Post-Test Calculation and Uncertainty Analysis of the Experiment - - PowerPoint PPT Presentation

▶
post test calculation and uncertainty analysis of the
SMART_READER_LITE
LIVE PREVIEW

Post-Test Calculation and Uncertainty Analysis of the Experiment - - PowerPoint PPT Presentation

Apr 10, 2024 380 likes •643 views

Post-Test Calculation and Uncertainty Analysis of the Experiment QUENCH-07 with the System Code ATHLET-CD H. Austregesilo, Ch. Bals, K.Trambauer Gesellschaft fr Anlagen- und Reaktorsicherheit (GRS), Germany Workshop on Evaluation of

slide-1
SLIDE 1

Post-Test Calculation and Uncertainty Analysis of the Experiment QUENCH-07 with the System Code ATHLET-CD

  • H. Austregesilo, Ch. Bals, K.Trambauer

Gesellschaft für Anlagen- und Reaktorsicherheit (GRS), Germany

Workshop on Evaluation of Uncertainties in Relation to Severe Accidents and Level 2 Probabilistic Safety Analysis Aix-en-Provence, November 7-9, 2005

slide-2
SLIDE 2

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 2

Topics

Introduction:

–

Short description of the code ATHLET-CD

–

Short description of the QUENCH facility and of test QUENCH-07

Post-Test Calculation of QUENCH-07

–

Input data

–

Main results of reference calculation

Sensitivity analysis

–

Methodology

–

Main results

Conclusions

slide-3
SLIDE 3

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 3

ATHLET-CD

Analysis of Thermal-Hydraulics of Leaks and Transients with Core Degradation

Mechanistic code for beyond design basis and severe accidents Assessment of accident management systems and procedures Simulation of processes in primary and secondary coolant systems:

–

Loss of coolant, core heat-up, degradation, melting, relocation

–

Release and transport of fission products and aerosols

–

Mechanical and thermal loads of reactor pressure vessel

Calculation of source terms for containment analyses Status of development: core damage before gross relocation

slide-4
SLIDE 4

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 4

ATHLET-CD: Main Modules (I)

RCS Thermal-Hydraulics (ATHLET):

–

two-fluid modelling

–

additional balance equations for non-condensable gases

–

  • ne-dimensional heat conduction within structures

–

control simulation module for the description of control, protection and balance-of-plant systems

Core Degradation (ECORE):

–

simulation of fuel and control rods, as well as BWR core structures

–

mechanical fuel rod behaviour, including thermal expansion, ballooning and cladding rupture

–

cladding oxidation (parabolic rate equations)

–

melting and relocation of cladding Zircaloy, absorber rods and guide tubes, fuel and oxidized cladding

slide-5
SLIDE 5

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 5

ATHLET-CD: Main Modules (II)

Fission Products and Aerosol Release (FIPREM):

–

fission products release and diffusion inside the grain

–

up to 24 elements or release groups considered

Fission Product and Aerosol Transport (SOPHAEROS):

–

deposition, transport and agglomeration processes

–

gas phase chemistry

Heat-up and Melting within Debris Bed (MESOCO):

–

2D balance equations with 3 components: particles, melt and gases

slide-6
SLIDE 6

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 6

QUENCH test section and bundle cross section (FZKA 6412)

slide-7
SLIDE 7

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 7

Main Features of Test QUENCH-07

First experiment with a boron carbide absorber rod in the bundle Main objectives:

–

Determination of the impact of a B4C absorber rod on a pre-oxidized LWR fuel rod bundle at high temperatures: Impact of absorber rod failure on fuel rod degradation Impact of absorber rod on bundle behaviour during cooldown

–

Information on gas generation after failure of absorber rod cladding and guide tube due to oxidation of B4C , in particular additional H2 generation and release of CO, CO2, CH4.

