ML14149A089

NRR E-mail Capture - STP-GSI-191 Presentation

ML14149A089
Person / Time
Site:	South Texas
Issue date:	05/21/2014
From:	Harrison A South Texas
To:	Balwant Singal Division of Operating Reactor Licensing
References
STP-GSI-191, TAC MF2400, TAC MF2401
Download: ML14149A089 (46)

Text

NRR-PMDA-ECapture Resource From: Harrison Albon <awharrison@STPEGS.COM>

Sent: Wednesday, May 21, 2014 11:03 AM To: Singal, Balwant Cc: Kee, Ernie; Blossom, Steven; Murray, Michael

Subject:

STP Sensitivity Presentation Attachments: STP-GSI-191 Presentation May 222014.pdf

Balwant, Our presentation is attached. Also, Alion confirmed the SWRI presentation has no proprietary information.

Wayne 1

CASA Grande Sensitivity Studies May 22,2014

Introductions

• Presenters Mike Murray Wayne Harrison Ernie Kee David Morton Bruce Letellier 2

Introductions

• Contributors Steve Blossom Zahra Mohaghegh Seyed Reihani Jeremy Tejada Don Wakefield 3

Summary

• Summarize approach to sensitivity analyses

• Demonstrate the approach on several parameters

• Review and explain counterintuitive results observed with certain parameters 4

Development Status and Potential Application o Development Status o NOT being proposed for NRC review/approval as part of the Risk-Informed Pilot Application o Will require appropriate validation and verification o Potential Applications o Facilitate evaluation of emergent plant conditions Facilitate identification/prioritization of compensatory actions Quantitative assessment may be required to formalize evaluation o Evaluation of plant modifications 5

Potential Application and Development Status, cont.

o Current and Potential Features (Ref. ML14072A211) o One-way sensitivity, tornado plots (Step 6) o One-way sensitivity studies, UQ plots (Step 7) o One-way sensitivity: Spider Plots (Step 8) o Two-way (two at a time) sensitivity studies (Step 9) o Meta models (Step 10) (response surface correlation)

To support multi-attribute sensitivity and simplified analysis of plant response o Compliance with standard software quality assurance o Importance o Helps show the relative importance of parameters o Helps to identify potential errors (counterintuitive results) and thereby contributes to analysis quality 6

CASA Grande Sensi-vity Studies

John Hasenbein David Morton Jeremy Tejada Alex Zolan May 22, 2014 7

Sensitivity Studies

• Approach

• Boron fuel limit

• Fiber penetration function

• Head loss studies 8

10-Step Sensitivity Analysis Process

• Step 1: Define the Model

• Step 2: Select Outputs of Interest

• Step 3: Select Inputs of Interest

• Step 4: Choose Nominal Values and Ranges for Inputs

• Step 5: Estimate Model Outputs under Nominal Input Values

• Step 6: One-Way Sensitivity Analysis: Sensitivity Plots & Tornado Diagrams

• Step 7: One-Way Sensitivity Analysis: UQ Plots

• Step 8: One-Way Sensitivity Analysis: Spider Plots

• Step 9: Two-way Sensitivity Analysis: Two-way Sensitivity Plots

• Step 10: Metamodels & Design of Experiments 9

Step 1: Define the Model

• We wont detail CASA Grande here (Volume 3)

• Use CASA Grande to estimate probability of sump failure and boron fiber limit failure, conditional on small, medium & large breaks

• Estimate change in core damage frequency (CDF) in events/year due to GSI-191 issues using these failure probability estimates and corresponding frequencies

• All results here are conditional on all pumps working 10

Step 2: Select Outputs of Interest

• Change in core damage frequency (CDF)

• Sometimes, we report ratio of CDF estimate for a scenario to CDF estimate for baseline and call this the risk ratio

• Use stratified sampling on initiating frequency

• Use IID replications within each cell of stratification

• Use common random numbers across scenarios; i.e., use CRNs across specified changes in input parameters 11

Step 2: Outputs: Estimating CDF Indices and Sets:

i = 1, . . . , F index for cells stratifying frequency replications k = 1, . . . , N index for set of pump states Events:

