November 13 16 2017 FUMAC 3 rd RCM IAEA Vienna Christine STRUZIK Antoine BOULORÉ 8 novembre 2017 PAGE 1 CEA 10 AVRIL 2012 Outline Introduction Objectives Evolutions of ALCYONE for LOCA simulation ID: 794891
Download The PPT/PDF document "Simulation of LOCA tests using ALCYONE f..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Simulation of LOCA tests using ALCYONE fuel performance code
November 13 – 16, 2017
FUMAC – 3rd RCM, IAEA, ViennaChristine STRUZIK, Antoine BOULORÉ
8 novembre 2017
| PAGE 1
CEA | 10 AVRIL 2012
Slide2OutlineIntroductionObjectives
Evolutions of ALCYONE for LOCA simulationResultsIFA 650.10 calculationsUncertainty quantification / Sensitivity analysisStudsvik 192 test calculation
Conclusions/Comments8 novembre 2017FUMAC 3rd RCM, Vienna |
November 13- 16, 2017
| PAGE 2
Slide3IntroductionInitial objectives of CEAImprove the validation of the advanced fuel modelling in ALCYONE to contribute to the interpretation of integral experiment based on single rod LOCA tests
Extend the knowledge of LOCA fuel behaviourInitially agreed work-planModelling with ALCYONE 1D scheme of LOCA tests
IFA650.9 and IFA650.10US NRC Studsvik 192 and 198 testsUncertainty and Sensitivity AnalysisIncluding UQ/SA on the state of fuel at the end of base irradiationLOCA tests boundary conditions
8 novembre 2017
| PAGE 3FUMAC 3rd RCM, Vienna | November 13- 16, 2017
Slide4ALCYONE fuel performance codeALCYONE = Fuel performance code for PWR developed in the PLEIADES software development framework
8 novembre 2017
| PAGE 4
FUMAC 3
rd RCM, Vienna | November 13- 16, 2017
Slide5ALCYONE fuel performance codeMultidimensional approach (1D, 2D, 3D)Multiphysics
modelling [1]Thermal/mechanical behaviourMaterial modifications under irradiationFission products inventoryValidation up to 80
GWd/tM (UO2) and 65 GWd/tM (MOX)
8 novembre 2017
| PAGE 5
Time
Time t
Time t+
D
t
Neutronics
:
Swelling
/
cladding
corrosion
Thermal
-
hydraulics
Thermic
behaviour
Fission
products
behaviour
Mechanical
behaviour
Convergence
test
He
Gas
release
Internal
pressure
N axial slice
loop
, 1D
Simplified flowchart of
ALCYONE 1D
[1]
Struzik C.,
Marelle
V., “Validation of fuel performance CEA code ALCYONE, scheme 1D, on extensive database”.
Transactions TopFuel2012 conference
, Manchester (UK), Sept. 2012.
FUMAC 3
rd
RCM, Vienna |
November
13- 16, 2017
Slide6ALCYONE 1D scheme : evolutions for LOCA simulationFission gas release model : CARACAS
Specific developments for LOCA conditions : intergranular fracturationPossibility to get grain boundary fracturation on the action of stress induced by pressurized fission bubbles has been introduced in the modelling
If stress applied on the coherent grain boundary > resistance fracturation and intergranular gas releasePossibility
of fine fragmentation is not yet taken into account
8 novembre 2017| PAGE 6
FUMAC 3
rd
RCM, Vienna |
November
13- 16, 2017
Slide7ALCYONE 1D scheme : evolutions for LOCA simulationCladding behaviour lawIt is established from tests on zirconium alloy pipe at various temperature and internal pressure conditions : EDGAR Test
To compute correctly the strain using the same formalism as in the infinitesimal strain [3]Logarithmic formalism :
log=Ln(1+ total)Burst time is calculated by post processing strain criteriaFuel pellet behaviour lawFree central point better assessment of stress in the
center part of the pellet when not in compression
| PAGE 7[3] Helfer T., Nucl. Eng. Des. 288 (2015) 75-81FUMAC 3rd RCM, Vienna | November 13- 16, 2017
Slide8IFA 650.10 calculation8 novembre 2017
| PAGE 8
Calculated T
~
measure at TC levels
Central fuel temperature ~ external fuel T +30°C
Test calculation boundary conditions/ input data
Initial plenum pressure and volume
Temperature plenum measured during test
