ML20308A733

Slides for GSI-191 Pre-Submittal Public Teleconference November 18, 2020

ML20308A733
Person / Time
Site:	Wolf Creek
Issue date:	11/18/2020
From:	Samson Lee Plant Licensing Branch IV
To:	Wolf Creek
Lee S, NRR/DORL/LPL4, 301-415-3168
References
EPID L-2020-LRM-0099
Download: ML20308A733 (41)

Text

Wolf Creek GSI-191 Resolution Plan and Current Status NRC Public Meeting November 18, 2020

Agenda

• Meeting Objectives

• Overview of Plant Specific Features

• Overview of Threshold Break Size Methodology

• Determination of Threshold Break Size

- Strainer evaluation

- In-vessel downstream effects

• Risk and Uncertainty Quantification

• Adoption of TSTF-567

• Methodology for Operability Evaluation

• Submittal Format and Schedule 2

Meeting Objectives

• Communicate current Wolf Creek plan for GL 2004-02 response

• Obtain staff feedback on the overall resolution path for Wolf Creek

• Identify areas of concern from the NRC on the approach 3

Wolf Creek Plant Layout

• Westinghouse 4-loop PWR (3,565 MWt)

• Two redundant ECCS and CS trains

- Each train has an RHR pump, CCP, SIP, and CS pump

- SIP and CCP piggyback off of the RHR pump discharge during recirculation

• Two independent and redundant containment air cooling trains 4

Overview of Sump Strainers 5

Strainer B Strainer A

Overview of Threshold Break Size Methodology

• Licensees have used various risk-informed GSI-191 methods including RoverD, the conditional failure probability (CFP) approach, and the alternate break methodology

• Wolf Creek has chosen to use a different approach called the threshold break size methodology

• This approach is more conservative than RoverD and the CFP approach, but can be implemented in a simplified manner and does not require risk integration software (e.g.,

NARWHAL or CASA Grande) 6

Overview of Threshold Break Size Methodology

• Intermediate analyses required for overall GSI-191 evaluation (e.g., debris generation and transport) are generally consistent with deterministic and risk-informed methods previously reviewed and accepted by the NRC

• Strainer head loss and in-vessel effects evaluations identify largest break size with no failures for any weld locationsthis is the threshold break size

• All breaks larger than threshold break size are conservatively assumed to fail

• Threshold break size is based on bounding equipment configuration and is conservatively assumed to apply to all equipment configurations 7

Overview of Threshold Break Size Methodology

• Risk quantification is performed outside the PRA model

• CDF is calculated with a simple interpolation of NUREG-1829 LOCA frequencies at the threshold break size

• LERF is calculated based on the conditional large early release probability (CLERP) for a large LOCA given core damage

• CLERP is determined from the PRA model and the CLERP value is multiplied by CDF to calculate LERF

• The base CDF and LERF values are obtained from the PRA model for comparison with RG 1.174 acceptance guidelines 8

Debris Generation and Transport Analyses Overall approach for debris generation and transport similar to Vogtle BADGER used for debris generation evaluation Debris transport analyzed for blowdown, washdown, pool fill and recirculation CFD models used for recirculation transport 9

Treatment of Reactor Cavity Breaks

• No breaks at the reactor nozzles are postulated due to plant geometry per previous PWROG letter*

• Hot and cold legs are held by whip restraints that limit lateral movement of piping

• Maximum allowable lateral movement is less than pipe wall thickness

• RCP tie rods preclude cold leg separation from reactor nozzle

• Steam generator lower lateral supports preclude hot leg separation from reactor nozzle 10

• ADAMS Accession No. ML100710710 and ML100570364

Determination of Threshold Break Size

• Threshold break size defined such that breaks up to this threshold do not fail any GSI-191 criteria

- Strainer head loss Strainer structural limit Pump NPSH margin Strainer degasification and flashing

- In-vessel downstream effects (core blockage)

- Air entrainment due to vortexing

- Ex-vessel downstream effects

- Upstream effects

• Threshold break sizes for strainer head loss and in-vessel effects determined separately; the smaller of the two is the overall threshold break size 11

Strainer Evaluation

• Threshold break size for strainer evaluation determined by meeting following criteria

- Strainer head loss lower than strainer structural limit

- Minimum pump NPSH margin stays positive

- Void fraction at pump suction < 2%

- No flashing downstream of the strainer

- No air-entraining vortexing

• Strainer evaluation used the bounding equipment configuration with single train failure