SL, M L, LL small, medium, large LOCA P Sk pumps in state k Fi initiating frequency in cell i S sump failure B boron fiber limit failure CD core damage Parameters:

fSL , fM L , fLL frequency (events/CY) of a small, medium, large LOCA P (P Sk ) probability mass of P Sk P (Fi ) probability mass of Fi P (SlLOCA, Fi , P Sk ) estimate of probability of S given LOCA = SL, M L, orLL, Fi , P Sk P (BlLOCA, Fi , P Sk ) estimate of probability of B given LOCA = SL, M L, orLL, Fi , P Sk RBASE non-GSI-191 core damage frequency (events/CY)

RCD estimate of core damage frequency (events/CY) 12

Step 2: Outputs: Estimating CDF CDF = RCD RBASE X F X N

= P (Fi )P (P Sk )

• i=1 k=1 h

fSL

• P (SlSL, Fi , P Sk ) + fSL

• P (BlSL, Fi , P Sk )

+fM L

• P (SlM L, Fi , P Sk ) + fM L

• P (BlM L, Fi , P Sk )

i

+fLL

• P (SlLL, Fi , P Sk ) + fLL

• P (BlLL, Fi , P Sk )

• We report results with:

- fSL , fML , fLL from Volume 2s Table 4-1

- P(all pumps working)=1

- P(Fi ): Bounded Johnson fit to NUREG-1829

• We form a variance estimate for the above estimator 13

Step 3: Select Inputs of Interest

• Amount of Latent Fiber in Pool: Existing dust/dirt in containment, based on plant measurement. Assumed to be in the pool at start of recirculation, uniformly mixed. During fill up, latent debris available to penetrate sump screen.

• Boron Fiber Limit: Refers to threshold where boron precipitation occurs for cold leg breaks. Fiber limit comes from vendor testing that shows no pressure drop occurs with full chemical effects. Assume all fiber that penetrates sump screen deposits uniformly on core.

• Debris Transport Fractions in ZOI: Refers to debris transport fractions involving three-zone ZOI. Each insulation type has characteristic ZOI divided in three sections to account for type of damage within each zone.

14

Step 3: Select Inputs of Interest

• Chemical Precipitation Temperature: CASA Grande assumes that, once a thin bed of fiber forms on strainer, chemical head loss factors apply when pool temperature reaches precipitation temperature.

• Total Failure Fraction for Debris Outside the ZOI: CASA Grande uses table of total failure fractions applied to transport logic trees. Fraction of each type (fiber, paint and coatings, etc.) that passes to the pool are used to understand what is in the pool as a function of time during recirculation.

• Chemical Head Loss Factor: Used as a multiplier on conventional head loss calculated in CASA Grande. Multiplier is applied if thin bed is formed and pool temperature is at or below precipitation temperature.

15

Step 3: Select Inputs of Interest

• Fiber Penetration Function: Fraction of fiber that bypasses the ECCS sump screen as a function of the amount of fiber on the screen.

• Size of ZOI: ZOI defined as direct function (multiplier) of break size and nominal pipe diameter; e.g., for NUKON fiber, ZOI is 17 times break diameter. ZOI is spherical unless break is not DEGB, in which case it is hemispherical. Truncated by any concrete walls within the ZOI.

• Time to Turn Off One Spray Pump: If three spray pumps start, then by procedure one is secured. Time to secure the pump is governed by operator acting on the conditional action step in procedure.

16

Step 3: Select Inputs of Interest

• Time to Hot Leg Injection: Similar to the spray pump turn off time, the time to switch one or more trains to hot leg injection operation is governed by procedure.

• Strainer Buckling Limit: Limit is the differential pressure across ECCS strainer at which strainer is assumed to fail mechanically. This limit is based on engineering calculations that incorporate safety factor.

• Water Volume in the Pool: Depending on break size, amount of water in pool, as opposed to amount in RCS and other areas in containment, varies. Smaller breaks tend to result in less pool volume than larger breaks.

17

Step 3: Select Inputs of Interest

• Debris Densities: Depends on amount and type of debris that arrives in pool. These densities are used in head loss correlations to calculate, for example, debris volume.

• Time Dependent Temperature Profiles: Temperature of water in sump affects air release and vaporization during recirculation. Time-dependent temperature profile comes from coupled RELAP5-3D and MELCOR simulations depending on break size.

• Spray Failure Fraction for Debris Outside ZOI: CASA Grande uses a table of failure fractions applied to transport logic trees. Fractions of each type of debris that passes to pool are used to understand what is in pool as function of time during recirculation. The spray failure fraction is fraction of failed coatings that wash to pool during spray operation.