Fissile power
External cladding temperature history and axial profile : 2 hypothesis tested
(
T
cladding
–
T
sat
) proportional to the axial power distribution
SOCRAT
FUMAC 3
rd
RCM, Vienna |
November
13- 16, 2017
Slide9IFA 650.10 calculationRod pressure evolutionBoth BC sets delay the burst time
8 novembre 2017
| PAGE 9
FUMAC 3
rd RCM, Vienna | November 13- 16, 2017
Slide10IFA 650.10 calculationRod geometrical evolutionRather good agreement with experimental elongation
Cladding diameter evolution (exp. and calc. burst time)Correct tendencyLittle underestimation
8 novembre 2017| PAGE 10
With SOCRAT profile
Diameter underestimated at experimental burst time
FUMAC 3
rd
RCM, Vienna |
November
13- 16, 2017
Slide11IFA 650.10 calculationFuel fragmentation ?Fracturation in HBS zones and some central zones
Fragmentation calculated after burst (loss of pressure in the rod)Amount of released gas ~ 8 cm3 (only intergranular)Negligible effect on rod pressure evolution during test
Model still to be improved (calibration)Quantification of fragmentation after burstProvide input for a relocation model …8 novembre 2017
| PAGE 11
FUMAC 3rd RCM, Vienna | November 13- 16, 2017
Slide12UQ/SA for IFA650.10UQ/SA Tool : URANIE“Uncertainty and sensitivity analysis” platform developed by CEA/DEN (rel.
3.11.1)Based on the ROOT data analysis framework (CERN)Open source project downloadable at http://sourceforge.net/projects/uranie
Sampling, sequential/parallel code launching
Sensitivity analysis : regression, Sobol’, FAST, Morris, …
Modelling : surrogate models (ANN, Copulas, …)Optimization : used for automatic calibration of codes8 novembre 2017FUMAC 3rd RCM, Vienna | November 13-16, 2017| PAGE 12
Slide13UQ/SA for IFA650.10Some differences with the “reference” caseUncertainty on H content
irradiated Zy4 cladding law necessaryOnly available for 600 ppm H content (only 300 ppm H for IFA650.10 cladding)
Hydrogen content impacts phase change temperatureHydrogen effect on creep rate not explicitCreep rate probably over-estimatedPlenum temperature considered = coolant temperature
8 novembre 2017
FUMAC 3rd RCM, Vienna | November 13-16, 2016| PAGE 13At calculated burst time
Slide14UQ/SA for IFA650.10Parameters not considered in the UQ/SA
Test rod power profileClad to coolant Heat transfer coefficient (cladding temperature imposed)Fuel solid swelling modelFuel
gaseous swelling modelCladding oxidation model at high temperature (model not available in ALCYONE)Thermal conductivity of the oxide layer (oxide layer conductivity not taken into account in ALCYONE)Fission gas releaseGap gas conductivity (parameter not accessible in input file)Fuel radial relocation (not relevant)
Fuel fragment packing (not relevant)Cladding strain threshold for fuel mobility (not relevant)
Cladding Meyer hardness (not relevant)Calculation performed until cladding burst only (ALCYONE not valid after that yet)8 novembre 2017FUMAC 3rd RCM, Vienna | November 13-16, 2016| PAGE 14
Slide15UQ/SA for IFA650.10Results (200 simulations)
Calculated burst time range : 320-345 seconds
beginning of blow down at 110 s exp burst time ~350 sExp. burst time between results obtained with ‘Fresh cladding’ and ‘600 ppm H Zy4 cladding’
Uncertainty on Cladding outer temperature profile = global translationEffect on the height of the balloon and on the time of burst
Max cladding diameter calculated always over measurement8 novembre 2017FUMAC 3rd RCM, Vienna | November 13-16, 2016| PAGE 15
Slide16UQ/SA for IFA650.10Sensitivity analysis (Partial Rank Correlation Coefficient)Ex. Sensitive parameters on calculated burst time
Cladding temperature cladding creep rateFilling gas pressure ballooning
Coolant temperature plenum temperature pressure increaseInitial cladding geometry (OD and ID) cladding thickness
