[aspect-devel] visualisation plugin

Timo Heister heister at clemson.edu
Thu Mar 30 14:10:39 PDT 2017


Lev,

several options:
1. Generate an AdditionalMaterialOutputs that contains a bool or int
or something for each quadrature point. You can then query this
information, example: melt.cc
2. Create a function inside your material model that your
postprocessor calls. For this you need to do a dynamic_cast to that
specific type of material model. This allows you to query whatever
function you want. Example: melt.cc and melt_statistics.cc
3. We introduce something more generic into the material model
interface to report something like "int local_material_type" in each
quadrature point and generate visualization plugins. This is
potentially useful for different plugins.
4. You could create a compositional field that represents this
information (by using reaction terms in the material model maybe?)



On Thu, Mar 30, 2017 at 4:48 PM, Lev Karatun <lev.karatun at gmail.com> wrote:
> Hi all,
>
> In my simulations I'm using my own visco-plastic material model plug-in. I
> also wrote a visualization plugin that outputs if material is behaving
> plastically or viscously. Currently the way it works is I have a function
> named "viscosity_ratio" in my material model that re-evaluates the viscous
> and plastic components of viscosity the same way as in the evaluate
> function, and returns their ratio. But the problem is - every time I want to
> change something in the formulas, I have to change it in 2 places, which is
> error-prone. So I wanted to change it so that it calls evaluate instead of
> re-calculating the viscosity, but the problem is - the functions take 2
> different types of input. I'm guessing I would have to make a
> MaterialModelInputs vector inside the viscosity_ratio function first to then
> use it as input for evaluate function? Should I be doing it through a
> FEValues object? In general, am I thinking in the right direction? Maybe
> there is an example of this already implemented in some existing plug-in (I
> couldn't find any)? Or maybe there is an easier way to implement this?
>
> Any advice would be appreciated.
>
> Best regards,
> Lev Karatun.
>
> _______________________________________________
> Aspect-devel mailing list
> Aspect-devel at geodynamics.org
> https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.geodynamics.org_cgi-2Dbin_mailman_listinfo_aspect-2Ddevel&d=DwIBaQ&c=Ngd-ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLxWPA_2Wlc4&r=c08Btfq4m9QEScXN3ZQwLZzzWQE7S8CYq1IYuzKV_Zk&m=sRhuIaAN-aPrlrEpsYjxYrvOpQU323JzbwMsBJFPix0&s=oV46wRg6brJWw7kqFKz0f7St9RqQiYl7j-mtErdaec4&e= 



-- 
Timo Heister
http://www.math.clemson.edu/~heister/


More information about the Aspect-devel mailing list