18

Step 4: Nominal Values and Ranges for Inputs Input Parameter Level 1 Level 2 Level 3 Level 4

Latent Fiber (63) 12.5 6.25 25 50

Boron Fiber Limit (g/FA) 7.5 4 15 50

Debris Transport Inside ZOI Base Low High

Water Volume in Pool Base -10% +10%

Chemical Precipita-on Temp (oF) 140o 160o

Total Failure Frac-on Outside ZOI Base Low

Chemical Head Loss Factor Base +50%

Fiber Penetra-on Func-on Base High

ZOI Size Base -33%

Turn o 1 Spray Pump (min.) 20 1440

Hot Leg Injec-on (min.) 345 450

Strainer Buckling Limit (6. H2O) 9.35 9.6

Debris Density (lbm/63) Base +25%

Temperature Pro"les (oF) Base -5%

Spray Transport Frac-on 6% 12%

19

Step 5: Estimate Outputs Under Nominal Values of Inputs

1. Sensi7vity Measure Expected 95% CI 95% CI 95% CI CI HW  %

Mean CDF

Direc7on Half-Width Low Limit Upper Limit of Mean

0 Baseline None 1.817E-08 1.914E-09 1.626E-08 2.009E-08 10.53%

20

Step 6: One-Way Sensitivity Analysis

1. Expected 0.95-level CI HW  % of

Sensi7vity Measure Mean CDF

Direc7on Mean

0 Baseline N/A 1.817E-08 10.53%

1 Latent Fiber Low (6.25 6^3) Decrease 1.905E-08 10.12%

2 Latent Fiber High (25 6^3) Increase 1.669E-08 10.61%

3 Latent Fiber Very High (50 6^3) Increase 3.394E-08 42.63%

4 Boron Low (4.0 g/FA) Increase 1.690E-06 67.79%

5 Boron Very High (50 g/FA) Decrease 1.308E-08 10.80%

6 Boron High (15 g/FA) Decrease 1.329E-08 10.65%

7 Debris Transport Inside ZOI High Increase 7.896E-08 28.50%

8 Debris Transport Inside ZOI Low Decrease 1.241E-08 12.03%

9 Chemical Temp High Increase 1.905E-08 10.17%

10 Debris Transport Outside ZOI Low Decrease 1.770E-08 10.61%

11 Chemical Head Loss Factors High Increase 2.287E-08 8.85%

12 Penetra-on: Low Envelope of Filtra-on Func-on Increase 1.552E-07 10.93%

13 ZOI Size Small Decrease 6.795E-09 12.18%

14 Turn O 1 Spray Longer Decrease 1.569E-08 11.23%

15 Hot Leg Injec-on Longer Increase 1.962E-08 9.96%

16 Strainer Limit Higher Decrease 1.639E-08 10.99%

17 Water Volume Low Increase 2.001E-08 10.13%

18 Water Volume High Decrease 1.655E-08 10.73%

19 Debris Density High Increase 2.567E-08 9.17%

20 Temperature Pro"les Low Increase 1.963E-08 10.14%

21 Debris Transport Outside ZOI High Increase 1.798E-08 10.65%

21

Step 6: One-Way Sensitivity Analysis Tornado Diagram: Total CDF

Ra-o of Risk under the Scenarios to Risk under Nominal Parameter Values

Decreasing Risk Increasing Risk

0.10 1.00 10.00 100.00 1,000.00

Boron Fuel Limit (4.0 g/FA - 50 g/FA)

Penetra-on Low Envelope

Debris Transport Inside ZOI

ZOI Size Small

Scenario Descrip-ons

Latent Fiber (6.25 6^3 - 50 6^3)

Debris Density High

Decreased Parameter Values

Chemical Head Loss Factors High

Water Volume Increased Parameter Values

Turn O 1 Spray Longer

Strainer Limit Higher

Temperature Pro"les Low

Hot Leg Injec-on Longer

Chemical Temp High

Total Failure  % Outside ZOI Low (80%)

Spray Transport  % Outside ZOI High (12%)

22

Step 6: One-Way Sensitivity Analysis CDF (Total)

1000.0

Mean Risk

Baseline

100.0

Risk Ra-os

Increasing Risk

10.0

1.0

Decreasing Risk

0.1

3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0

Boron Fuel Limit (g/FA)

