Consistent Code-to-Code Comparison of Pin-cell Depletion Benchmark Suite

Transcription

1 Consistent Code-to-Code Comparison of Pin-cell Depletion Benchmark Suite September 27, 26 Jinsu Park, Deokjung Lee * COmputational Reactor Physics & Experiment lab

2 Contents VERA depletion benchmark suite VERA 2D pin problem specifications Sensitivity study Burnup interval sensitivity Depletion intra-zone sensitivity Q-value sensitivity Comparison method Compare with a burnup Compare with a number density of isotope Compare with a number density of element Compare with an accumulative flux Conclusion 2

3 VERA depletion benchmark suite Consistent Code-to-Code Comparison of Pin-cell Depletion Benchmark Suite 3

4 VERA depletion benchmark suite When verifying the depletion calculation module of reactor analysis codes, the code-to-code comparisons of depletion benchmark calculation results can be one of the viable methods. A depletion benchmark suite has been developed based on The VERA core physics benchmark progression problems*. *AT. Godfrey, The VERA Core Physics Benchmark Progression Problem Specifications, CASL-U , Rev. 4, Oak Ridge National Laboratory, 24. 4

5 VERA depletion benchmark suite The VERA depletion benchmark problems include single fuel pin problems with various fuel temperatures, enrichment of 235 U, and burnable poisons. Problem Description Temperature (K) Moderator Cladding Fuel Moderator density (g/cm 3 ) 235 U w/o Power de nsity (w/gu) A Pin (3.w/o) B Pin (3.w/o) C Pin (3.w/o) D Pin (3.w/o) E Pin (IFBA) F Pin (2.w/o) G Pin (3.6w/o) H Pin (4.6w/o) I Pin (Gadolinia) J Pin (3.w/o) 6 6 6/9/

6 VERA depletion benchmark suite The VERA depletion benchmark suite provides a recommended burnup step and depletion intrazone for each problems. 4 burnup steps Step MWD MWD MWD MWD Step Step Step / kgu / kgu / kgu / kgu depletion zones for UO 2 pins and 5 zones for gadolinia pin. 6

7 VERA depletion benchmark suite Code-to-Code comparison Depletion calculation information. Code SERPENT2 MCODE (MCNP5 + ORIGEN2.2) MCNP6 MCS CASMO-5 STREAM McCARD ntracer MPACT Developer VTT MIT LANL UNIST Studsvik UNIST SNU SNU CASL Transport MC MC MC MC MOC MOC MC MOC MOC Energy group CE CE CE CE CE XS Library ENDF/B-VII. ENDF/B-VII. ENDF/B-VII. ENDF/B-VII. ENDF/B-VII. ENDF/B-VII. ENDF/B-VII. ENDF/B-VII. ENDF/B-VII. Kappa data SERPENT2 ORIGEN2.2 MCNP6 VERA CASMO-5 VERA ORIGEN2.2 ENDF/B-VII. + (nu-) 6.MeV for parasitic γ ENDF/B-VII. + Inhouse Decay library ENDF/B-VII ORIGEN2.2 (ENDF/B-VI) CINDER9 ORIGEN2.2 ENDF/B-VII + α ORIGEN2.2 ORIGEN2.2 ORIGEN ORIGEN-S (ENDF/B-VII.) Yield library ENDF/B-VII ORIGEN2.2 (ENDF/B-VI) CINDER9 ENDF/B-349 ENDF/B-VII + α ENDF/B-349 ENDF/B-349 ORIGEN ORIGEN-S (ENDF/B-VII.) Depletion solver CRAM MEM (ORIGEN2.2) MEM (CINDER9) CRAM CRAM CRAM MEM Krylov MEM (ORIGEN-S) Predictor-Corrector Full-PC Semi-PC Full-PC Full-PC Semi-PC Full-PC Semi-PC Semi-PC Full-PC Equilibrium Xe Possible Impossible Impossible Impossible Possible Impossible Impossible Possible Possible Critical spectrum Possible Impossible Impossible Impossible Possible Possible Impossible Possible Possible Gd treatment QD QD - QD Post correction 7

