Hi,
I'v read a paper about statistical tests in Sepent 2 (http://montecarlo.vtt.fi/download/VTTR0037114.pdf), but is can't find how to use these options. It says "The statistical tests are performed automatically for the analog, implicit and collision
estimates of kef.." but there is no “<input>_stat.m” file generated by Serpent. Is there any way to get statistical test results?
I would like to find out how many inactice cycle needed for source convergence.
Thanks,
Bálint
Statistical test

 Posts: 11
 Joined: Sat Jun 27, 2015 3:31 pm
 Security question 1: No
 Security question 2: 92
Re: Statistical test
Hi,
You have to use the dhis option with a detector to active the tests. You can use a dummy detector like
if you only want to test the keff. Note that the tests are performed only on active cycles, and they might not help you for detecting source convergence, depending on your case.
You have to use the dhis option with a detector to active the tests. You can use a dummy detector like
Code: Select all
det 0 dhis 1
Re: Statistical test
Hi,
Thanks for your reply, i will try that.
I run the same script twice, only modifying parameters for group constant generation and i don't get the same keff within statistic error. I guess the problem is source convergence. Is there any way to test that?
Best wishes,
Bálint
Thanks for your reply, i will try that.
I run the same script twice, only modifying parameters for group constant generation and i don't get the same keff within statistic error. I guess the problem is source convergence. Is there any way to test that?
Best wishes,
Bálint
 Jaakko Leppänen
 Site Admin
 Posts: 2387
 Joined: Thu Mar 18, 2010 10:43 pm
 Security question 2: 0
 Location: Espoo, Finland
 Contact:
Re: Statistical test
Add
in the input file, and you'll get an output file <input>_his0.m which contains the fission source entropies in x, y and zdirections.
Code: Select all
set his 1
 Jaakko