxxxxxxxxxxxxxxxx>]

Question #690942 on Yade changed:
https://answers.launchpad.net/yade/+question/690942

Jan Stránský proposed the following answer:
Hello,

> I am new to YADE

welcome :-)

> There are some of the questions ...

next time please open separate "ask a question" for each question [1]

> 1...
> peri3dController
> How to determine the dimensions of the concrete cube
> there is an initial size = 1.2, but apparently, this is not the cube dimension.

there is no "cube", the simulation is periodic, simulating infinite space.
To get dimension of a periodic cell (which is in no way related to physical dimensions - as it is infinite), you can use O.cell.size [2].
The initSize is passed to randomPeriPack. It creates a loose packing using makeCloud using the initSize size, then compress it. The parameter basically control number of particles in the simulation.

Have a look at yade/examples/concrete [5] for simulation of "physical"
samples.

> 2. How to determine the variation of aggregate granules if specified;

It is not possible "as is", randomPariPack [3] just supports uniform distribution (defined by radius and rRelFuzz parameters).
To use defined particle size distribution, one option is to "rewrite" the randomPeriPack function [4] with adjusted makeCloud call.

> in YADE Book (pg.280)

it is better to use links to online documentation, like [6]

> 3. How to determine the value (goal, xxPath, yyPath, zzPath, zxPath,
xyPath), I'm confused with the determination of the numbers . I have
read it in YADE Book (pg.280), but i dont understand about it.

Please describe more what you do not understand.
The path arguments are just a list of (time,value) pair, i.e. what value is prescribed at what time/iter.
In certain cases, the values may be relative (e.g. the time on scale [0,1], internally rescaling the values with nSteps argument)
The prescribed value is linearly interpolated between these defined points.
Have a look at comments in the example you have posted and compare them with the simulation results.


A note about modeling of concrete using periodic simulations: in case of strain localization, the periodicity has influence on cracks/strain localized area and therefore influence on overall response.
Which is not a problem "by default", one just need to (should) have this in mind while using it.

cheers
Jan

[1] https://www.yade-dem.org/wiki/Howtoask
[2] https://yade-dem.org/doc/yade.wrapper.html#yade.wrapper.Cell.size
[3] https://yade-dem.org/doc/yade.pack.html#yade.pack.randomPeriPack
[4] https://gitlab.com/yade-dev/trunk/-/blob/master/py/pack/pack.py#L566
[5] https://gitlab.com/yade-dev/trunk/-/tree/master/examples/concrete
[5] https://yade-dem.org/doc/yade.wrapper.html#yade.wrapper.Peri3dController.xxPath

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