8 VERA depletion benchmark suite How can we trust this results? Pin C (4 steps, 3 zones), Reference: SERPENT2 MCODE MCNP6 MCS McCARD CASMO-4E CASMO-5 STREAM ntracer MPACT MWD/kgU 8

9 Sensitivity study Consistent Code-to-Code Comparison of Pin-cell Depletion Benchmark Suite 9

10 Sensitivity study Solver-dependent Burnup interval sensitivity CRAM and MEM Problem-dependent Depletion intra-zone sensitivity UO 2 and Gadolinia Data-dependent Q-value sensitivity Kappa from ORIGEN, SERPENT2, and VERA

11 Sensitivity study Burnup interval sensitivity Test was performed using SERPENT2 : CRAM MCNP6 : MEM w/o sub-step For the case: C, I. Monte Carlo histories, Pin: 2, history/cycle, 2 inactive cycle, 8 active cycle Depletion intra-zone, UO 2 : 3 zones Gadolinia: 5 zones Based on 4 burnup points, splitting intervals, 4 intervals 79 intervals 57 intervals 33 intervals

12 Sensitivity study Burnup interval sensitivity Difference of multiplication factor from 33 steps using SERPENT2 and MCNP6 for problem C Pin C : Difference of k inf with 33 steps MWD/kgU 4 steps 8 steps 57 steps Pin C : Difference of k inf with 33 steps 4 steps 79 steps 57 steps MWD/kgU SERPENT2 MCNP6 *For normal UO 2 pin problem, 4 burnup intervals is sufficient by using CRAM. However, it is insufficient by using MEM. 2

13 Sensitivity study Burnup interval sensitivity Difference of multiplication factor from 33 steps using SERPENT2 and MCNP6 for problem I Pin I : Difference of k inf with 33 steps MWD/kgU 4 steps 8 steps 57 steps Pin I : Difference of k inf with 33 steps MWD/kgU 57 steps SERPENT2 MCNP6 *For normal UO 2 pin problem, 57 burnup intervals is sufficient by using CRAM. However, it is insufficient by using MEM. 3

14 Sensitivity study Depletion intra-zone sensitivity Test was performed using SERPENT2 For the case: C, I Monte Carlo histories, Pin: 2, history/cycle, 2 inactive cycle, 8 active cycle Burnup intervals, C: 4 burnup intervals I: 57 burnup intervals Depletion intra-zone test cases, C: /3/5/ zones I: /5//5 zones 4

15 Sensitivity study Depletion intra-zone sensitivity of Problem C and I Pin C : Difference of k inf with zones zone 3 zones 5 zones Pin I : Difference of k inf with 5 zones zone 5 zones zones MWD/kgU Problem C MWD/kgU Problem I *For normal UO 2 pin problem, depletion intra-zone is sufficient. For normal gad pin problem, depletion intra-zones are required to obtain converged solution. 5

16 Sensitivity study Q-value sensitivity Test was performed using SERPENT2. For case, C and I. Q value from VERA, ORIGEN2.2, and SERPENT2. Isotope Q-value (MeV) Difference with VERA (%) VERA ORIGEN2.2 SERPENT2 ORIGEN2.2 SERPENT

17 Sensitivity study Q-value sensitivity SERPENT2 test results.2..9 Pin C Pin I Burnup (MWD/kgHM) Burnup (MWD/kgHM) *Q-value from VERA is larger than that from ORIGEN2.2 and SERPENT2, So, flux level can be lower and the fuel is burned slower. 7

18 Comparison method Consistent Code-to-Code Comparison of Pin-cell Depletion Benchmark Suite 8

19 Comparison method Compare with Burnup Code: SERPENT2 The k inf depletion calculation result with various kappa value (VERA, appa) by using SERPENT2. When the difference is compare in terms of burnup (x-axis), the difference from different kappa is remarkable. Pin C Pin I Burnup (MWd/kgHM) Burnup (MWd/kgHM) 9

