Re: [fluka-discuss]: About BIASING option to get particle distributions

From: Answers <answers_at_pcfluka.mi.infn.it>
Date: Mon, 16 Mar 2020 11:15:42 +0100 (CET)

Dear Vladimir

in general BIASING helps when there is attenuation or anyway further
scattering between the source and the detector(s). My
interpretation is that in your case biasing does not help
because it is applied in regions where particles are more or less
free-flying (air), hence you just get exact copies of the original
particle with reduced weights.

In order to improve statistics it could help to try to bias already
in the target/moderator region, maybe subdividing them into thinner
slices and using biasing to make progressively more likely for particle to
exit. Also it is advisable not to exceed a factor 3 increase in
importance between adajacent regions.

                 Hope this helps

On Fri, 13 Mar 2020, babintsev wrote:

>
>   Dear experts,
>
>  I want to get E-, T- distributions (and large statistics - as the 1-st
>  step) at the large distance from the target.
>  I have checked the BIASING option for this task. - simple detector example.
>
>  Lead target in the polyethilen , P beam with E= 2 GeV.
>  On the right and on the left of the target there are 2 virtual
>  counters (box of air) : C2 and C22.
>
>  C2 is disposed at the interior of 3 air-boxes with BIAS-coefficients:
>   10-100-1000
>
>  C22 - on the left has no BIAS option : direct filling for particles
>          to compare results with C2.
>
>  I saved also input particles at C2 and C22
>  (initial lines are presented)
>
>
>  Results are in zip-file : result.zip.
>
>  My questions :
>  1)  for C2 with BIASING I have got large multiplication of secondary
>      particles - BUT they have the same parameters !
>      As a result, E-distributions for C2 and C22 (no BIAS) are the same.
>
>      So, BIASING here can improve only integral values , for example :
>          particle density in the given volume
>          But, distributions are the same .. - am I right ?
>
>  2) Are there any methods (ideas ?) how to change parameters of the generated
>     particles for C2 : E-values, angles,... to enlarge statistics
>     as the 1-st step for the following 2-nd step of simulation ?
>     ... or I have to to use only direct C2 filling (no BIAS)
>
>  Thank you,
>  Vladimir
>
>

__________________________________________________________________________
You can manage unsubscription from this mailing list at https://www.fluka.org/fluka.php?id=acc_info
Received on Mon Mar 16 2020 - 12:37:54 CET

This archive was generated by hypermail 2.3.0 : Mon Mar 16 2020 - 12:37:56 CET