8 novembre 2017FUMAC 3
rd RCM, Vienna | November 13-16, 2017| PAGE 16Stress and Strain
Slide17USNRC Studsvik 192 test calculationFather rod
Behaviour law of Zy4 for cladding (instead of ZIRLO)Important thermal loadingFGR calculated = 6.4% (measure ~10%)Calculated HBS width (full or partial restructuring)
at least 500 µmCentral precipitation zone 2.5mm (only 1.5mm for IFA650.10)
Test 192 boundary conditionsOuter cladding axial temperature profile
imposedInitial inner pressure set upPlenum temperature imposedHypothesis : all parts of free volume are connectedCalculation of inner pressure evolution and cladding deformationCalculation of fuel temperature (radial distribution)8 novembre 2017| PAGE 17FUMAC 3
rd
RCM, Vienna |
November
13- 16, 2017
Slide18USNRC Studsvik 192 test calculationEvolution of temperature and pressure
Burst time delayed compared to experimental value (~20s)Behaviour law for Zy4 alloy, whereas cladding = ZIRLO~100°C temperature differenceCladding profile
Calculation of uniform strain (just before burst time) / measure = total strain (after burst)8 novembre 2017| PAGE
18
FUMAC 3rd RCM, Vienna | November 13- 16, 2017
Slide19USNRC Studsvik 192 test calculation
Fuel fragmentationIntergranular fragmentation calculation before burst only in HBS zoneIntergranular fragmentation occurs after burst (inner pressure decrease)Central zones
HBS zonesRelated to the amount of gas on GB at the end of base irradiation and the intergranular bubbles interconnexionThe fractured GB ratio cannot be used to determine fuel fragment sizeOnly gas release mechanism
8 novembre 2017| PAGE
19FUMAC 3rd RCM, Vienna | November 13- 16, 2017
Slide20ConclusionSpecific evolutions of ALCYONE for LOCA test calculation (1D scheme)
IFA650.10 experimentFather rod simulation OKLOCA test simulation
Results depend a lot on the BC set used in the simulationThis experiment cannot be used to discriminate between different formulation for cladding behaviour law for instanceUQ/SA : uncertainty on results very much dependent on uncertainty on boundary conditions (cladding outer temperature)Same problem for IFA650.9 not calculated in this project
Studsvik 192 testFather rod simulation
OKTest simulationCalculated delay for burst timeRelated to the behaviour law used for the cladding (Zy4 instead of ZIRLO)Fuel fragmentation occurs after burst in the calculation (pressure decrease)Studsvik 198 test : data for base irradiation not provided not simulated in this project8 novembre 2017| PAGE 20FUMAC 3
rd
RCM, Vienna |
November
13- 16, 2017
Slide21Recommendations and future workDetermine precise uncertainty on boundary conditions necessary to make these experiments discriminant
Uncertainty on cladding outer temperature uniform on the whole length of the rodlet
Maybe an uncertainty on the axial profile of cladding outer temperature better representative of the real uncertaintyCladding behaviour law available does not correspond to the cladding of the experiment (IFA650.10)
Future work
Fuel fragmentation relocation model integrated in ALCYONESimulation until the end of the experimentCladding oxidation during the experiment has to be modelled as wellCladding profile of IFA650.10 probably 3D effects in pellet-cladding interaction (sticking between pellet and cladding)Interesting to perform 3D simulationsStudsvik 198 : need for base irradiation history8 novembre 2017| PAGE 21
FUMAC 3
rd
RCM, Vienna |
November
13- 16, 2017
Slide22DEN DECSESC
Commissariat à l’énergie atomique et aux énergies alternativesCentre de Saclay
| 91191 Gif-sur-Yvette CedexT. +33 (0)1 XX XX XX XX | F. +33 (0)1 XX
XX XX XX
Etablissement public à caractère industriel et commercial | RCS Paris B 775 685 0198 novembre 2017
| PAGE
22
CEA | 10 AVRIL 2012