ML19214A065
| ML19214A065 | |
| Person / Time | |
|---|---|
| Site: | 99902043 |
| Issue date: | 08/06/2019 |
| From: | Prosanta Chowdhury NRC/NRO/DLSE/LB1 |
| To: | |
| Chowdhury P / 415-1647 | |
| References | |
| TR-0915-17772 | |
| Download: ML19214A065 (21) | |
Text
Staff Feedback on NuScales Response to RAI 9666 Related to the EPZ Sizing Methodology Topical Report (TR-0915-17772)
Public Meeting August 6, 2019
Abbreviations and Acronyms
- CDF - core damage frequency
- NPP - nuclear power plant
- CEMP - comprehensive
- PAG - protective action guide emergency management plans
- PGA - peak ground acceleration
- CNV - containment vessel
- CSDRS - certified seismic design
- PSHA - probabilistic seismic response spectrum hazards analysis
- DCA - design certification
- RAI - request for additional application information
- DID - defense-in-depth
- EPZ - emergency planning zone
- SER - safety evaluation report
- HCLPF - high confidence low
- SMA - seismic margins assessment probability of failure
- SSC - structures, systems, and
- LRF - large release frequency components
- LWR - light water reactor
- TR - topical report
RAI 9666, Question 1.05-34: Acceptability of a DC PRA for Risk Informed Applications
- RAI Response- the applicant will need to demonstrate the technical acceptability of the PRA is sufficient to support risk-informed decision making.
- Staff feedback - The response is satisfactory and staff will incorporate a condition of use in the SER on acceptability, including the impact of uncertainties on numerical thresholds.
Aug 6, 2019, Public Meeting on EPZ TR 3
RAI 9666, Question 1.05-36: a) How insights from Level 2 PRA are considered, and b) statement that severe accident phenomena are not credible
- RAI response
- Insights will be considered for accident sequences that screen in.
- TR updates:
- Removes references to severe accident phenomena not being credible
- Adds, methodology does not utilize the DCA PRA and assessment of containment integrity will be performed with the PRA which is associated with the application.
(Section 3.4.3)
- Adds, application of the EPZ methodology should consider the assessment of severe accident phenomena available at the time. (Section 3.8.2)
- Staff feedback
- Level 2 PRA insights issue will be addressed with RAI Question 1.05-38
- Clarification requested regarding added text relative to finality of design certifications Aug 6, 2019, Public Meeting on EPZ TR 4
RAI 9666, Question 1.05-37: Evaluation of module drop scenarios
- In TR, accidents are "less severe" if containment does not fail.
Less severe accidents are evaluated against the early phase PAGs.
- In TR, accidents are more severe if containment fails or if the containment is bypassed.
- More Severe accidents are compared against the 200 rem LRF criterion.
- Approach implies less severe accidents are more frequent/more severe accidents are less frequent.
- Appropriate for current LWR designs evaluated in NUREG 0396.
- Module drops are the most likely cause of core damage.
- Drop of a fully assembled module is assumed to cause a containment breach and is evaluated against the 200 rem criterion.
- Drop of the upper CNV and upper RPV on fuel located in the RFT is screened.
- The staff expected module drops to be evaluated to the early phase PAGs.
Aug 6, 2019, Public Meeting on EPZ TR 5
RAI 9666, Question 1.05-37: continued
- RAI response:
- Drop of upper CNV and upper RPV screened from the PRA (no Large release) and therefore, screened from the methodology.
- Staff response:
- Drop of an intact module and drop of the upper RPV and CNV on fuel are more likely scenarios but evaluated against the more severe/less likely criterion.
- NUREG 0396 (page I-9) states design basis accidents and less severe core-melt accidents should be considered for protective actions.
- To provide the same level of protection as NUREG 0396, module drops of fully assembled and partially assembled module should be assessed against the PAGs since they are most likely.
