general concepts

Questions and discussion about applications, input, output and general user topics
tarek
Posts: 42
Joined: Fri Oct 12, 2018 9:41 pm
Security question 1: No
Security question 2: 96

general concepts

Post by tarek » Mon Aug 23, 2021 5:54 pm

the questions that are still confused and not understood for me :

1- Can Detector entry be used in power calculations in both criticality
and burnup modes and what is the difference for each case?

2-the total power and power density which got from res. file during
criticality mode are accurate values in general?

3- If I did not know the power of the assembly because I have changed the
fissile mass and moderator volume, how could I set up the new power normalization
value for burnup, and if I could calculate it with the serpent by
detector e.g, det power dr -8 void dm fuel_3.7 without need for normalization as in the case of transport mode?

4-During the normalization process of power density, how does the code fix
the power density to the value which I set up in the input file, for
example, if the code calculates total fission energy and divided it by the
fissile mass and then evaluate the value of power density and found that it
does not converge with user-defined normalization value ?

Ana Jambrina
Posts: 660
Joined: Tue May 26, 2020 5:32 pm
Security question 1: No
Security question 2: 7

Re: general concepts

Post by Ana Jambrina » Mon Aug 23, 2021 10:24 pm

Detectors are set in the same way in both criticality and depletion calculations. Power/power density as total reaction rates either are fixed by the normalization or calculated based on it. So, yes, they are accurate if the calculation values are correctly set (e.g. volumes, etc.).

A new assembly definition implies a new evaluation to set up a new depletion history. Changing both fissile mass and moderator volumes means a complete new assembly. Understand which parameter you want to fix for the given geometry that is invariant (before/after the changes).

In neutron transport (e.g. critically), there is a normalization set by default to unit loss rate. Depletion depends on normalization. If you are interested in applying the same normalization as in the same neutron transport calculation, use: ‘set lossrate 1’. Analogously, Serpent will calculate the correspondence to power (or other total reaction rates).

The power tallied in the neutron transport part is based on estimating the fission heat release from different nuclides with:

power = flux*Sigma_fiss*H_fiss

The normalization coefficient is used as a multiplication factor for the neutron flux when the results are collected - integral values: which are normalised by the flux level (e.g. reaction rates).

norm * [flux*Sigma_fiss*H_fiss] / fmass = set_powerdens

The list of the different reaction rates can be checked within “Normalized total reaction rated (neutrons)” section in the ‘_res.m’ file.
- Ana

tarek
Posts: 42
Joined: Fri Oct 12, 2018 9:41 pm
Security question 1: No
Security question 2: 96

Re: general concepts

Post by tarek » Tue Aug 24, 2021 12:46 am

now it became more clear, thank you very much for these explanations

tarek
Posts: 42
Joined: Fri Oct 12, 2018 9:41 pm
Security question 1: No
Security question 2: 96

Re: general concepts

Post by tarek » Tue Aug 24, 2021 4:12 am

Ana Jambrina wrote:
Mon Aug 23, 2021 10:24 pm
Detectors are set in the same way in both criticality and depletion calculations. Power/power density as total reaction rates either are fixed by the normalization or calculated based on it. So, yes, they are accurate if the calculation values are correctly set (e.g. volumes, etc.).

A new assembly definition implies a new evaluation to set up a new depletion history. Changing both fissile mass and moderator volumes means a complete new assembly. Understand which parameter you want to fix for the given geometry that is invariant (before/after the changes).

In neutron transport (e.g. critically), there is a normalization set by default to unit loss rate. Depletion depends on normalization. If you are interested in applying the same normalization as in the same neutron transport calculation, use: ‘set lossrate 1’. Analogously, Serpent will calculate the correspondence to power (or other total reaction rates).

The power tallied in the neutron transport part is based on estimating the fission heat release from different nuclides with:

power = flux*Sigma_fiss*H_fiss

The normalization coefficient is used as a multiplication factor for the neutron flux when the results are collected - integral values: which are normalised by the flux level (e.g. reaction rates).

norm * [flux*Sigma_fiss*H_fiss] / fmass = set_powerdens

The list of the different reaction rates can be checked within “Normalized total reaction rated (neutrons)” section in the ‘_res.m’ file.

OK, as you mentioned above, the power is tallied in the neutron transport through formula; flux*Sigma_fiss*H_fiss, so I have checked the materials volume and masse according to out.file and no significant errors found, But in res.file, the TOT_POWER and TOT_POWDENS still have significant small values along with the very small value of TOT_FISSRATE which I think it came from an error in flux estimation during the criticality mode (TOT_FLUX has also very small value).
Last edited by tarek on Tue Aug 24, 2021 4:18 am, edited 1 time in total.

