ML12030A081: Difference between revisions

From kanterella
Jump to navigation Jump to search
(Created page by program invented by StriderTol)
(Created page by program invented by StriderTol)
 
Line 1: Line 1:
{{Adams
#REDIRECT [[NOC-AE-11002773, Supplemental Response to Requests for Additional Information for the South Texas Project License Renewal Application - SAMA]]
| number = ML12030A081
| issue date = 01/19/2012
| title = Project, Units 1 and 2 - Supplemental Response to Requests for Additional Information for the South Texas Project License Renewal Application - SAMA
| author name = Rencurrel D W
| author affiliation = South Texas Project Nuclear Operating Co
| addressee name =
| addressee affiliation = NRC/Document Control Desk, NRC/NRR
| docket = 05000498, 05000499
| license number =
| contact person =
| case reference number = G25, NOC-AE-11002773, STI: 33163064, TAC ME4938, TAC ME5122
| document type = Letter
| page count = 59
| project = TAC:ME4938, TAC:ME5122
| stage = Response to RAI
}}
 
=Text=
{{#Wiki_filter:Nuclear Operating Company South Texas Prolect Electric GeneratinS Station PO Box 289 Wadsworth, Texas 77483 A^VVýJanuary 19, 2012 NOC-AE-1 1002773 10CFR54 STI: 33163064 File: G25 U. S. Nuclear Regulatory Commission Attention:
Document Control Desk One White Flint North 11555 Rockville Pike Rockville, MD 20852-2738 South Texas Project Units 1 and 2 Docket Nos. STN 50-498, STN 50-499 Supplemental Response to Requests for Additional Information for the South Texas Proiect License Renewal Application
-SAMA (TAC Nos. ME4938 and ME5122)
 
==References:==
: 1. STPNOC Letter dated October 25, 2010, from G. T. Powell to NRC Document Control Desk, "License Renewal Application" (NOC-AE-10002607) (ML103010257)
: 2. STPNOC Letter dated September 22, 2011, from G. T. Powell to NRC Document Control Desk, "Response to Requests for Additional Information for the South Texas Project License Renewal Application (TAC Nos. ME4938 and ME5122)" (NOC-AE-1 1002735) (ML1I1270A060)
By Reference 1, STP Nuclear Operating Company (STPNOC) submitted a License Renewal Application (LRA) for South Texas Project (STP) Units 1 and 2. By Reference 2, STPNOC responded to NRC staff requests for additional information for review of the STP LRA regarding severe accident mitigation alternatives (SAMA). In Reference 2, STPNOC stated that an assessment would be completed in order to respond to requests for additional information and the results of the assessment would be reported to the NRC. The enclosure to this letter provides the results of the assessment.
There are no regulatory commitments in this letter.Should you have any questions regarding this letter, please contact either Arden Aldridge, STP License Renewal Project Lead, at (361) 972-8243 or Ken Taplett, STP License Renewal Project regulatory point-of-contact, at (361) 972-8416.I declare under penalty of perjury that the foregoing is true and correct.Executed on //1-1 It,,,/ t Eat-e" enucrre~l
/Senior Vice President, Technical Support & Oversight KJT
 
==Enclosure:==
 
STPNOC Supplemental Response to Requests for Additional Information-A t L-Y7 NOC-AE-1 1002773 Page 2 cc: (paper copy)(electronic copy)Regional Administrator, Region IV U. S. Nuclear Regulatory Commission 1600 East Lamar Boulevard Arlington, Texas 76011-4511 Balwant K. Singal Senior Project Manager U.S. Nuclear Regulatory Commission One White Flint North (MS 8B1)11555 Rockville Pike Rockville, MD 20852 Senior Resident Inspector U. S. Nuclear Regulatory Commission P. 0. Box 289, Mail Code: MN1 16 Wadsworth, TX 77483 C. M. Canady City of Austin Electric Utility Department 721 Barton Springs Road Austin, TX 78704 John W. Daily License Renewal Project Manager (Safety)U.S. Nuclear Regulatory Commission One White Flint North (MS O11-Fl)Washington, DC 20555-0001 Tam Tran License Renewal Project Manager (Environmental)
U. S. Nuclear Regulatory Commission One White Flint North (MS O11FO1)Washington, DC 20555-0001 A. H. Gutterman, Esquire Kathryn M. Sutton, Esquire Morgan, Lewis & Bockius, LLP John Ragan Chris O'Hara Jim von Suskil NRG South Texas LP Kevin Polio Richard Pena City Public Service Peter Nemeth Crain Caton & James, P.C.C. Mele City of Austin Richard A. Ratliff Alice Rogers Texas Department of State Health Services Balwant K. Singal John W. Daily Tam Tran U. S. Nuclear Regulatory Commission Enclosure NOC-AE-1 1002773 Enclosure STPNOC Supplemental Response to Requests for Additional Information Attachment 1: Attachment 2: Fire PRA Impact Assessment Seismic Impact Assessment Enclosure NOC-AE-1 1002773 Page 1 of 4 STPNOC Supplemental Response to Requests for Additional Information SOUTH TEXAS PROJECT LICENSE RENEWAL APPLICATION REQUESTS FOR ADDITIONAL INFORMATION REGARDING THE ANALYSIS OF SEVERE ACCIDENT MITIGATION ALTERNATIVES (SAMA)
 
==References:==
: 1. STPNOC Letter dated July 5, 2011, from G. T. Powell to NRC Document Control Desk,"Response to Request for Additional Information for the South Texas Project License Renewal Application (TAC Nos. ME4938)" (NOC-AE-1 1002687) (ML1 1 193A016)2. STPNOC Letter dated September 22, 2011, from G. T. Powell to NRC Document Control Desk, "Response to Requests for Additional Information for the South Texas Project License Renewal Application (TAC Nos. ME4938 and ME5122)" (NOC-AE-1 1002735) (ML1 1270A060)References 1 and 2 provide the previous STPNOC responses to the NRC Requests for Information (RAI) numbers 3.b and 3.c. The RAIs are repeated below with the STPNOC supplemental response provided.NRC RAI 3.b (Reference 1): Requested information:
In the May 9, 2007 STPNOC response to RAls for risk-managed technical specifications (RMTS), it was stated that a review of the fire frequency data presented in NUREG/CR-6850 was planned for a future reanalysis of fire hazards at STP. If the results of this review have not been incorporated in STPREV6, assess the impact the fire frequency data on the SAMA assessment.
NRC RAI 3.b (Reference 2): Fire-analysis individual plant examination (IPE) -use of technical report NUREGICR-6850 Backgoround:
The response to this RAI simply indicates that a review of the NUREG/CR-6850 will be performed in the future. However, the recent research and guidance reported in NUREG/CR-6850, specifically in the areas of hot short probabilities, fire ignition frequencies, and non-suppression probabilities, indicate that the fire analysis methodologies utilized for the IPE may underestimate fire risk.Requested information:
Provide assurance that consideration of this new information is not expected to impact the selection of cost beneficial severe accident mitigation alternatives (SAMAs) for South Texas Project (STP).
Enclosure NOC-AE-1 1002773 Page 2 of 4 STPNOC Response: STPNOC performed an assessment using the new information provided in NUREG/CR-6850,"EPRI/NRC-RES Fire PRA Methodology for Nuclear Power Facilities," for assessing impact on the selection of cost beneficial severe accident mitigation alternatives (SAMAs) for the South Texas Project (STP). The assessment considered insights provided in NUREG/CR-6850 regarding hot short probabilities, fire ignition frequencies, and non-suppression probabilities.
The STP Fire Probabilistic Risk Analysis (PRA) model was modified only for the purpose of performing the assessment.
No revision to the STP Fire PRA resulted from this activity.The insights from NUREG/CR-6850 were applied to the fire analyses from the IPE and as augmented through the STPREV 6 version of the PRA model. The insights from NUREG/CR-6850 were applied to the fire scenarios currently in the PRA model. The resultant core damage frequency (CDF) from the assessment was used to screen the list of SAMAs currently considered in the South Texas Project License Renewal Application (LRA) to reevaluate if any SAMAs would be cost-beneficial.
An uncertainty multiplier of 2.7 from the response to RAI ld (
 
