yade-users team mailing list archive

[Jan Stránský <question690872@xxxxxxxxxxxxxxxxxxxxx>]

Question #690872 on Yade changed:
https://answers.launchpad.net/yade/+question/690872

    Status: Open => Answered

Jan Stránský proposed the following answer:
> concerning the class computeNewAlfa

it is a method/function, not class :-)

> How do i precise ...

however you want / your model needs..

> phys->alfa = alfa + om->dt *( ... )
> How do i precise ... that OldAlfa is equal to NewAlfa for the previous time step

depnding on the formulation, e.g. you can do it like in the original
post, just the "phys->alfa = alfa + om->dt *( ... )" is defined in
computeNewAlfa function (which can be composed of several other
functions, like computeTerm1(...) etc.)

bool Law2_ScGeom_VsmPhys_Vsm::go(...)
{
   VsmPhys::computeAnything(phys->alfa); // phys->alfa is "oldAlfa"
   ...
   phys->alfa = computeNewAlfa(phys->alfa,...); // phys->alfa argument is "oldAlfa", phys->alfa result is "newAlfa"
}

or if you want the code being more "self-explanatory", the same can be
rewritten as:

bool Law2_ScGeom_VsmPhys_Vsm::go(...)
{
   Real oldAlfa = phys->alfa;
   computeAnything(oldAlfa);
   ...
   Real newAlfa = VsmPhys::computeNewAlfa(oldAlfa,...);
   // now you can do computations with both oldAlfa and newAlfa if needed
   phys->alfa = newAlfa;
}
/////

> corresponding to a specific interaction

being member of IPhys = corresponding to a specific interaction

> how is the value of NewAlfa stored between two time steps... ?)

as phys->alfa (??)
if you need values in two time steps (current and previous), you can store two values in IPhys

cheers
Jan

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