20 Comparison method Compare with N.D. of 48 Nd Code: SERPENT2 When the k inf were comparing with number density of specific isotope ( 48 Nd), it was covered that the difference from different kappa value. Pin C - Pin I Number density of 48 Nd (#/barn-cm) x Number density of 48 Nd (#/barn-cm) - x -5 2

21 Comparison method Compare with Burnup Code: STREAM In order to clearly show the difference without statistical error, previous comparison procedure was performed again by using STREAM (UNIST in-house MOC code) with SERPENT, VERA, and appa value. Pin C 2 Pin I Burnup (MWD/kgHM) Burnup (MWD/kgHM) 2

22 Comparison method Compare with N.D. of 44 Nd Code: STREAM The number density of 44 Nd was used as x-axis instead of burnup for comparison. Then -4~8pcm difference of k inf were reduced to -2~3pcm. Pin C Pin I Number density of 44 Nd (#/barn-cm) x Number density of 44 Nd (#/barn-cm) x -5 22

23 Comparison method Compare with N.D. of 45 Nd Code: STREAM The number density of 45 Nd was used as x-axis instead of burnup for comparison. Then -4~8pcm difference of k inf were reduced to -2~pcm. Pin C Pin I Number density of 45 Nd (#/barn-cm) x Number density of 45 Nd (#/barn-cm) x -5 23

24 Comparison method Compare with N.D. of 48 Nd Code: STREAM The number density of 48 Nd was used as x-axis instead of burnup for comparison. Then -4~8pcm difference of k inf were reduced to -2~5pcm. Pin C Pin I Number density of 48 Nd (#/barn-cm) x Number density of 48 Nd (#/barn-cm) x -5 24

25 Comparison method Compare with Nd (neodymium) element Code: STREAM In order to compare the results by computer code with the experiment data, the number density of Nd element was used as x-axis instead of burnup for comparison. Then -4~8 pcm difference of k inf were reduced to.2 -~pcm. Pin C Pin I Number density of Target element (#/barn-cm) x Number density of Target element (#/barn-cm) x

26 Comparison method Best options of isotope and element for C and I problem. C I Isotope 48 Nd (2pcm) 66 Er (3pcm) Element Iodine (8pcm) Erbium (3pcm) Pin I *It means the maximum difference of k inf Burnup (MWD/kgHM) 26

27 Comparison method Compare with Er (erbium) element Code: STREAM Pin C 2 Pin I Number density of Target element (#/barn-cm) x Number density of Target element (#/barn-cm) x

28 Comparison method Compare with cumulative flux Code: STREAM The cumulative flux will be used as x-axis instead of burnup for comparison. Plot the cumulative flux along the depletion burnup steps. Cumulative Flux (#/cm 2 ) x 22 Pin C Difference (%) Cumulative Flux (#/cm 2 ) x 22 Pin I Difference (%) Burnup (MWD/kgHM) Burnup (MWD/kgHM) 28

29 Comparison method Depletion comparison using Cumulative flux Code: STREAM If the cumulative flux was used as the amount of depletion, then -4 ~8pcm difference of k inf were reduced to -2~2pcm. Pin I Cumulative Flux (#/cm 2 ) x 22 29

30 Conclusions Consistent Code-to-Code Comparison of Pin-cell Depletion Benchmark Suite 3

31 Conclusions VERA depletion benchmark suite was calculated using SERPENT2 with various set of kappa value. Optimum depletion calculation options are determined through the sensitivity study of the burnup interval and the number of depletion intrazones Difference of kappa value is main reason occurring difference of multiplication factor in code-to-code comparison. By using number density of specific isotope and element or accumulative flux as a criteria of comparison, the discrepancy of k inf is reduced remarkably in code-to-code comparison. 3

32 32