- Maximizes strainer flow rate and debris load on the active strainer 12

Strainer Evaluation

• Head loss testing performed in 2016 at Alden 13

- Overall approach consistent with tank tests observed by NRC at Alden

- Performed one full debris load test and one thin-bed test

- Used two prototypical strainer stacks with no modifications

- Followed NEI guidance on fiber preparation

- Used pre-made AlOOH to represent chemical debris

- Bounded breaks up to 10 for debris loads and strainer approach velocity under single train operation Flow Direction Debris Introduction

& Mixing Section Plenum Box Test Strainer To Flow Loop

Strainer Evaluation

• Determined total strainer head loss for breaks up to 10

- Measured debris head losses adjusted to plant conditions (e.g., temperature and flow rate) using flow sweep data taken from testing

- Debris head loss combined with clean strainer head loss to determine total strainer head loss

• Demonstrated strainer evaluation acceptance criteria are met for breaks up to 10 14

In-Vessel Downstream Effects 15

• Threshold break size for in-vessel was determined based on HLB debris limit following NRC review guidance

• Performed a fiber-only penetration test in 2016 at Alden

- Removed every other disks and seismic cables to avoid bridging

- Used 5-µm filter bags to collect penetrated fiber

- Bounded breaks up to 10 for fiber load and strainer approach velocity

• Developed curve-fit from test data for fiber penetration as function of fiber loading on strainer Flow Direction

In-Vessel Downstream Effects

• Determined in-vessel fiber load using WCAP-17788 methodology

- Divided recirculation phase into smaller time steps

- Calculated debris arrival at sump strainers for each time step based on pool volume and pump flow rates

- Evaluated fiber penetration fractions based on strainer fiber load for each time step using curve-fit from testing

- Analyzed most limiting equipment configurations (both RHR pumps operating with failure of one or both CS pumps)

- Performed sensitivity to capture the worst combination of inputs (e.g., pool volume, RHR pump flow rate)

- Assumed all fiber that reaches reactor accumulate at core inlet with no credit of alternate flow paths (AFPs)

• Used Box 4 path from NRC review guidance to demonstrate applicability of WCAP-17788 AFP analysis to Wolf Creek for breaks up to 10 16

In-Vessel Downstream Effects 17 Parameters WCAP-17788 Revision 1 Values WCGS Values Nuclear Steam Supply System (NSSS) Design Various Westinghouse Fuel Type Various Westinghouse 17 x 17 Barrel/Baffle Configuration Various Upflow Minimum Chemical Precipitation Time (tchem) 143 minutes (tblock, WCAP-17788, Vol 1, Table 6-1) 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> Maximum HLSO Time 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> (tchem) 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> Maximum Core Inlet Fiber Load for 10 HLB WCAP-17788, Volume 1, Table 6-3 94.29 g/FA (Failure of both CS pumps)

Total In-Vessel Fiber Limit for 10 HLB WCAP-17788, Volume 1, Section 6.4 Minimum Sump Switchover (SSO) Time 20 minutes 13 minutes Maximum Rated Thermal Power 3658 MWt 3565 MWt Maximum AFP Resistance WCAP-17788, Volume 4, Table 6-1 WCAP-17788, Volume 4, Table RAI-4.2-24 ECCS Flow per FA 8 - 40 gpm/FA 37.8 to 52.9 gpm/FA

In-Vessel Downstream Effects

• Maximum in-vessel fiber load for breaks up to 10 exceeds core-inlet fiber limit but are bounded by total in-vessel fiber limit in WCAP-17788

- WCAP core-inlet fiber limit conservatively low based on assumption of uniform fiber bed at core inlet

- Licensees may justify that a non-uniform debris bed will form at the core inlet allowing adequate flow to assume LTCC, even though the average debris load per FA metric is exceeded 18

In-Vessel Downstream Effects

• Earliest Wolf Creek SSO time (13 min) not bounded by that assumed in WCAP analysis (20 min)

- The 13 min SSO time represents shortest injection model duration and was calculated very conservatively Maximum pump flow rates based on 0 psig containment pressure All pumps operating with no credit for pump startup time Minimum RWST volume based on Tech Spec limit

- Wolf Creek decay heat at SSO lower than that used in WCAP 19 Decay Heat at SSO (MWt)

SSO Time (min)

Thermal Power (MWt)