Aug 6, 2019, Public Meeting on EPZ TR 6
RAI 9666, Question 1.05-38: Consistency of proposed criteria for low Defense-in-Depth with Commission expectations for advanced light water reactors
- RAI Response
- DID analysis is aimed at establishing PRA technical adequacy
- If only an approved, technically adequate PRA can be used to support EPZ method, purpose of the DID analysis becomes extraneous and unnecessary
- Based on response to Question 01.05-34, DID evaluation section of TR should be deleted
- Staff Feedback
- Response appears to be inconsistent with NRCs PRA policy statement
- Confidence that more severe accidents (i.e. containment bypass sequences) would not produce significant off-site consequences would provide DID Aug 6, 2019, Public Meeting on EPZ TR 7
References on Defense-in-depth Regulatory Guide 1.174, Revision 3 The defense-in-depth philosophy has traditionally been applied in plant design and operation to provide multiple means to accomplish safety functions and prevent the release of radioactive material. It has been and continues to NUREG/KM-0009, Historical Review and be an effective way to account for uncertainties in equipment and human performance and, in particular, to Observations of Defense-in-Depth account for the potential for unknown and unforeseen ... the ultimate purpose of defense-in-depth is to compensate failure mechanisms or phenomena that, because they are for uncertainty (e.g., uncertainty due to lack of operational unknown or unforeseen, are not reflected in either the PRA experience with new technologies and new design features, or traditional engineering analyses. uncertainty in the type and magnitude of challenges to safety).
1995 Commission Policy Statement on Use of Defense-in-depth, in the NUREG, is defined as . . . an element PRA Methods in Nuclear Regulatory Activities of NRCs safety philosophy that is used to address uncertainty The use of PRA technology should be increased in by employing successive measure including safety margins to all regulatory matters to the extent supported by prevent and mitigate damage if a malfunction, accident or the state-of-the-art in PRA methods and data and in naturally caused event occurs at a nuclear facility.
a manner that complements the NRCs deterministic approach and supports the NRC's traditional defense-in-depth philosophy.
Aug 6, 2019, Public Meeting on EPZ TR 8
RAI 9666, Question 1.05-35: Adequacy of the PRA based SMA to determine EPZ size RAI response:
- As stated in NUREG-1738 (p. 3-37), for high PGA earthquake, it was reasoned that there would be no effective evacuation and many structures would be uninhabitable.
- Pre-planned response as defined in emergency plans would be significantly less effective than an integrated national response, which assesses and accounts for damaged infrastructure over a wide area.
- Seismic events, seismic PRA, and SMA were not considered in determining the 10-mile plume exposure distance for EPZ.
Aug 6, 2019, Public Meeting on EPZ TR 9
RAI 9666, Question 1.05-35: Adequacy of the PRA based SMA to determine EPZ size (continued)
Staff response:
- Evacuation is site specific.
- The EP rule does not rely on comprehensive emergency management plans (CEMPs). These CEMPs are not reviewed and approved by FEMA.
- NUREG 0396 Appendix III states, no specific design basis accident or Class 9 accident scenario can be isolated as the one for which to plan because each accident would have different consequences. It appears that no accidents or accident initiators were explicitly eliminated on an a-priori basis.
Aug 6, 2019, Public Meeting on EPZ TR 10
Seismic Risk
- Does the PRA based seismic margins approach (SMA) appropriately consider credible seismic induced core damage events?
- New application for the SMA.