–

Additional heat-up due to the oxidation of B4C and its contribution to the temperature escalation of fuel rods

–

Information on the B4C - stainless steel - Zry interactions

slide-8
SLIDE 8

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 8

Boundary conditions for QUENCH-07

ATHLET−CD Post−Test Calculation of QUENCH−07 500 1000 1500 2000 2500 3000 3500 4000 4500 Time (s) 5 10 15 20 Power (kW), Mass Flow Rate (g/s)

Electric power steam flow

Heat−up Oxidation Cooldown Transient

slide-9
SLIDE 9

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 9

Nodalization of QUENCH test section for ATHLET-CD

  • 0.5
  • 0.4
  • 0.3
  • 0.2
  • 0.1

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 BUNDLE BYPASS CROSSFLOW JACKETIN JACKETTUBE JACKETOUT JACKETAR TOPJACIN TOPJACTUBE TOPJACOUT TOPJACH2O OUTERLP SHROUD SHRTOP TOPHS OUTERWALL OUTERTOP1 OUTERTOP2 GRID1 GRID2 GRID3 GRID4 GRID5 ROD2 ROD3

slide-10
SLIDE 10

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 10

Input Data for ATHLET-CD

Basis: standard data sets used for the calculation of previous QUENCH

experiments, specially for QUENCH-06 (ISP-45)

Modelling options as recommended in the code User’s Manual, except

calculation of Zr oxidation at temperatures above 1773K (correlation of Prater-Courtright instead of Urbanic-Heidrick)

Steam/argon inlet temperatures:

Temperatures measured by thermocouple T511 minus 100K

External resistance per heated rod: 4 mΩ

slide-11
SLIDE 11

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 11

Calculated and measured axial temperature profiles

ATHLET−CD Post−Test Calculation of QUENCH−07 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 Elevation (m) 600 800 1000 1200 1400 1600 1800 Temperature (K) Rod2 calc. TFS2 exp. Rod3 calc. TFS5 exp. SHROUD calc. TSH exp. ATHLET−CD Post−Test Calculation of QUENCH−07 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 Elevation (m) 600 800 1000 1200 1400 1600 1800 Temperature (K) Rod2 calc. TFS2 exp. Rod3 calc. TFS5 exp. SHROUD calc. TSH exp.

Oxidation phase (t = 2700 s) Start of transient phase (t = 3150 s)

slide-12
SLIDE 12

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 12

Test section temperatures at elevation 950 mm

code reproduces satisfactorily thermal behaviour of test bundle clad temperatures at the end of first heat-up phase are underestimated good agreement with respect to the start of temperature escalation due to oxidation code overestimates shroud temperature excursion due to

  • xidation at the final heat-up phase

shroud failure was not reproduced by the code

ATHLET−CD Post−Test Calculation of QUENCH−07 500 1000 1500 2000 2500 3000 3500 4000 4500 Time (s) 500 1000 1500 2000 2500 Temperature (K)

ROD2, calc. TFS 2/13, 0.95m, exp. ROD3, calc. TFS 5/13, 0.95m, exp. TIT A/13, 0.95m, exp. SHROUD TT/2 Node 13, cal. TSH 13/90, 0.95m, exp.

slide-13
SLIDE 13

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 13

Integral H2 production

good agreement up to start of

  • xidation escalation

code strongly underestimates H2 production during quench calculation did not consider:

  • xidation of external shroud surface
  • xidation of Mo-electrodes

(about 22 g H2)

  • possible cracking of oxide layer

due to thermal shock total amount of H2 generation:

  • calc. : 92 g
  • exp. : 177 g

(130 g for simulated components) H2 mass due to CR oxidation: calc.: 8.7 g exp.: 8 g

ATHLET−CD Post−Test Calculation of QUENCH−07 500 1000 1500 2000 2500 3000 3500 4000 4500 Time (s) 50 100 150 200 Mass (g)

  • acc. H2 out calc.