Decay Heat Model WCAP-17788 87.4 20 3,658 10CFR50 Appendix K model (1971 ANS Standard + 20%)

Wolf Creek 78.8 13 3,565 1971 ANS Standard + 2

In-Vessel Downstream Effects

- WCAP analysis assumed all debris arrives at core inlet within 60 sec after start of SSO Wolf Creek core inlet fiber load reaches WCAP limit 7.1 min after SSO

- Wolf Creek core inlet fiber load reaches 94.29 g/FA >1 hr after SSO Sensitivity runs in WCAP-17788 Vol 4 showed much reduced peak cladding temperature and no core-wide uncover when core inlet resistance linearly ramps up over 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> or 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />

• Wolf Creek ECCS flow per FA bounded by WCAP analysis as it exceeds min flow analyzed in WCAP

- Debris bed with highest resistance formed at min flow

- Unstable debris bed at higher flow rates 20

Risk Quantification

• GSI-191 risk quantification considered the following events

- Small, medium, and large LOCAs due to:

Pipe breaks Failure of non-piping components Water hammer

- Secondary side breaks inside containment that result in a consequential LOCA that requires sump recirculation

- Fire-induced RCP seal LOCAs

- Seismically-induced LOCAs

• Events were evaluated using a combination of quantitative (conservative or bounding) and qualitative methods 21

Risk Quantification

• Given a threshold break size of 10 inches for pipe break LOCAs, CDF was calculated to be 6.6E-07 yr-1 22

Risk Quantification

• Water hammer induced LOCAs, fire induced LOCAs, and other external events were determined to have no GSI-191 risk contribution

• Frequency of seismically induced large breaks was calculated using two separate methods:

- 6.9E-07 yr-1 based on representative fragility parameters from EPRI 3002000709

- 3.9E-07 yr-1 based on site-specific fragility parameters and the guidance in NUREG-1903

• All seismically induced large breaks were conservatively assumed to result in strainer failure, so frequency is equivalent to CDF 23

Risk Quantification

• Secondary side breaks do not generally require ECCS recirculation for long term decay heat removal

• However, subsequent failures following the initiating event (e.g., a stuck open PORV or loss of aux feedwater) could require recirculation to support feed and bleed cooling

• The PRA model was used to calculate a bounding risk contribution assuming that all secondary side breaks that require sump recirculation will fail due to the effects of debris

- CDF = 6.5E-08 yr-1

- LERF = 1.1E-10 yr-1 24

Risk Quantification

• Baseline CDF and LERF values are relatively high due to fire risk contribution

• CDF and LERF are outside the RG 1.174 guidelines for Region II (1E-04 and 1E-05, respectively) 25 PRA Model CDF (yr-1)

LERF (yr-1)

Internal Events 7.25E-06 7.31E-08 Internal Flooding 9.06E-06 3.77E-08 Internal Fire 5.49E-04 1.33E-05 High Winds 3.40E-06 7.98E-09 Total 5.69E-04 1.34E-05

Risk Quantification

• GSI-191 risk quantification results are within RG 1.174 Region III guidelines

• CDF and LERF values from various hazards are not added together since bounding methods were used to calculate values 26 Hazard CDF (yr-1)

LERF (yr-1)

Piping and Non-Piping LOCAs 6.6E-07 1.9E-11 Water Hammer Induced LOCAs 0.0 0.0 Secondary Side Breaks 6.5E-08 1.1E-10 Fire Induced LOCAs 0.0 0.0 Seismically Induced LOCAs 6.9E-07 2.0E-11 Other External Hazards 0.0 0.0

Uncertainty Quantification

• Uncertainty quantification considers:

- Parametric uncertainty

- Model uncertainty

- Completeness uncertainty

• Completeness uncertainty was qualitatively determined to be low

• Most parameters and models used for Wolf Creek GSI-191 risk quantification are conservative inputs or consensus models that do not require uncertainty quantification 27

Uncertainty Quantification

• An evaluation of GSI-191 inputs identified only one parameter that was not conservative or bounding:

- Mean LOCA frequency values

• CDF was recalculated using the 5th and 95th percentile values, which showed a range of 3.1E-09 yr-1 to 2.2E-06 yr-1 (compared to the base value of 6.6E-07 yr-1) 28

Uncertainty Quantification

• An evaluation of GSI-191 models identified only three models that are not consensus models:

- Continuum break model

- Geometric aggregation of LOCA frequencies

- Seismic LOCA frequency based on EPRI 3002000709

• CDF was recalculated using alternative models:

- DEGB-only model is qualitatively less conservative than continuum break model for threshold break methodology

- Arithmetic aggregation of LOCA frequencies are almost an order of magnitude higher than geometric aggregation

- Seismic LOCA frequency is lower based on site-specific fragilities and the guidance in NUREG-1903 29

Uncertainty Quantification 30 Base Case Input or Model Sensitivity Case Input or Model CDF (yr-1)

LERF (yr-1)

Pipe Break Risk Based on 25-year GM LOCA Frequency Input Pipe Break Risk Based on 25-year Geometric 5th Percentile Input 3.1E-09 8.8E-14 Pipe Break Risk Based on 25-year Geometric 95th Percentile Input 2.2E-06 6.2E-11 Pipe Break Risk Based on Continuum Break Model Pipe Break Risk Based on DEGB-Only Model

< 6.6E-07

< 1.9E-11 Pipe Break Risk Based on Geometric LOCA Frequency Model Pipe Break Risk Based on Arithmetic LOCA Frequency Model 5.2E-06 1.5E-10 Seismic Risk Model Based on Representative Fragility Parameters from EPRI 3002000709 Seismic Risk Model Based on Site-Specific Fragility Parameters and the Guidance in NUREG-1903 3.9E-07 1.1E-11

Adoption of TSTF-567

• Wolf Creek Tech Spec is consistent with NUREG-1431

• Wolf Creek plans to implement Tech Spec changes following the TSTF-567 model application

• Wolf Creek will review TSTF-567 and the NRCs SE to ensure that the justifications in TSTF-567 and the SE are applicable to Wolf Creek 31

Operability Evaluation

• With approval of risk-informed GSI-191 LAR, new design basis for Wolf Creek will be that risk increase due to GSI-191 failures is within RG 1.174 Region III (i.e., a CDF less than 1E-06 yr-1)

• The current NRC guidance does not allow the use of risk to address operability issues

• Debris limits are therefore defined to ensure plant stays within its design basis and can be used for operability determinations

• The plant design basis is maintained if none of the breaks smaller than threshold break size (10 inches) cause any GSI-191 failures 32

Operability Evaluation

• Strainer and in-vessel debris limits were developed to ensure that breaks 10 inches do not fail

• The debris limits were derived based on worst equipment configurations for strainer and in-vessel

- Single train failure for strainer evaluation

- Two RHR pumps operating with failure of both CS pumps at the start of recirculation for in-vessel effects

• The 10-inch threshold break size conservatively assumed to apply to all equipment configurations 33

Debris Limits

• For fiber fines, more limiting debris margin between strainer and in-vessel is used 34

Debris Limits

• Debris limits for all other debris types are based on strainer evaluation 35

Debris Limits Debris Type Debris Limit Max Debris Quantity for Breaks 10 Available Margin Fiber Fines (lbm) 144.1 119.6 24.5 Total Fiber Fines, Small Pieces, and Large Pieces (lbm) 322.5 235.8 86.7 Latent Particulate (lbm) 122.2 54.2 68.0 ThermoLag Particulate (ft3) 0.50 0.51 0

Coatings Particulate (ft3) 2.43 1.67 0.76 Degraded Paint Chips (ft2) 158.4 0

158.4 Miscellaneous Debris (ft2) 20.0 7.1 12.9 36

Operability Evaluation 37 Next Slide

Operability Evaluation 38 Previous Slide

Submittal Content

• Proposed LAR submittal includes the following:

- Attachment 1: License Amendment Request Implementation of risk-informed approach for GSI-191 Implementation of TSTF-567

- Attachment 2: Request for Exemption from certain requirements of 10 CFR 50.46 (a)(1)

- Attachments 3 to 6: Proposed Changes to Tech Spec (markup and clean version), Tech Spec Bases, and USAR

- Attachment 7: Overview of Risk-Informed Approach

- Attachment 8: Updated GL 2004-02 Responses

- Attachment 9: Defense in Depth and Safety Margins 39

Submittal Schedule

• Wolf Creek is currently working on the updated responses to GL 2004-02

• Final review by Wolf Creek licensing scheduled 2/3/2021 - 3/4/2021

• Current projected date for submittal to the NRC: April 2021 40

Closing

• Questions?

41