==Reference:==
 
STPNOC response to NRC Requests for Additional Information for the South Texas Project License Renewal Application dated August 23, 2011 [NOC-AE-1 1002711][ML 1250A067])
was applied to the PRA results for the updated quantifications and the cost benefit analysis was updated to determine if the use of the updated fire scenario initiating event frequencies would impact the conclusions of the SAMA analysis.
The conclusions of the SAMA analysis provided in the LRA did not change.See the Table below.Cost Benefit Analysis Results for Phase II SAMAs Supported by Detailed PRA Quantifications (SAMAFR14 Model, 9 5 th Percentile PRA Results)Cost of Total Averted Change in Implementation Cost-Risk Status?SAMA 3b $796,677 $8,905 -$787,772 No SAMA 4 $100,000 $70,621 -$29,379 No SAMA 10 $100,000 $9,828 -$90,172 No SAMA 12 $100,000 $4,147 -$95,853 No SAMA 13 $100,000 $5,346 -$94,654 No SAMA 15 $100,000 $18,932 -$81,068 No SAMAFR14 -the name given to the SAMA fire PRA model used for the assessment.
Attachment 1 to this Enclosure provides the details of the assessment.
Enclosure NOC-AE-1 1002773 Page 3 of 4 NRC RAI 3.c (Reference 1): Requested information:
Identify the seismic hazard curves used to determine the seismic CDF in STPREV6. If the seismic CDF is based on the Electric Power Research Institute (EPRI) hazard curve, provide the seismic CDF using the Lawrence Livermore National Laboratory (LLNL) hazard curve or the more recent USGS 2008 assessment and include a description of the dominant seismic CDF sequences.
Discuss the impact of these results on the SAMA assessment.
NRC RAI 3.c (Reference
: 2) -Seismic-analysis IPE of external events (IPEEE) -use of Lawrence Livermore National Laboratory (LLNL) or United States Geological Survey (USGS)hazard curves
 
==Background:==
 
The response to this RAI did not provide the requested updated seismic core damage frequency (CDF) results. Instead, the applicant cited Notice IN 2010-18. While this Information Notice concluded that the US plants had adequate safety margin, it did indicate that the seismic CDF for STP could be as high as 3E-06 per year (for spectral accelerations of 5 hz and 10 hz). This is 40 times the total seismic CDF given in the environmental report (ER). Also, note that the STP IPE gives a seismic CDF using the LLNL hazard curve of 1.7E-05 per year which is over 200 times the value used in the probabilistic risk assessment (PRA). Since the seismic CDF was determined using point estimates, the seismic CDF for the analysis STPREV6, based on the LLNL hazard curve, can be obtained from the LLNL seismic frequencies from Table 3.4.4-9 of the IPE or IPEEE or both and the conditional core damage probabilities (CCDPs) from Table F.2-1 of the ER. The result is a seismic core damage frequency (SCDF) of 8.7E-06 per year.Comparing the USGS hazard curves for the STP site with the Electric Power Research Institute (EPRI) hazard curves indicates that the frequency for the USGS curves is 60 to 150 times those for the EPRI curves over the range of 0.4 to 0.6 g which is the range for the largest contributors to STP seismic CDF. Furthermore, the USGS hazard curve is higher than the LLNL hazard curve by a factor of 1.5 to 2 over the same range.Using the above method for determining the CDF for SEIS3 and SEIS4 initiators for the seismic event scenarios gives seismic CDFs of 4E-06 and 5E-06 per year, respectively.
This indicates that applying the LLNL hazard curves or the 2008 USGS hazard curves to the SEIS3 and SEIS4 initiators could lead to CDF contributions of about 60 to 150% of the STPREV6 total CDF.Requested information:
Provide an assessment of the seismic CDF contribution due to the updated USGS hazard curves and the potential for cost beneficial SAMAs.STPNOC Response: STPNOC performed an assessment of the seismic Level 1 and Level 2 contributions due to the updated USGS hazard curves for impact on the selection of cost beneficial SAMAs for STP.The STP PRA model was modified only for the purpose of performing the assessment.
No revision to the STP PRA resulted from this activity.
Enclosure NOC-AE-1 1002773 Page 4 of 4 The acceleration values from the USGS hazard curves were applied to the current STP PRA model and the impact on those key components previously identified in the current STP PRA model was examined.
Because the fragilities for many of these key components were conservative, revised fragilities were assumed, if necessary and where justified, to determine the resultant CDF and Level 2 release category frequencies.
The results from the assessment were used to update both the SAMA identification process and the cost benefit analysis.An uncertainty multiplier of 2.7 from the response to RAI ld (
 