H2−rods calc. H2−CR (Zr+B4C) calc. intl.H2 exp.

slide-14
SLIDE 14

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 14

Combined Sensitivity Analysis

Application of some features of the GRS methodology for

code uncertainty analyses:

–

Simultaneous variation of uncertain parameters

–

Statistical evaluation of the code results

–

Application of the Wilks formula: number of code calculations independent of number

  • f input and output parameters

number of code calculations only depends only on the desired statistical tolerance limits (e.g. for a two-sided statistical tolerance, a minimum of 93 calculations are required for a 95% probability and a 95% confidence level)

Support software SUSA:

–

input of uncertain parameters and probability distributions

–

automatic generation of ATHLET-CD input data sets (total: 100)

–

statistical evaluation of code results

slide-15
SLIDE 15

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 15

Input parameters for sensitivity analysis (I)

index parameter input data unit

distribution

min. value max. value reference calculation discrete options

1 External heater rod resistance WHRES0 mΩ uniform 3.6 4.4 4.0 2 Deviation of steam inlet temperature to T511 DELTA K uniform

  • 80
  • 120
  • 100

3 Factor for conductivity of shroud insulation FSUSA1

  • uniform

0.8 1.2 1.0 4 Factor for heat capacity of shroud insulation FSUSA2

  • uniform

0.8 1.2 1.0 5 Factor for conductivity of heater rods FSUSA3

  • uniform

0.8 1.2 1.0 6 Factor for heat capacity of heater rods FSUSA4

  • uniform

0.8 1.2 1.0 7 Factor for argon conductivity (top of shroud) FSUSA5

  • uniform

0.5 2.0 1.0 8 Nodalization at top of bundle PAR08

  • discrete

1 3 1

1: node length 10 cm 2: node length 5cm 3: node length 3.3cm

9 Correlation for Zr oxidation IOXMOD

  • discrete

15 19 15 15: Cathcart + Prater/Courtright 16: Cathcart + Urbanic/Heidrick 19:Leistikov + Prater/Courtright 10 Correlation for B4C –SS interaction IB4CSS

  • discrete

3 7 7 3:JAERI 5:Belovsky 7:Nagase

slide-16
SLIDE 16

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 16

Input parameters for sensitivity analysis (II)

index parameter input data unit

distribution

min. value max. value reference calculation discrete options

11 Correlation for B4C oxidation ICRB4C

  • discrete

1 7 7 1:mod Steinbrück 5:Steinbrück 7:VERDI/BOX data 12 Factor for HTC due to convection to vapour FSUSA6

  • uniform

0.8 1.2 1.0 13 HTC to argon (jacket tube) PAR13 W/m2K uniform 10 20 15 14 HTC to air (containment) PAR14 W/m2K uniform 5 20 10 15 Limit of steam availability for oxidation red. OXXLIM

  • log. uniform

0.1 0.5 0.1 16 Limitation of protective oxide layer HROXLM mm uniform 0.1 0.3 0.2 17 Fraction of bundle area assigned to bypass flow PAR17

  • uniform

0.01 0.1 0.05 18 Gap conductivity in the heated rods HTCGAP W/m2K uniform 500 1500 500 19 Emissivity of heated rods EPS

  • normal

σ = 0.1 0.6 1.0 0.8 20 Factor for surface area due to porosity FAREA

  • uniform

1.0 3.0 2.0 21 Melt temperature of absorber guide tube CRTAM K uniform 1423 1523 1473

slide-17
SLIDE 17

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 17

Uncertainty and Sensitivity Analysis of Test QUENCH 07 with ATHLET-CD 1.1L Two-sided tolerance limits Sample Size = 100, BETA = 0.95, GAMMA = 0.95

400 800 1200 1600 2000 2400 2800 500 1000 1500 2000 2500 3000 3500 4000 4500 Time (s) Cladding temperature outer ring at 950 mm (K) calc min calc max TIT A/13 exp calc medians

Cladding temperature of outer ring at elevation 950 mm

slide-18
SLIDE 18

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 18

Total generated H2 mass

Uncertainty and Sensitivity Analysis of Test QUENCH 07 with ATHLET-CD 1.1L Two-sided tolerance limits Sample Size = 100, BETA = 0.95, GAMMA = 0.95

