svalinn / DAGMC

Direct Accelerated Geometry Monte Carlo Toolkit
https://svalinn.github.io/DAGMC
Other
100 stars 63 forks source link

Parsing or handling of standard FLUKA volume scoring definitions #37

Closed makeclean closed 11 years ago

makeclean commented 11 years ago

Develop routines to appropriately parse and thereby create appropriate handles to the FLUKA scoring routines, i.e. by parsing metadata found within the h5m created via Cubit group names

makeclean commented 11 years ago

The routine should make use of the dagmc functions that return the volume of entities to ensure correct normalisation of rallies. I.e. When writing tally information at similar stage to material data include volume of entity in score definition.

makeclean commented 11 years ago

Suggest we add tally parsing as an argument to the executable, in the way we currently use

$> fludag input_geom.h5m

To produce the material input data, add flags to output scoring data in a similar manner

$> fludag input_geom.h5m -scores score_data.dat

Also suggest that by default the routine prints the physical volume, however we should retain the capability to set the volume of tallies to an arbitrary value as well. This is needed in the case of stochastic volume calculations, which in this instance will serve as a validation tool.

gonuke commented 11 years ago

Regarding scoring volume... is this normally something the user has to provide? Or does this really mean that we have to provide a way for FLUKA's scoring routines to query the volumes?

makeclean commented 11 years ago

In fluka for a volume tally, like a track length tally, the user within the tally specification must supply the volume of the region, you can specify any value you wish greater than 0.

makeclean commented 11 years ago

Perhaps to clarify, this is data that should be provided in the input deck as opposed to via argument or routine.

makeclean commented 11 years ago

Suggest using similar methodology as dagmc where groups are used to create scores. Fluka can score some 60 different particle types as well as derived types like dose or energy.

Something like a group naming convention like

usrtrack_proton_flux_emin_emax

___ There are sufficent variations that some user intervention will be necessary after parsing and printing
julamot commented 11 years ago

Work was done on this (a cubit file was given group names with this convention) but the underscores presented a problem to dagmc_preproc, and the names also overran a 32-char tag limit on the name.

makeclean commented 11 years ago

This has been subsumed by #87

makeclean commented 11 years ago

And #86