==Reference:==
 
STPNOC response to NRC Requests for Additional Information for the South Texas Project License Renewal Application dated August 23, 2011 [NOC-AE-1 1002711][ML1 1250A067])
was applied to the Phase II quantification results using the 2008 USGS Seismic Hazard Curves and the updated fragility data. The cost benefit analysis was updated to determine if the use of updated seismic data would impact the conclusions of the SAMA analysis.
The conclusions of the analysis did not change. See the Table below.Cost Benefit Analysis Results for Phase II SAMAs Supported by Detailed PRA Quantifications (2008 USGS Seismic Hazard Curves, Fragility Data Updates, 9 5 te Percentile PRA Results)Cost of Total Averted Change in Implementation Cost-Risk Status?SAMA 3b $796,677 $10,514 -$786,163 No SAMA 4 $100,000 $71,906 -$28,094 No SAMA 10 $100,000 $8,122 -$91,878 No SAMA 12 $100,000 $2,187 -$97,813 No SAMA 13 $100,000 $5,081 -$94,919 No SAMA 15 $100,000 $18,587 -$81,413 No Using the 2.7 uncertainty multiplier, the cost benefit analysis was updated to determine if the simultaneous use of the updated fire and seismic data together in one model would impact the conclusions of the SAMA analysis.
As documented in the Table below, the conclusions of the analysis did not change Cost Benefit Analysis Results for Phase II SAMAs Supported by Detailed PRA Quantifications (Updated Fire and Seismic Data, 9 5th Percentile PRA Results)Cost of Total Averted Change in AAIDNet Value Stus Implementation Cost-Risk Status?SAMA 3b $796,677 $17,771 -$778,906 No SAMA 4 $100,000 $82,890 -$17,110 No SAMA 10 $100,000 $8,197 -$91,803 No SAMA 12 $100,000 $2,597 -$97,403 No SAMA 13 $100,000 $15,741 -$84,259 No SAMA 15 $100,000 $22,000 -$78,000 No Attachment 2 to this Enclosure provides the details of the assessment.
Attachment 1 to Enclosure NOC-AE-1 1002773 Attachment I Fire PRA Impact Assessment Attachment 1 to Enclosure NOC-AE-1 1002773 Page 1 of 21 Fire PRA Impact Assessment Notes: 1. This assessment used the new information provided in NUREG/CR-6850,"EPRI/NRC-RES Fire PRA Methodology for Nuclear Power Facilities," for assessing impact on the selection of cost beneficial severe accident mitigation alternatives (SAMAs) for the South Texas Project (STP).2. The references for this assessment report are provided on page 21 of this attachment.
Discussion The scope of RAI 3.b has been subsequently clarified to request the impacts of applying the NUREG/CR-6850 (Reference
: 1) approach to the following three technical issues: (1) The impact of fire frequency data on fire scenario initiating event frequency in the PRA;(2) The impact of addressing hot shorts in the PRA fire scenarios; and (3) The impact of fire suppression (or, more specifically, non-suppression probability) assumptions applied in the PRA fire scenarios.
The response for this RAI is limited to the potential impacts on fire scenarios retained in the existing South Texas Project (STP) PRA (STPREV6) applied for SAMA analysis, and, therefore, does not include potential impacts on the detailed fire scenario screening analyses performed to identify and limit the fire scenarios retained for quantification in STPREV6. The following four documents describe, in detail, fire PRA technical work that has been performed for the STPEGS Units 1 and 2 PRA to date:* the original STP Probabilistic Safety Assessment (PSA) (Reference 2), specifically Sections 8, 9, and Appendix D;" the STP Level 2 Probabilistic Safety Assessment and Individual Plant Examination (Reference 3), specifically Sections 3.4.1 and 3.4.2;" the Fire Analysis Update for the STPEGS PSA for Selected Fire Zones (Reference 4);" and the Fire Sensitivity Case Studies for Risk-Managed Technical Specifications (RMTS) (Reference 5).There are eight fire scenarios retained for quantification following screening (using the 1.7E-07 CDF screening threshold applied in the STP PRA) in the STP_REV6 PRA model. Table 1 below identifies these fire scenarios and summarizes their Level 1 PRA impacts, as currently assessed in STPREV6. These results are sorted by decreasing contribution to reactor core damage frequency (CDF). Table 1 also shows the percent contribution to total CDF and the initiating event conditional core damage probability (CCDP) value. Table 2 below summarizes the Level 2 PRA impacts of these eight scenarios, as currently assessed in STPREV6. Table 2 presents the contribution of each fire scenario initiating event to Level 2 PRA release category frequencies.
The Level 2 PRA tracks four release categories identified as follows:
Attachment 1 to Enclosure NOC-AE-1 1002773 Page 2 of 21* Release Category I (RELI) also known as large early release frequency (LERF);* Release Category II (RELII) also know as small early release frequency;" Release Category III (RELIII) also know as late containment failure release frequency;
* and Release Category IV (RELIV) also know as intact containment scenario frequency.
Note that the values presented in Tables 1 and 2 are rounded.
Attachment 1 to Enclosure NOC-AE-1 1002773 Page 3 of 21 Table 1. STPREV6 Level I PRA Fire Scenario Results Summary (for CDF Contribution)
Fire Scenario IE Freq.* CDF* %CDF CCDP 1 1.46E-05 3.98E-07 6.23 2.72E-02 2 2.34E-07 2.12E-07 3.32 9.06E-01 3 2.72E-03 1.83E-07 2.87 6.74E-05 4 2.12E-06 1.22E-07 1.91 5.77E-02 5 3.18E-06 6.40E-08 1.00 2.01E-02 6 1.61 E-06 2.61E-08 0.41 1.62E-02 7 1.08E-03 1.1OE-08 0.17 1.01E-05 8 3.43E-06 1.02E-09 0.02 2.98E-04 All Fire Initiating Events in the PRA: 3.83E-03 1.02E-06 15.93 2.66E-04 All Initiating Events in the 2.33E+00 6.39E-06 100.00 2.74E-06 PRA: 0 Events per reactor year Note: Each fire scenario number above represents a specific fire zone in the plant. The fire zone locations are provided in the reports used to support this letter response.For purposes of this letter, the locations in the plant are only represented by a fire scenario number.
Attachment 1 to Enclosure NOC-AE-1 1002773 Page 4 of 21 Table 2. STPREV6 Level 2 PRA Fire Scenario Results Summary (by Release Category Contribution)
RELI Freq. RELII RELIII RELIV Fire Scenario IE Freq. (LERF) Freq. Freq. Freq.1 1.46E-05 2.02E-08 1.93E-07 7.81E-08 1.03E-07 2 2.34E-07 4.47E-09 1.79E-07 1.22E-08 1.60E-08 3 2.72E-03 3.63E-09 7.77E-08 2.04E-08 7.17E-08 4 2.12E-06 9.55E-09 4.75E-08 2.75E-08 3.63E-08 5 3.18E-06 2.66E-09 4.24E-08 8.15E-09 9.71E-09 6 1.61E-06 3.14E-10 5.61E-11 4.82E-10 2.45E-08 7 1.08E-03 0.OOE+00 1.16E-10 4.03E-09 5.64E-09 8 3.43E-06 1.13E-11 0.OOE+00 9.59E-12 8.66E-10 All Fire Initiating Events in the PRA: 3.83E-03 4.08E-08 5.40E-07 1.51E-07 2.68E-07 All Initiating Events in the 2.33E+00 5.01E-07 1.16E-06 1.48E-06 3.1OE-06 PRA:
Attachment 1 to Enclosure NOC-AE-1 1002773 Page 5 of 21 Reference 4 provides the most current description of the technical analyses supporting the following five fire scenarios retained in the PRA:.1* 2,* 3,* 5, and* 7.Section 3.4.2.4 of Reference 3 provides the most current description of the technical analyses supporting fire scenarios 4, 6, and 8.Detailed discussions of all the factors applied in deriving the fire scenario initiating event (IE)frequencies applied in the PRA, and shown in Tables 1 and 2 above are provided in References 3 and 4. These discussions are quite lengthy and complex. The general equations applied in deriving these IE frequencies are reviewed herein.Fire Scenario 1 (FS1) IE Frequency:
IEFS1 = fl*f2*(fB+fC+fM+fG+fP+fR+fs+fT+fN)*RFs1 where (Table 5-18 of Reference 4)IEFs1 = FS1 initiating event frequency applied in the PRA model (= 1.46E-05 in Table 1), f, = fire damage fraction representing the fraction of fires that affects at least two adjacent cable trays, f2 = fire damage fraction representing the fraction of fires that affects more than two cable trays, given two adjacent cable trays are affected, fl
* f2 = XS
* fNSSF + (1-Xs)*fNSLF
= 4.54E-03 for FS1 (see discussion in Section 5.5.2.2 of Reference 4); in effect, this is the consolidated fire non-suppression factor for all fires in the target location, where Xs = the fraction of transient fires at the target location that are small fires, fNSSF = non-suppression factor for small transient fires that can damage all target cable trays, fNSLF = non-suppression factor for large transient fires that can damage all target cable trays, fB = cable fraction from Table 5-9 of Reference 4 = 8.54E-01, fc = cable fraction from Table 5-9 of Reference 4 = 6.81 E-02, fM = cable fraction from Table 5-9 of Reference 4 = 2.81 E-02, Attachment 1 to Enclosure NOC-AE-1 1002773 Page 6 of 21 fG = cable fraction from Table 5-9 of Reference 4 = 9.79E-03, fp = cable fraction from Table 5-9 of Reference 4 = 5.55E-03, fR = cable fraction from Table 5-9 of Reference 4 = 5.25E-03 (includes hot short factor of 0.30), fs = cable fraction from Table 5-9 of Reference 4 = 2.53E-03, fT = cable fraction from Table 5-9 of Reference 4 = 7.13E-04, fN = cable fraction from Table 5-9 of Reference 4 = 2.61 E-02 (includes open circuit factor of 0.70 = 1-0.30), and RFs1 = FS1 fire ignition frequency (from Reference
: 4) = 3.22E-03 events/Rx-year.
Table 3 shows how these values were modified to apply the NUREG/CR-6850 methodology regarding fire zone ignition frequency, hot shorts, and non-suppression factors in an effective bounding sensitivity case study of this fire scenario frequency.
The fire zone ignition frequency was modified by applying the sum of items 5, 6, and 7 in Table 6-1 of Reference 1, a location weighting factor of 1.0, and an ignition source weighting factor of 0.02832 based on floor area fraction of the building from Table 3.4.1-14 of Reference
: 3. The hot short factor of 0.30 applied in Reference 4 was increased to 0.60 based on information provided in Reference 1 regarding the possible range of hot short fractions from 0.2 to 1.0. The fire non-suppression factor of 4.54E-03 was increased to 6.90E-02 based on the representative fire non-suppression factor derived in Appendix P of Reference
: 1. As shown in Table 3, the modified FS1 frequency is 2.95E-05 events per year.
Attachment 1 to Enclosure NOC-AE-1 1002773 Page 7 of 21 Table 3. FSI Scenario Frequency Case Parameter Il"12 fB fc fM fG fP fR- fs fT fN* RFS1 IE FSI Original 4.54E-03 8.54E-01 6.81E-02 2.81E-02 9.79E-03 5.55E-03 5.25E-03 2.53E-03 7.13E-04 2.61E-02 3.22E-03 1.46E-05 For RAI 6.90E-02 8.54E-01 6.81E-02 2.81 E-02 9.79E-03 5.55E-03 1.05E-02 2.53E-03 7.13E-04 1.49E-02 4.30E-04 2.95E-05 Attachment 1 to Enclosure NOC-AE-1 1002773 Page 8 of 21 Fire Scenario 2 (FS2) IE Frequency:
IEFs2 = fl*f2*(fA+fB+fC+fD+fP+fG+fL+fN)*RFS2 where (Table 5-17 of Reference 4)IEFs2 = FS2 initiating event frequency applied in the PRA model (= 2.34E-07 in Table 1), f, = fire damage fraction representing the fraction of fires that affects at least two adjacent cable trays, f2 = fire damage fraction representing the fraction of fires that affects more than two cable trays, given two adjacent cable trays are affected, fl
* f 2 = XS
* fNSSF + (1-Xs)*fNSLF
= 4.67E-03 for FS2 (see discussion in Section 5.5.2.2 of Reference 4), where Xs = the fraction of transient fires at the target location that are small fires, fNSSF = non-suppression factor for small transient fires that can damage all target cable trays, fNSLF = non-suppression factor for large transient fires that can damage all target cable trays, fA = cable fraction from Table 5-8 of Reference 4 = 2.71 E-01, fB = cable fraction from Table 5-8 of Reference 4 = 1.96E-01, fc = cable fraction from Table 5-8 of Reference 4 = 2.02E-01, fD = cable fraction from Table 5-8 of Reference 4 = 1.91 E-01, fp = cable fraction from Table 5-8 of Reference 4 = 5.59E-02, fG = cable fraction from Table 5-8 of Reference 4 = 3.87E-03 (includes hot short factor of 0.30), fL = cable fraction from Table 5-8 of Reference 4 = 3.87E-02, fN = cable fraction from Table 5-8 of Reference 4 = 4.13E-02 (includes open circuit factor of 0.70 = 1-0.30 for one raceway), and RFS2 = FS2 fire ignition frequency (from Reference
: 4) = 5.01 E-05 events/Rx-year Table 4 shows how these values were modified to apply the NUREG/CR-6850 methodology regarding fire zone ignition frequency, hot shorts, and non-suppression factors in an effective bounding sensitivity case study of this fire scenario frequency.
The fire zone ignition frequency was modified by applying the sum of items 5, 6, and 7 in Table 6-1 of Reference 1, a location weighting factor of 1.0, and an ignition source weighting factor of 0.00099 based on floor area fraction of the building from Table 3.4.1-14 of Reference
: 3. The hot short factor of 0.30 applied Attachment 1 to Enclosure NOC-AE-1 1002773 Page 9 of 21 in Reference 4 was increased to 0.60 based on information provided in Reference 1 regarding the possible range of hot short fractions from 0.2 to 1.0. The fire non-suppression factor of 4.67E-03 was increased to 6.90E-02 based on the representative fire non-suppression factor derived in Appendix P of Reference
: 1. As shown in Table 4, the modified FS2 frequency is 1.04E-06 events per year.
Attachment 1 to Enclosure NOC-AE-1 1002773 Page 10 of 21 Table 4. FS2 Scenario Frequency Case Parameter fl*f 2  fA fB fc fD fp fG* fL fN* Rzo07i IE FS2 Original 4.67E-03 2.71E-01 1.96E-01 2.02E-01 1.91E-01 5.59E-02 3.87E-03 3.87E-02 4.13E-02 5.01E-05 2.34E-07 For RAI 6.90E-02 2.71E-01 1.96E-01 2.02E-01 1.91E-01 5.59E-02 7.74E-03 3.87E-02 3.74E-02 1.50E-05 1.04E-06 Attachment 1 to Enclosure NOC-AE-1 1002773 Page 11 of 21 Fire Scenario 3 (FS3) IE Frequency:
IEFs3 = (1-fl)*fB*RFs3 where (Table 5-18 of Reference 4)IEFs3 = FS3 initiating event frequency applied in the PRA model (= 2.72E-03 in Table 1), f, = fire damage fraction representing the fraction of fires that affects at least two adjacent cable trays = 1.31 E-02 (from Reference 4), fa = cable fraction from Table 5-9 of Reference 4 = 8.54E-01, and RFS3 = FS3 fire ignition frequency (from Reference
: 4) = 3.22E-03 events/Rx-year.
