Re: [fluka-discuss]: Resampling using pre-defined phase space data

From: Stefan E. Mueller <stefan.mueller_at_hzdr.de>
Date: Fri, 10 Jun 2022 11:13:28 +0200 (CEST)

Dear Arpad,

I am not sure if I understand the question. The way you set up your
source.f-file, FLUKA will randomly extract data points from your phase
space, and propagate the thus retrieved primary particle through your
geometry. Unless your geometry is very simple (e.g. there is only
vacuum, which would be anyway not very useful), each
interaction will add further randomness to the event. So even if a
following event happens to retrieve the same information from the phase
space data points, the event will look different, because FLUKA will use
different random numbers than for the previous event.

Therefore you can run your FLUKA input for more than 6 million events, the
results will be still (enough) statistically independent to combine the
results in a meaningful way (for example if you process several runs to
get the statistical uncertainty with the usual post-processing programs or
with FLAIR).

Hope this information helps,

         Stefan

--
Stefan E. Mueller
Department of Information Services and Computing - Computational Science
and Institute of Radiation Physics
Helmholtz-Zentrum Dresden-Rossendorf
Tel: +49 (0351) 260 3847
Stefan.Mueller_at_hzdr.de
http://www.hzdr.de
Vorstand: Prof. Dr. Sebastian M. Schmidt, Dr. Diana Stiller
Vereinsregister: VR 1693 beim Amtsgericht Dresden
On Thu, 2 Jun 2022, Arpad Lenart wrote:
> 
> Dear FLUKA team,
> 
>  
> 
> I have successfully imported 6D phase space data of electrons (x, y, z, x
> momentum, y momentum, energy, weights) using the source.f, currently I have
> 6 million primary particles. I would like to ask, how do I tell the source.f
> file to sample from my phase space data (which has 6 million data points) to
> get better statistics? I have attached to this email my current source.f
> file. How would the monte carlo re-sampling be carried out in such a case,
> so that my simulation errors decrease and I get better statistical behavior?
> I understand the FLUKA user manual includes this, but I am having trouble
> knowing how to actually modify the file for this purpose.
> 
>  
> 
> For example, the source.f script would read all the 6 million lines of 6D
> phase space data, I suppose I could split this into 6 separate cycles each
> with 1 million primary particles, and combine the final results from all 6
> cycles, but how would I tell FLUKA to do this? I am using FLUKA Flair (with
> user interface) and I don’t fully understand how this resampling is meant to
> be done ..
> 
>  
> 
> Thank you for your help,
> 
> Best regards,
> 
> Arpad
> 
> 
>

__________________________________________________________________________
You can manage unsubscription from this mailing list at https://www.fluka.org/fluka.php?id=acc_info

Received on Fri Jun 10 2022 - 12:34:22 CEST