yade-users team mailing list archive

[Václav Šmilauer <question136034@xxxxxxxxxxxxxxxxxxxxx>]

Question #136034 on Yade changed:
https://answers.launchpad.net/yade/+question/136034

Václav Šmilauer posted a new comment:
I think the criterion be derived from an exact definition of "quasi-
static", which we don't have (the definition is necessarily also
"quasi"). It is clear that it is a measure that will be zero for static
case (no motion, or more generally no accelerations, since we are in
newtonian physics); since exact zero will be only rarely achieved, one
needs to set some tolerance, and therefore it should be an admimensional
measure, since otherwise the threshold will have different meanings for
different problem scale.

The easiest adimensional params are ratios of 2 quantities with the same
dimension; unbalanced force is one, but there are other possible: I am
not sure if energy balance would make any difference (kinetic/elastic
potential) -- it could, since physical systems conserve energy rather
than forces, therefore even if they are both adimensional and one will
be monotnically increasing function of the other, it can be a better
measure. Never tried, though.

The disadvantage of unbalanced force (or the energy ratio) is that it is
not defined when there are no interactions (leading to conjuncted
conditions like O.iter>1000 && utils.unabalncedForce()>0.05), which is
unavoidable, seems to me. The subsequent evolution, though, as
interactions get established, could be made more corresponding to
expectations by multiplying e.g. by Mi/M (where Mi would be mass of
particles which are in interactions, and M total mass of all particles)
or something similar.

(I am sorry to have no good answer, take it just as a comment...)

You received this question notification because you are a member of
yade-users, which is an answer contact for Yade.