Table 5 shows how these values were modified to apply the NUREG/CR-6850 methodology regarding fire zone ignition frequency, hot shorts, and non-suppression factors in an effective bounding sensitivity case study of this fire scenario frequency.
The fire zone ignition frequency was modified by applying the sum of items 5, 6, and 7 in Table 6-1 of Reference 1, a location weighting factor of 1.0, and an ignition source weighting factor of 0.02832 based on floor area fraction of the building from Table 3.4.1-14 of Reference
: 3. Both hot shorts and open circuits are assumed to lead to the failure modes of interest in this scenario, so no conditional probability of hot short was applied. The fire non-suppression factor of 1.31 E-02 was increased to 6.90E-02 based on the representative fire non-suppression factor derived in Appendix P of Reference 1. As shown in Table 5, the modified FS3 frequency is 3.42E-04 events per year.Table 5. FS3 Scenario Frequency Case Parameter 1-fl fB RFs3 IE FS3 Original 9.87E-01 8.54E-01 3.22E-03 2.71 E-03 For RAI 9.31 E-01 8.54E-01 4.30E-04 3.42E-04 Fire Scenario 5 (FS5) IE Frequency:
IEFs5 = [(1/8)*fB*fl*(1-f2)
+ (1/8)*fc*fl*(1-f 2)]*RFs 5 where (Table 5-18 of Reference 4)IEFs5 = FS5 initiating event frequency applied in the PRA model (= 2.72E-03 in Table 1), f, = fire damage fraction representing the fraction of fires that affects at least two adjacent cable trays = 1.31 E-02 (from Reference 4), f2 = fire damage fraction representing the fraction of fires that affects more than two cable trays, given two adjacent cable trays are affected = 3.47E-01 (from Reference 4), fB = cable fraction from Table 5-9 of Reference 4 = 8.54E-01, fc = cable fraction from Table 5-9 of Reference 4 = 6.81 E-02, and RFS5 = FS5 fire ignition frequency (from Reference
: 4) = 3.22E-03 events/Rx-year.
Table 6 shows how these values were modified to apply the NUREG/CR-6850 methodology regarding fire zone ignition frequency, hot shorts, and non-suppression factors in an effective Attachment 1 to Enclosure NOC-AE-1 1002773 Page 12 of 21 bounding sensitivity case study of this fire scenario frequency.
The fire zone ignition frequency was modified by applying the sum of items 5, 6, and 7 in Table 6-1 of Reference 1, a location weighting factor of 1.0, and an ignition source weighting factor of 0.02832 based on floor area fraction of the building from Table 3.4.1-14 of Reference
: 3. Both hot shorts and open circuits are assumed to lead to the failure modes of interest in this scenario, so no conditional probability of hot short was applied. The fire non-suppression factor of 1.31 E-02 was increased to 6.90E-02 based on the representative fire non-suppression factor derived in Appendix P of Reference 1. As shown in Table 6, the modified FS5 frequency is 3.42E-06 events per year.Table 6. FS5 Scenario Frequency Case Parameter fl f2 f. fc RFS5 IE FS5 Original 1.31E-02 3.47E-01 8.54E-01 6.81E-02 3.22E-03 3.18E-06 For RAI 6.90E-02 0.OOE+00 8.54E-01 6.81 E-02 4.30E-04 3.42E-06 Fire Scenario 7 (FS7) IE Frequency:
IEFs7 = fo*(1-fl)*RFs7 where (Table 5-16 of Reference 4)IEFS7 = FS7 initiating event frequency applied in the PRA model (= 1.08E-03 in Table 1), f, = fire damage fraction representing the fraction of fires that affects at least two adjacent cable trays = 7.71 E-03 (from Reference 4), fo = cable fraction from Table 5-7 of Reference 4 = 9.07E-01, and RFS7 = FS7 fire ignition frequency (from Reference
: 4) = 1.20E-03 events/Rx-year.
Table 7 shows how these values were modified to apply the NUREG/CR-6850 methodology regarding fire zone ignition frequency, hot shorts, and non-suppression factors in an effective bounding sensitivity case study of this fire scenario frequency.
The fire zone ignition frequency was modified by applying the sum of items 5, 6, and 7 in Table 6-1 of Reference 1, a location weighting factor of 1.0, and an ignition source weighting factor of 0.03163 based on floor area fraction of the building from Table 3.4.1-14 of Reference
: 3. Both hot shorts and open circuits are assumed to lead to the failure modes of interest in this scenario, so no conditional probability of hot short was applied. The fire non-suppression factor of 7.71 E-03 was increased to 6.90E-02 based on the representative fire non-suppression factor derived in Appendix P of Reference 1. As shown in Table 7, the modified FS7 frequency is 4.06E-04 events per year.
Attachment 1 to Enclosure NOC-AE-1 1002773 Page 13 of 21 Table 7. FS7 Scenario Frequency Case Parameter fl fo RFS7 IE FS7 Original 7.71 E-03 9.07E-01 1.20E-03 1.08E-03 For RAI 6.90E-02 9.07E-01 4.81 E-04 4.06E-04 Refined zone fire frequency utilizes Items 5, 6, and 7 from Table 6-1, and floor area percentage for weighting.
Fire Scenario 4 (FS4) IE Frequency:
For fire scenarios 4, 6, and 8, the following general equation applies (see Section 3.4.2.4.2 of Reference 3): IEi = RCR
* FGSi
* FFMi
* FHEi where (Section 3.4.2.4.2 of Reference 3)RCR = the annual frequency of fire of any severity in the fire area associated with fire scenarios 4, 6 and 8.FGSi = the conditional frequency, given a control room fire, of having an impact on the area of interest.
This relative frequency is derived from integration of the product of the geometric factor and the severity factor over the affected area.FFMi = the conditional frequency of the fire causing the combination of failure modes (e.g., hot shorts and open circuits) in the various equipment items that characterize scenario i.FHEi = the conditional frequency, given all the hardware associated with scenario i fails after a fire and the operators fail to conduct the appropriate recovery actions for restoring the required system functions.
IEFs4 = RCR
* FGs18
* FFM18
* FHE18 where RCR = 4.90E-03 (from Section 3.4.2.4.3 of Reference 3), FGS18 = 3.7E-03 (from Section 3.4.2.4.8.8 of Reference 3), FFM18 = 6.7E-01 (from Section 3.4.2.4.8.8 of Reference 3), and FHE18 = Fsc = 2.OE-01 (from Section 3.4.2.4.5 of Reference 3).Table 8 shows how these values were modified to apply the NUREG/CR-6850 methodology regarding fire zone ignition frequency, hot shorts, and non-suppression factors in an effective bounding sensitivity case study of this fire scenario frequency.
The fire zone ignition frequency was modified by applying the 4 in Table 6-1 of Reference 1, a location weighting factor of 1.0, and an ignition source weighting factor of 1.0 based on floor area fraction of the target fire zone.Both hot shorts and open circuits are assumed to lead to the failure modes of interest in this scenario, so no conditional probability of hot short was applied. For this fire area, the fire non-suppression factor is incorporated within the value assigned for FGSi. In this case study, the value of FGs18, 3.7E-03, was increased to 1.38E-02 (= 2.OE-01
