Decay heat - burnup step dependency

Questions and discussion about applications, input, output and general user topics
Post Reply
novako
Posts: 32
Joined: Sun Feb 15, 2015 1:01 pm
Security question 1: No
Security question 2: 92
Location: Prague, Czech Republic

Decay heat - burnup step dependency

Post by novako » Thu Aug 26, 2021 11:43 am

Hello,
I do observed suspicious behavior of TOT_DECAY_HEAT results. If burnsteps are significantly shortened after longer burnup period, decay heat jumps up. The reactor power remained the same, so no such significant changes in decayheat should be observed. See the enclosed figures (burnstep 20, cca 16 MWd/kg). Did someone experienced similar behavior? IS there an explanation / recommendation how to set the decay heat calculation?
The calculation was done for fuel pin in infinite lattice. Similar behavior was observed for fuel assembly in infinite lattice as well.
Thank you
Ondrej

Burn steps:

Code: Select all

dep butot
0.0001
0.09
0.26
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
6
7
8
9
10
13
16
16.01
16.02
16.04
16.5
16.8
18
Attachments
decheatproblem.png
decheatproblem.png (32.61 KiB) Viewed 241 times

Ana Jambrina
Posts: 661
Joined: Tue May 26, 2020 5:32 pm
Security question 1: No
Security question 2: 7

Re: Decay heat - burnup step dependency

Post by Ana Jambrina » Thu Aug 26, 2021 12:01 pm

Have you checked the statistics/convergence of the calculation? Could it be some iteration/instability of the coupled burnup/transport scheme? The burnup steps might be longer than they should (0.5-1 MWd/kgHM) at those stages. Then, setting much smaller steps might be popping up the underneath problem. (see the collection of inputs: Simple burnup examples).
- Ana

novako
Posts: 32
Joined: Sun Feb 15, 2015 1:01 pm
Security question 1: No
Security question 2: 92
Location: Prague, Czech Republic

Re: Decay heat - burnup step dependency

Post by novako » Thu Aug 26, 2021 3:09 pm

Hi,
thank you for advice, I will try to run the same calculation with shorter burnup steps. Regarding the convergence, the calculation had enough neutrons and cycles as it was set to set pop 40000 400 50.
Ondrej

novako
Posts: 32
Joined: Sun Feb 15, 2015 1:01 pm
Security question 1: No
Security question 2: 92
Location: Prague, Czech Republic

Re: Decay heat - burnup step dependency

Post by novako » Fri Aug 27, 2021 2:37 pm

Hi,
I did the the calculation with much shorter steps and results are better, but still small increase is there for the significantly shorter BU step. I do expect, that all steps ideally needs to be super short.No with the steps the jump is around 0,2 % which is ok, if the other inaccuracies are considered.
Thank you for help
Ondrej
Attachments
decheatproblem2.png
decheatproblem2.png (38.89 KiB) Viewed 212 times

Ville Valtavirta
Posts: 539
Joined: Fri Sep 07, 2012 1:43 pm
Security question 1: No
Security question 2: 92

Re: Decay heat - burnup step dependency

Post by Ville Valtavirta » Fri Aug 27, 2021 2:54 pm

Hi Ondrej,

the momentary decay heat estimation during full power operation has a significant component coming from very short lived nuclides, the concentrations of which are very sensitive to the potential time-discretization error in the depletion calculations. I would imagine that comparing decay heats even only a couple of seconds after reactor shutdown might be less sensitive.

The concentrations of short lived nuclides cannot be well predicted with predictor corrector methods unless very short timesteps are used. See e.g. the discussion in this paper (https://doi.org/10.1016/j.anucene.2011.04.022):
Short-lived nuclides are in secular equilibrium, and their concentrations should at all times be determined by the momentary reaction rates. In particular, the EOS concentrations should be determined by the EOS reaction rates. The cross-sections and flux used in the Bateman solutions are, however, averaged over the entire step and thus differ from the EOS values leading to errors for short-lived nuclides. The largest difference between average and EOS values is observed for the flux, which is constantly increasing to maintain the constant power normalization. This mechanism is not significant for long-lived nuclides as their behavior is dominated by the average, rather than momentary, reaction rates.
Luckily Serpent also includes substep methods for depletion (see https://doi.org/10.1016/j.anucene.2011.07.012) which should help you without needing further transport solutions (see set pcc in the input manual).

Compare the (momentary) decay heats obtained with your depletion stepping by the default settings (celi 1 1) and higher order methods (leli 10 10 and leqi 50 50) below:
H.png
H.png (11.69 KiB) Viewed 209 times
-Ville

User avatar
Jaakko Leppänen
Site Admin
Posts: 2446
Joined: Thu Mar 18, 2010 10:43 pm
Security question 2: 0
Location: Espoo, Finland
Contact:

Re: Decay heat - burnup step dependency

Post by Jaakko Leppänen » Mon Aug 30, 2021 8:38 am

The momentary results probably also depend on decay data library. JEFF data includes much more very short-lived isotopes compared to ENDF.
- Jaakko

Post Reply