Aug 6, 2019, Public Meeting on EPZ TR 11
Staffs confirmatory calculations
- Use fragility information provided for NuScale structural SSCs to estimate failure frequency due to seismic loading
- Use hazard curves from Near-Term Task Force Recommendation 2.1 (R2.1)
Seismic Hazard reevaluations to model seismic loading of structural SSCs
- R2.1 hazard results incorporate latest seismic source and ground motion models
- Detailed geologic siting information for each site allows determination of local site response
Table 19.1-35: Structural Fragility Parameters and Results Structures Am (g) r u HCLPF (g) Controlling Failure Mode Assumed consequence Reactor Building Crane 2.64 0.28 0.39 0.88 Bridge seismic restraint Core damage / Large Release weldment yielding Reactor Building Exterior Walls 1.92 0.12 0.33 0.92 Out-of-plane shear Core damage / Large Release NPM Supports 1.98 0.12 0.35 0.92 Shear failure of multiple shear Core damage/Large Release lugs Bio Shield - horizontal shear flexure 11.62 0.28 0.37 3.99 Horizontal shield slab bending Core damage / Large Release
-normal operation failure Bio shield - pool wall bolt failure - 5.37 0.28 0.35 1.91 Shear Failure of pool wall Anchor Core damage / Large Release normal operation Bolts Bio shield - horizontal shear flexure - 4.05 0.28 0.41 1.30 Bending failure of both stacked Core damage / Large Release double stacked for refueling of adj. shield slabs when configuration present model Bio shield - pool wall bolt failure - 3.05 0.28 0.35 1.08 Shear Failure of pool wall Anchor Core damage / Large Release double stacked for refueling of adj. Bolts when configuration present model Pool Walls 2.31 0.21 0.33 0.95 Out-of-plane shear Core damage / Large Release Crane Support Walls 2.61 0.12 0.34 1.23 Out-of-plane shear Core damage / Large Release Bay Walls 2.65 0.12 0.31 1.31 In-plane flexure Core damage / Large Release Roof 2.22 0.12 0.26 1.20 In-plane shear Core damage / Large Release Basemat 3.57 0.27 0.31 1.38 Out-of-plane shear Core damage / Large Release Am = median seismic capacity; u = uncertainty in the median seismic capacity; r = randomness of the fragility evaluation; HCLPF = High-Confidence (95%) of a Low Probability (5%) of Failure, Reference 19.1-57 Aug 6, 2019, Public Meeting on EPZ TR 13
Crane Fragility
- NuScale Crane HCLPF High Confidence Low Probability of Failure
- HCLPF=0.88 g
- Uncertainties
- u=0.39
- r=0.28
- The natural or resonance frequency for large structural SSCs ranges from about 0.5 to 2.5 Hz
Crane Fragility
- Use NuScale Certified Seismic Design Response Spectrum (CSDRS) to develop HCLPF frequency adjustment factors Spectral Freq (Hz) Spectral Acc (g) Freq Adj Factor HCLPF (g) 0.5 0.3 0.3/0.5=0.6 0.528 1.0 0.6 0.6/0.5=1.2 1.056 2.5 1.0 1.0/0.5=2.0 1.760 100 (PGA) 0.5 0.5/0.5=1 0.880 Aug 6, 2019, Public Meeting on EPZ TR 15
Crane Fragility for Various Spectral Frequencies 0.5 Hz 100 Hz (PGA) 1 Hz 2.5 Hz 16 Aug 6, 2019, Public Meeting on EPZ TR
Near-Term Task Force Recommendation 2.1 (R2.1) Seismic Hazard Curves
- Seismic Source Characterization Model
- Seismic Ground Motion Characterization Model
- Site Response Evaluation based on dynamic properties of local site geology Aug 6, 2019, Public Meeting on EPZ TR 17
Estimated Structural SSC Failure Frequency
- Convolve seismic hazard with SSC seismic fragility to determine failure frequency
- Average SSC failure frequencies for 0.5, 1.0, and 2.5 Hz to determine composite failure frequency Aug 6, 2019, Public Meeting on EPZ TR 18
Response Spectra Comparisons Aug 6, 2019, Public Meeting on EPZ TR 19
Results for NuScale Structures Estimated Estimated HCLPF Assumed Seismic Failure Seismic Failure (g) Consequence DCA Frequency Frequency Component Table 19.1-35 Table 19.1-35 Plant 1 Plant 2 Reactor Building .88g Core Damage (CD)/ 2.61E-07 6.02E-07 Crane Large Release (LR)
Reactor Building .92 CD/LR 3.38E-07 8.91E-07 Exterior Walls NPM Supports .92 CD/LR 3.15E-07 8.26E-07 Pool Walls .95 CD/LR 2.58E-07 6.72E-07 Crane Support Walls 1.23 CD/LR 1.01E-07 2.56E-07 Bay Walls 1.31 CD/LR 8.63E-08 2.17E-07 Roof 1.20 CD/LR 1.52E-07 3.95E-07 Basemat 1.38 CD/LR 5.33E-08 1.31E-07 20 Aug 6, 2019, Public Meeting on EPZ TR
Notes
- Each key structure that screens-in would represent a separate seismic sequence.
- Staff is open to alternate ways of handling seismic core damage risk.
- Staff needs to ensure that all credible seismic accident scenarios are being considered.