* 6.90E-02) based on application of the screening value, 2.OE-01, for the geometric factor from Appendix D of Reference 2 and Attachment 1 to Enclosure NOC-AE-1 1002773 Page 14 of 21 the representative fire non-suppression factor, 6.90E-02, derived in Appendix P of Reference 1 applied as the severity factor. As shown in Table 8, the modified FS 4 frequency is 4.62E-06 events per year.Table 8. FS4 Scenario Frequency Case Parameter RCR FGS18 FFMI8 FHE18 IEFS4 Original 4.90E-03 3.70E-03 6.70E-01 2.OOE-01 2.12E-06 For RAI 2.50E-03 1.38E-02 6.70E-01 2.OOE-01 4.62E-06 Fire Scenario 6 (FS6) IE Frequency:
IEFs6 = RCR
* FGS23
* FFM23
* FHE23 where (Section 3.4.2.4.2 of Reference 3)RCR = 4.90E-03 (from Section 3.4.2.4.3 of Reference 3), FGS23 = 3.2E-03 (from Section 3.4.2.4.8.10 of Reference 3), FFM23 = 7.5E-01 (from Section 3.4.2.4.8.10 of Reference 3), and FHE23 = Fsc = 2.OE-01 (from Section 3.4.2.4.5 of Reference 3).Table 9 shows how these values were modified to apply the NUREG/CR-6850 methodology regarding fire zone ignition frequency, hot shorts, and non-suppression factors in an effective bounding sensitivity case study of this fire scenario frequency.
The fire zone ignition frequency was modified by applying the 4 in Table 6-1 of Reference 1, a location weighting factor of 1.0, and an ignition source weighting factor of 1.0 based on floor area fraction of the target fire zone.Both hot shorts and open circuits are assumed to lead to the failure modes of interest in this scenario, so no conditional probability of hot short was applied. For this fire area, the fire non-suppression factor is incorporated within the value assigned for FGSi. In this case study, the value of FGS23, 3.7E-03, was increased to 1.38E-02 (= 2.OE-01
* 6.90E-02) based on application of the screening value, 2.OE-01, for the geometric factor from Appendix D of Reference 2 and the representative fire non-suppression factor, 6.90E-02, derived in Appendix P of Reference 1 applied as the severity factor. As shown in Table 9, the modified FS 6 frequency is 5.18E-06 events per year.Table 9. FS6 Scenario Frequency Case Parameter RCR FGS2 3  FFM23 FHE23 IEFs6 Original 4.90E-03 3.20E-03 7.50E-01 2.OOE-01 1.61E-06 For RAI 2.50E-03 1.38E-02 7.50E-01 2.OOE-01 5.18E-06 Attachment 1 to Enclosure NOC-AE-1 1002773 Page 15 of 21 Fire Scenario 8 (FS8) IE Frequency:
IEFs8 = RCR
* FGS10
* FFM10
* FHE10 where (Section 3.4.2.4.2 of Reference 3)RCR = 4.90E-03 (from Section 3.4.2.4.3 of Reference 3), FGsl0 = 1.2E-02 (from Section 3.4.2.4.8.4 of Reference 3), FFM10 = 4.5E-01 (from Section 3.4.2.4.8.4 of Reference 3), and FHE10 = Fsc = 2.OE-01 (from Section 3.4.2.4.5 of Reference 3).Table 10 shows how these values were modified to apply the NUREG/CR-6850 methodology regarding fire zone ignition frequency, hot shorts, and non-suppression factors in an effective bounding sensitivity case study of this fire scenario frequency.
The fire zone ignition frequency was modified by applying the 4 in Table 6-1 of Reference 1, a location weighting factor of 1.0, and an ignition source weighting factor of 1.0 based on floor area fraction of the target fire zone.Both hot shorts and open circuits are assumed to lead to the failure modes of interest in this scenario, so no conditional probability of hot short was applied. For this fire area, the fire non-suppression factor is incorporated within the value assigned for FGSi. In this case study, the value of FGSl0, 1.2E-02, was increased to 1.38E-02 (= 2.OE-01
* 6.90E-02) based on application of the screening value, 2.OE-01, for the geometric factor from Appendix D of Reference 2 and the representative fire non-suppression factor, 6.90E-02, derived in Appendix P of Reference 1 applied as the severity factor. As shown in Table 10, the modified FS 8 frequency is 3.11 E-06 events per year.Table 10. FS8 Scenario Frequency Case Parameter RCR FGS23 FFM23 FHE23 IEFS8 Original 4.90E-03 1.20E-02 4.50E-01 2.OOE-01 3.43E-06 For RAI 2.50E-03 1.38E-02 4.50E-01 2.OOE-01 3.11E-06 Attachment 1 to Enclosure NOC-AE-1 1002773 Page 16 of 21 Risk Impact: To calculate the impact of these modified fire scenario initiating event frequencies on plant risk, the RISKMAN 11.2 computer code was applied to save the STP_REV6 PRA model as a model named SAMAFIRE.
In the SAMAFIRE model, the fire scenario initiating event frequencies were replaced with the values presented herein. All point estimate initiating event groups were then quantified, using the point estimate method. A new point estimate master frequency file named SFIREPE was created. The new SFIREPE master frequency file and the modified initiating events were exported to the event tree quantification module of RISKMAN, and the SAMAFIRE model was quantified in two separate batch point estimate quantification runs, one designated LVL1 for Level 1 (yielding CDF) and one designated LVL2 for Level 2 (yielding modified frequencies of Release Categories I (LERF), II, Ill, and IV). The results of this analysis are presented in Tables 11 and 12 below, which can be compared with the results shown in Tables 1 and 2 for the original STP_REV6 PRA model.Table 11. SAMAFIRE Level I PRA Fire Scenario Results Summary (for CDF Contribution)
Fire Scenario IE Freg. CDF %CDF CCDP 1 2.95E-05 8.04E-07 10.62 2.73E-02 2 1.04E-06 9.43E-07 12.46 9.07E-01 3 3.42E-04 2.23E-08 0.29 6.52E-05 4 4.62E-06 2.67E-07 3.53 5.78E-02 5 3.42E-06 6.88E-08 0.91 2.01 E-02 6 5.18E-06 8.42E-08 1.11 1.63E-02 7 4.06E-04 4.OOE-09 0.05 9.84E-06 8 3.11E-06 9.24E-10 0.01 2.97E-04 All Fire Initiating Events in the PRA: 7.95E-04 2.19E-06 28.99 2.76E-03 All Initiating Events in the PRA: 2.32E+00 7.57E-06 100 3.26E-06 Table 12. SAMAFIRE Level 2 PRA Fire Scenario Results Summary (by Release Category Contribution)
RELI Freq. RELII Fire Scenario IE Freg. (LERF) Freg. RELIII Freg. RELIV Freg.1 2.95E-05 4.11E-08 3.91E-07 1.59E-07 2.09E-07 2 1.04E-06 2.01E-08 7.96E-07 5.44E-08 7.14E-08 3 3.42E-04 3.68E-10 9.06E-09 1.99E-09 8.05E-09 4 4.62E-06 2.09E-08 1.04E-07 6.04E-08 7.96E-08 5 3.42E-06 2.86E-09 4.56E-08 8.78E-09 1.04E-08 6 5.18E-06 1.05E-09 2.07E-10 1.70E-09 7.97E-08 0.OOE+0 7 4.06E-04 0 2.29E-11 1.39E-09 1.98E-09 0.OOE+0 8 3.11E-06 1.02E-11 0 8.69E-12 7.80E-10 All Fire Initiating Events in the PRA: 7.95E-04 8.63E-08 1.35E-06 2.87E-07 4.61 E-07 All Initiating Events in the PRA: 2.32E+00 5.46E-07 1.97E-06 1.61 E-06 3.29E-06 It is important to note that the values in Tables 11 and 12, as those in Tables 1 and 2, are rounded. The information in Tables 11 and 12 can be applied to support refined SAMA cost-benefit analyses, as desired. The results show that total CDF and LERF increased slightly with the modified fire initiating event frequencies applied in the SAMAFIRE model.