Ana Jambrina
Posts: 660
Joined: Tue May 26, 2020 5:32 pm
Security question 1: No
Security question 2: 7

Re: general concepts

Post by Ana Jambrina » Tue Aug 24, 2021 9:06 am

When you fix one total reaction rate via normalization (e.g., loss rate - default), the rest of them, e.g., total flux, power, fission rate, etc., are adjusted accordingly. (The absolute value of the estimates is computed, and even if it is very small, it has its physical meaning within the calculation). The normalization, as a multiplier, is applied several times over the calculation (every time that “physical”/“non-physical” detectors are collected).
Also, keep in mind that every estimate has its associated statistical error that always plays a role in the simulation.

Exactly, what are you doing? And what are those discrepancies that you are observing? What are the main parameters of the simulation? And the statistics?
- Ana

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

Re: general concepts

Post by Ville Valtavirta » Tue Aug 24, 2021 9:07 am

Hi,

you have to set the normalization for the simulation yourself using one of the options in Serpent. See Serpent manual section "5.8 Source rate normalization".

For pure neutronic calculations the choice of a power level in the system is arbitrary and, for that reason the normalization needs to be set by the user. (Default normalization was explained by Ana.)

-Ville

tarek
Posts: 42
Joined: Fri Oct 12, 2018 9:41 pm
Security question 1: No
Security question 2: 96

Re: general concepts

Post by tarek » Tue Aug 24, 2021 4:15 pm

Ana Jambrina wrote:
Tue Aug 24, 2021 9:06 am
When you fix one total reaction rate via normalization (e.g., loss rate - default), the rest of them, e.g., total flux, power, fission rate, etc., are adjusted accordingly. (The absolute value of the estimates is computed, and even if it is very small, it has its physical meaning within the calculation). The normalization, as a multiplier, is applied several times over the calculation (every time that “physical”/“non-physical” detectors are collected).
Also, keep in mind that every estimate has its associated statistical error that always plays a role in the simulation.

Exactly, what are you doing? And what are those discrepancies that you are observing? What are the main parameters of the simulation? And the statistics?
I am working on examining the softening and hardening of the neutron flux and its impact on other assembly parametres through varying the water to fuel volume ratio within the VVER fuel assembly through many different ways . e.g. ; in standard vver 1000 fuel assembly there are 312 FR but in my design i have decreased it to 264 FR and inroduced Zr rods instead .
so for the new design with new fuel and water volume i want to calculate thr power during transport mode to know the effects of changing Moderator fule ratio on the assembly power and to use this power farthur in burnup normalizatio . but during transport mode , i got very low value of total power corresponding with low values of power density , fission rate and flux , also i noted that in out.file the fuel mass has normal values in grams , but in res.file the fissile mass , fmass , has a very small value in grams .

tarek
Posts: 42
Joined: Fri Oct 12, 2018 9:41 pm
Security question 1: No
Security question 2: 96

Re: general concepts

Post by tarek » Tue Aug 24, 2021 4:20 pm

Ville Valtavirta wrote:
Tue Aug 24, 2021 9:07 am
Hi,

you have to set the normalization for the simulation yourself using one of the options in Serpent. See Serpent manual section "5.8 Source rate normalization".

For pure neutronic calculations the choice of a power level in the system is arbitrary and, for that reason the normalization needs to be set by the user. (Default normalization was explained by Ana.)

-Ville

ok understood but for any modifications could be happened in assembly fuel masses and moderator volume , is there a possibilty to evalute the new design parametres like power values which correspond to the new modifications without any need to be calculated analytically and to be used after that for normalization during depletion .

Ana Jambrina
Posts: 660
Joined: Tue May 26, 2020 5:32 pm
Security question 1: No
Security question 2: 7

Re: general concepts

Post by Ana Jambrina » Wed Aug 25, 2021 6:59 am

My question was referring to how you were setting the simulation. Providing a vague description of some output parameters does not give much information about the simulation.

In any case, what is the problem setting an invariant normalization? As Ville pointed out, the power level is a user-defined choice in a neutronic simulation. You fix one total reaction rate (normalization), setting the power level of the calculation, directly or indirectly, depending on the normalization choice and, from there, you can evaluate easily the neutronic effects that your changes have on the assembly.
- Ana

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

Re: general concepts

Post by Ville Valtavirta » Wed Aug 25, 2021 8:59 am

To put it even more simply,

there is no specific power level that corresponds to some assembly that can be solved with neutronics. If the assembly is critical at the power level of, say 1 kW, it is critical at the power level of 1 MW too when only considering neutronics.

-Ville

Post Reply