0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 500 1000 1500 2000 2500 3000 3500 4000 4500 Time (s) Total generated H2 mass (kg) calc min calc max H2 exp calc medians 130 g

slide-19
SLIDE 19

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 19

Sensitivity Coefficients

  • 1
  • 0.8
  • 0.6
  • 0.4
  • 0.2

0.2 0.4 0.6 0.8 1 2 4 6 8 10 12 14 16 18 20 22 24 Index of Parameter T im e p o in t o f t e m p e r a tu r e e s c a la t io n a t 9 5 0 m m

  • 1
  • 0.8
  • 0.6
  • 0.4
  • 0.2

0.2 0.4 0.6 0.8 1 2 4 6 8 10 12 14 16 18 20 22 24 Index of Parameter T o t a l g e n e r a t e d H 2 m a s s

Total generated H2 mass Time point of temperature escalation

slide-20
SLIDE 20

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 20

Sensitivity coefficients of main uncertainty parameters

Sensitivity coefficients Index Description Range of variation Temperature escalation Total H2 mass H2 due to B4C 1 electrical resistance ± 10% 0.436

  • 0.322
  • 0.324

3 conductivity of shroud insulation ± 20% 0.360

  • 0.304
  • 0.174

5 conductivity of fuel pellets ± 20% 0.324

  • 0.093
  • 0.049

7 Ar conductivity in top of shroud ± 50% 0.178

  • 0.261
  • 0.094

8 nodalization upper part of bundle 3.3-10 cm

  • 0.095

0.139 0.215 9 correlation for Zr oxidation 15,16,19 0.195 0.134 0.005 11 correlation for B4C oxidation 1,5,7 0.151

  • 0.051
  • 0.047

12 HTC convection to cooling gas ± 20% 0.488

  • 0.453
  • 0.547
slide-21
SLIDE 21

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 21

Main findings of the sensitivity analysis (I)

Main sources of code uncertainties:

–

convective heat transfer between cladding and steam-argon mixture

–

input value for the external electrical resistance (controls the actual power generated within rods)

–

input of material properties, mainly the thermal conductivity of test section components and cooling gases

Non-prototypic aspects of the test facility have a considerable

influence on code results

Small variations of input parameters can affect considerably the good

agreement between experiment and prediction obtained with the reference calculation

slide-22
SLIDE 22

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 22

Main findings of the sensitivity analysis (II)

Influence of chosen nodalization not so strong as expected a priori.

Finer nodalization in the upper part of test bundle leads to:

–

increased cladding temperatures

–

earlier start of temperature escalation

–

higher hydrogen production

Choice of correlations for Zr oxidation plays a secondary role on code

uncertainties

Main uncertainty parameters affect mostly the H2 generation before

  • cooldown. Their influence on H2 production during quench is

considerably reduced.

Modelling of B4C-SS interaction and B4C oxidation does not affect

significantly the calculated global thermal behaviour nor total H2 production

slide-23
SLIDE 23

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 23

Summary and Conclusions (I)

Post-test calculation of QUENCH-07 showed a good overall

agreement with experimental results concerning thermal-hydraulic behaviour of test bundle

Combined sensitivity analysis provided additional information about the

influence of several code input parameters and modelling options on the simulation of main phenomena observed experimentally

Results indicate that the main experimental measurements lay within

the uncertainty range of the calculated data, except for the increased hydrogen production after cooldown initiation: . Potential need for further modelling improvement

slide-24
SLIDE 24

7-9 November 2005 CSNI Workshop on Evaluation of Uncertainties - Aix-en-Provence 24

Summary and Conclusions (II)

Main contributors to the uncertainty of code results:

–

heat transfer coefficient due to forced convection to steam/argon

–

thermal conductivity of shroud insulation

–

input value for the external heater rod resistance

Uncertainties on modelling of B4C oxidation do not affect significantly

the total calculated H2 release rates

This study was a first step towards a more general application of the

GRS uncertainty methodology for the simulation of severe accidents

Further steps shall include similar analyses for in-pile experiments

(e.g. PHÉBUS) and for the TMI-2 accident, in order to extend:

–

the range of phenomena covered by the simulation

–

the database needed for the determination of uncertainty parameters