Attachment 1 to Enclosure NOC-AE-1 1002773 Page 17 of 21 Subsequent to the first application of fire scenario updates described herein, it was determined that, for purposes of evaluating seismic model updates, a general model truncation level of 1.OOE-14 was preferred over the standard 1.OOE-12 truncation level applied in the STPREV6 and SAMAFIRE models. Therefore, the STPREV6 model was saved as a new model named STPR6T14, and this model was quantified applying an event tree module initiating event global cutoff (truncation) level of 1.OOE-14 to yield the results presented in Tables 13 and 14. Tables 13 and 14 present the same information presented in Tables 1 and 2, but now applying the baseline risk model truncated at 1.OOE-14 instead of 1.OOE-12.Table 13. STPR6T14 Level I PRA Fire Scenario Results Summary (for CDF Contribution)
Fire Scenario IE Freq. CDF %CDF CCDP 1 1.46E-05 3.99E-07 6.17 2.73E-02 2 2.34E-07 2.12E-07 3.29 9.07E-01 3 2.72E-03 1.86E-07 2.89 6.85E-05 4 2.12E-06 1.23E-07 1.90 5.78E-02 5 3.18E-06 6.43E-08 1.00 2.02E-02 6 1.61 E-06 2.64E-08 0.41 1.64E-02 7 1.08E-03 1.15E-08 0.18 1.07E-05 8 3.43E-06 1.07E-09 0.02 3.13E-04 All Fire Initiating Events in the PRA: 3.83E-03 1.02E-06 15.85 2.67E-04 All Initiating Events in the PRA: 2.33E+00 6.46E-06 100 2.77E-06 Table 14. STPR6T14 Level 2 PRA Fire Scenario Results Summary (by Release Category Contribution)
RELI REI RELII RELIII Fire Scenario IE Freq. Freq. req. req. RELIV Freq.(LERF) Freq. Freq, 1 1.46E-05 2.04E-08 1.95E-07 7.94E-08 1.04E-07 2 2.34E-07 4.55E-09 1.79E-07 1.23E-08 1.61 E-08 3 2.72E-03 4.22E-09 8.11E-08 2.37E-08 7.62E-08 4 2.12E-06 9.68E-09 4.81E-08 2.80E-08 3.67E-08 5 3.18E-06 2.73E-09 4.30E-08 8.56E-09 9.93E-09 6 1.61E-06 3.46E-10 8.16E-11 5.85E-10 2.53E-08 7 1.08E-03 1.51E-11 2.31E-10 4.77E-09 6.32E-09 8 3.43E-06 1.35E-11 2.72E-12 2.16E-11 1.02E-09 All Fire Initiating Events in the PRA: 3.83E-03 4.20E-08 5.46E-07 1.57E-07 2.76E-07 All Initiating Events in the 2.33E+00 PRA: 5.13E-07 1.20E-06 1.54E-06 3.18E-06 Similarly, the SAMAFIRE model was saved as a new model named SAMAFR14, and this model was quantified applying an event tree module initiating event global cutoff (truncation) level of 1.OOE-14 to yield the results presented in Tables 15 and 16. Tables 15 and 16 present the same information presented in Tables 11 and 12, but now applying the baseline risk model truncated at 1.OOE-14 instead of 1.OOE-12.
Attachment 1 to Enclosure NOC-AE-1 1002773 Page 18 of 21 Table 15. SAMAFR14 Level 1 PRA Fire Scenario Results Summary lfor CDF Contribution)
Initiating Event Designator IE Freq. CDF %CDF CCDP 1 2.95E-05 8.05E-07 10.55 2.73E-02 2 1.04E-06 9.43E-07 12.36 9.07E-01 3 3.42E-04 2.34E-08 0.31 6.83E-05 4 4.62E-06 2.67E-07 3.50 5.79E-02 5 3.42E-06 6.91 E-08 0.91 2.02E-02 6 5.18E-06 8.48E-08 1.11 1.64E-02 7 4.06E-04 4.32E-09 0.06 1.06E-05 8 3.11E-06 9.74E-10 0.01 3.13E-04 All Fire Initiating Events in the PRA: 7.95E-04 2.20E-06 28.80 2.77E-03 All Initiating Events in the PRA: 2.32E+00 7.63E-06 100 3.29E-06 r -Table 16. SAMAFR14 Level 2 PRA Fire Scenario Results Summary (by Release Category Contribution)
Initiating Event RELI Freq. RELII RELIII Designator IE Freq. (LERF) Freq. Freq. RELIV Freq.1 1.60E-2.95E-05 4.13E-08 3.93E-07 07 2.10E-07 2 5.46E-1.04E-06 2.02E-08 7.97E-07 08 7.15E-08 3 2.88E-3.42E-04 5.14E-10 1.01E-08 09 9.46E-09 4 6.12E-4.62E-06 2.11E-08 1.05E-07 08 8.01E-08 5 9.21E-3.42E-06 2.93E-09 4.62E-08 09 1.07E-08 6 1.90E-5.18E-06 1.12E-09 2.70E-10 09 8.13E-08 7 1.77E-4.06E-04 4.59E-12 8.06E-11 09 2.35E-09 8 1.95E-3.11E-06 1.22E-11 2.47E-12 11 9.22E-10 All Fire Initiating Events 2.92E-in the PRA: 7.95E-04 8.72E-08 1.35E-06 07 4.66E-07 All Initiating Events in the 1.68E-PRA: 2.32E+00 5.59E-07 2.01E-06 06 3.37E-06 Tables 13 through 16 are provided for comparison with results developed for the seismic model updates performed to respond to NRC RAI item 3.c. It is important to note that the results presented in Tables 11 through 16 are rounded.
Attachment 1 to Enclosure NOC-AE-1 1002773 Page 19 of 21 SAMA Impact: The updated fire model (SAMAFR14) was then used to update the SAMA analysis in order to determine how its use would impact the results of that analysis.
The assessment included the following steps:* Update the STP Maximum Averted Cost-Risk (MACR) using the new fire results,* Check the importance review threshold for the overall SAMA identification process," Determine if any new, potentially cost beneficial SAMAs can be identified,* Update the Phase I analysis,* Update the Phase II analysis.Note: The combined impact of the updated fire and seismic results are examined in the response to RAI 3.c. (See Attachment 2 of this Enclosure)
The updated STP MACR was calculated in the same manner as described in Section F.4 of the STP Environmental Report (ER) using the total CDF of 7.63E-06 from Table 15 and the release category frequencies for "all initiating events" from Table 16. The result is a single unit MACR of$318,000 and $636,000 MACR for the site, which is $118,000 greater than the $518,000 site MACR used in the ER.Because the importance review threshold was artificially reduced for STP in order to provide a more robust review of the PRA results, the increase in the MACR does not result in a reduction of the review threshold below that which was used in the ER. The minimum SAMA implementation cost of $100,000 for the site correlates to a risk reduction worth (RRW) value of about 1.18, which is significantly higher than the RRW threshold of 1.022 that was applied in the ER. The overall SAMA identification process, therefore, has not been updated to address these fire modeling changes as it could not capture any new, non-fire related events.The importance list review documented in the ER includes many fire related split fractions and SAMAs were identified to address several of the higher contributing fire scenarios, including fire scenarios 1, 2, 3 and 5. The change in fire scenario frequencies based on this update does not impact the design of the SAMAs that were proposed to mitigate them and no new SAMAS are required.
For the fire scenarios that were not explicitly addressed by the SAMA list (due to low contribution), the revised fire results indicate that the potential averted cost-risk (PACR) values associated with these scenarios are still less than the minimum SAMA implementation cost and no new SAMAs are warranted.
For fire scenarios 4, 6, 7, and 8, the PACR values were calculated using the method described in Section F.4 of the ER, the CDF for each fire scenario in Table 15, and the release category frequencies from Table 16. As can be seen in Table 17, the largest PACR for these scenarios is fire scenario 4 at $24,000 (for the site).Table 17. PACR Values for Fire Scenarios Not I Explicitly Addressed by the SAMA List I Initiating Event Description Baseline PACR (site)FS4 $24,000 FS6 $6,000 FS7 <$2,000 FS8 <$2,000 Attachment 1 to Enclosure NOC-AE-1 1002773 Page 20 of 21 While the updated fire results do not result in the introduction of any new SAMAs, the Phase I analysis may be impacted due to the change in the MACR. As identified above, the site MACR increased from $518,000 to $636,000 for the base case. When the 9 5 th percentile PRA results are applied, this value becomes $1,717,200 (using the multiplier of 2.7 from the response to RAI ld (
 