23

Step 6: One-Way Sensitivity Analysis CDF (Vessel)

1000.0

Mean Risk

100.0 Baseline

Risk Ra-os

Decreasing Risk Increasing Risk

10.0

1.0

0.1

3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0

Boron Fuel Limit (g/FA)

24

Step 6: One-Way Sensitivity Analysis CDF (Sump)

10.0

Mean Risk

Baseline

Decreasing Risk Increasing Risk

Risk Ra-os 1.0

0.1

3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0

Boron Fuel Limit (g/FA)

25

Step 6: One-Way Sensitivity Analysis Filtration Function Envelope 1$

0.95$

0.9$

0.85$

Filtra'on*

Test$1:$353gpm$

0.8$

Test$2:$353gpm$

0.75$ Test$3:$353gpm$

Test$5:$358gpm$

0.7$

Test$7:$220gpm$

0.65$ Fit$

Upper$Envelope$

0.6$

Lower$Envelope$

0.55$

0$ 500$ 1000$ 1500$ 2000$ 2500$ 3000$ 3500$ 4000$

Strainer*Mass*in*Grams*

26

Step 6: One-Way Sensitivity Analysis CDF (Total)

100.0

Mean Risk

Increasing Risk

10.0

Risk Ra-os

1.0

Decreasing Risk

0.1

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Fiber Envelope (0-Lower, 1-Upper)

27

Step 6: One-Way Sensitivity Analysis CDF (Vessel)

100.0

Mean Risk

Baseline

10.0

Risk Ra-os

Increasing Risk

1.0

Decreasing Risk 0.1

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Fiber Envelope (0-Lower, 1-Upper)

28

Step 6: One-Way Sensitivity Analysis CDF (Sump)

10.0

Mean Risk

Baseline

Risk Ra-os

Increasing Risk

1.0

Decreasing Risk 0.1

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Fiber Envelope (0-Lower, 1-Upper)

29

Step 6: One-Way Sensitivity Analysis

• Alternative distributions used for chemical head loss factor:

1. No chemical head loss

2. Factors constant at 1x mean (S=2.25/M=2.50/L=3.00)

3. Factors constant at 2x mean (S=3.50/M=4.00/L=5.00)

4. Factors constant at 3x mean (S=4.75/M=5.50/L=7.00)

5. Truncated exponential (S=6/M=3.5/L=2.25) at tail probability of 1E-5

6. Truncated exponential (S=3/M=2.5/L=2.25) at tail probability of 1E-5

7. Truncated normal (Mean, St Dev = 3x Mean)

Means have base case values of (S=2.25/M=2.50/L=3.00) 30

Step 6: One-Way Sensitivity Analysis Total CDF

2.0

Min Temp 140

1.8

No Minimum Temp

1.6

Increasing Risk

Ra-o of Total CDF to Base Case

Baseline

1.4

1.2

1.0

Decreasing Risk

0.8

0.6

0.4

0.2

0.0

Distribu-on of Chemical Head Loss Factor

31

Step 9: Two-Way Sensitivity Analysis CDF (Total) as a Func7on of Fiber Penetra7on Limit and Filtra7on Func7on

0 - 1.E-08

1.E-09-1.E-08

1.E-08-1.E-07

1.E-04

1.E-07-1.E-06

1.E-05

1.E-06-1.E-05

CDF

1.E-05-1.E-04

1.E-06

1.E-07

0.000

Filtra7on Func7on

1.E-08 0.167

0.333

1.E-09

4.0 0.500

5.0

6.0

7.0 1.000

8.5

Fiber Penetra7on Limit (g/FA)

32

Step 9: Two-Way Sensitivity Analysis CDF (Total) as a Func7on of Fiber Penetra7on Limit and Filtra7on Func7on

0.000

0 - 1.E-08

1.E-09-1.E-08

1.E-08-1.E-07

CDF

0.167

1.E-07-1.E-06

Filtra7on Func7on

1.E-06-1.E-05

0.333

1.E-05-1.E-04

0.500

1.000

4.0 5.0 6.0 7.0 8.5

Fiber Penetra7on Limit (g/FA)

33

Step 9: Two-Way Sensitivity Analysis CDF (Vessel) as a Func7on of Fiber Penetra7on Limit and Filtra7on Func7on

0.000

0 - 1.E-08

1.E-09-1.E-08

1.E-08-1.E-07

CDF 0.167

1.E-07-1.E-06

Filtra7on Func7on

1.E-06-1.E-05

1.E-05-1.E-04 0.333

0.500

1.000

4.0 5.0 6.0 7.0 8.5

Fiber Penetra7on Limit (g/FA)

34

Explaining Subtle Modeling Trends Bruce Letellier - Alion Science Jeremy Tejada - University of Texas Austin 35

Nonintuitive Trends

• Extensive usage of CASA Grande for parameter studies reveals four (4) subtle nonintuitive trends in quantitative risk

• Two issues were opened as Error Reports. Two issues were investigated during parameter study. All dispositioned using case study analysis:

- ER01 - Unqualified Coatings Spray Fraction

• Increasing the fraction of UC transport under spray leads to slight reduction in risk

- ER03 - Fiber Inventory Mass Conservation

• Over time, fiber mass increases slightly

- PS01 - Latent Fiber Effect at T0

• Increasing latent fiber slightly decreases risk

- PS02 - TimeStep Effect

• Decreasing time step can significantly decrease risk 36

ER01 - Unqualified Coatings Spray Fraction

• Parameter studies of transport fractions indicate that increasing the 6%

spray washdown fraction to 12% for failed epoxy slightly reduces risk (1% Reduction).

• Spreadsheet calculations for assumed inventory of failed coatings shows very definite reduction in SV for increased spray fraction from 6%

to 25%.

• Holds for both linear mass and volume weighting and for quadratic volume weighting. Simply competition between particulate properties.

• SV is the amount of drag area per unit solid debris volume. SV is independent of porosity. More debris does not imply higher average SV.

• A proper formalism would use total surface area rather than average surfacetovolume ratio so that more of anything always adds drag.

37

ER01 - Unqualified Coatings Spray Fraction (example)

Spherical Material Initial Initial Final Final Diam Density Mass (m1) Mass (m1)

(kg/m3) (kg) (kg) 1 10 1490 100 100 2 150 1986 25 30 SV 512,000 1581

• Surfacetovolume ratio can decrease when the proportion of larger particles increases (volume increases faster than area)

• True at STP for spray fractions because large inventory of 10m enamel has zero spray fraction (other particulates increase in proportion and SV decreases) 38

ER03 - Fiber Inventory Mass Conservation

• Parameter studies and code verification exercises show that fiber inventory increases slightly over the 36h calculation

• Explicit timeforward integration uses leading concentrations for each time step.

Coarse resolution delivers artificial mass at each time step that accumulates above initial inventory.

• Smaller time steps reduce this effect.

39

PS01 - Latent Fiber Effect at T0

• Parameter studies of latent fiber quantity show that increasing latent fiber slightly decreases risk.

• Fiber filtration and shedding model accounts for improved filtration with increasing debris load.

• More latent fiber initializes slightly higher filtration at beginning of recirculation

• User option added to allow some fraction of latent fiber to pass through the strainer.

40

PS01 - Latent Fiber Effect at T0 (cont)

Sensitivity Plot: Common Random Numbers 3.5E08 Mean Risk 3.0E08 Total Core Damage Frequency Increasing Risk 2.5E08 2.0E08 Decreasing Risk 1.5E08 1.0E08 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 Latent Fiber (ft3) 41

PS02 - Time-Step Effect

• Reduction of time step from 5 min to 1 min found to reduce quantitative risk by a factor of 1.5.

• Explict timeforward integration introduces numerical diffusion that advances debris to the strainer artificially rapidly.

• NPSH related failures occur before reduction in temperature regains margin.

• Possible nonconservatism may be introduced by improved filtration on the debris bed that protects core.

No numerical evidence that this dominates.

42

PS02 - TimeStep Effect Trend 43

Summary:

• The four nonintuitive observations discussed here have been explained at a fundamental level

• Changes in numerical approximations and physical descriptions can eliminate undesired behavior

• Subtle trends revealed in model interactions

• Essential to confirming or correcting engineering intuition

• Sensitivity studies create essential QA opportunities to exercise physical models over full parameter ranges

- Identify input errors

- Identify code level errors 44

Conclusions

• Summarized approach to sensitivity analyses

• Demonstrated the approach on several parameters

• Reviewed and explained counterintuitive results observed with certain parameters 45