==Reference:==
 
STPNOC response to NRC Requests for Additional Information for the South Texas Project License Renewal Application dated August 23, 2011 [NOC-AE-1 1002711][ML1 1250A067]).
Some of the SAMAs (i.e., 3b, 7, 8, 9, 11, and 16) that were screened in the Phase I analysis of the ER can no longer be screened when the updated 95th percentile MACR is applied and are transferred to the Phase II screening process for further evaluation.
One method of screening SAMAs in the Phase II analysis is to perform a PRA model run to support a detailed cost benefit analysis for the SAMAs. However, PRA insights can also be used to demonstrate that the SAMAs would not be cost beneficial.
This type of screening was performed for these same SAMAs as part of the response to the follow up RAI on question 1 .d, which considered a 95th percentile MACR of $2.4 million. The process is performed for SAMAs 7, 8, 9, 11 and 16 for Case 2 of the follow up response to RAI 3.c (integrated fire and seismic results) as the results of that screening evaluation will bound those for the SAMAFR14 model.Due to the difficulty associated with characterizing the potential cost benefit of SAMA 3b using only PRA insights, its averted cost-risk was determined using a PRA model quantification along with the original Phase II SAMAs from the Environmental Report (see Tables 18 and 19 below).In order to determine how the fire scenario initiating event frequency updates would impact the STP Phase II SAMA quantifications, each of the quantifications was updated to include the new fire scenario initiating event frequency data. Table 18 documents the PRA results for the updated quantifications.
Table 18: Phase II Quantification Results (SAMAFR14 model)Case Identifier CDF REL I REL II REL III REL IV BASESAMAFR14 7.63E-06 5.59E-07 2.01 E-06 1.68E-06 3.37E-06 SAMA 3b 7.60E-06 5.59E-07 1.97E-06 1.68E-06 3.37E-06 SAMA 4 7.51 E-06 4.33E-07 2.01 E-06 1.68E-06 3.37E-06 SAMA 10 7.63E-06 5.50E-07 2.02E-06 1.53E-06 3.52E-06 SAMA 12 7.63E-06 5.56E-07 2.01E-06 1.68E-06 3.37E-06 SAMA 13 7.61 E-06 5.57E-07 2.OOE-06 1.67E-06 3.36E-06 SAMA 15 7.55E-06 5.53E-07 1.99E-06 1.65E-06 3.33E-06 Attachment 1 to Enclosure NOC-AE-1 1002773 Page 21 of 21 Using the 2.7 uncertainty multiplier from the response to RAI ld and the results from Table 18, the cost benefit analysis was updated to determine if the use of the updated fire scenario initiating event frequencies would impact the conclusions of the SAMA analysis.
As documented in Table 19 below, the conclusions of the analysis did not change: Table 19: Cost Benefit Analysis Results for Phase II SAMAs Supported by Detailed PRA Quantifications (SAMAFR14 Model, 9 5 th Percentile PRA Results)fTotal Change in SAMA ID Cost ofToa SAMA I D Averted Net Value Status?Implementation Cost-Risk SAMA 3b $796,677 $8,905 -$787,772 No SAMA 4 $100,000 $70,621 -$29,379 No SAMA 10 $100,000 $9,828 -$90,172 No SAMA 12 $100,000 $4,147 -$95,853 No SAMA 13 $100,000 $5,346 -$94,654 No SAMA 15 $100,000 $18,932 -$81,068 No
 
==References:==
: 1. Electric Power Research Institute and U. S. Nuclear Regulatory Commission, "EPRI/NRC-RES Fire PRA Methodology for Nuclear Power Facilities," Final Report, EPRI 1011989, NUREG/CR-6850, Volumes 1 and 2, September 2005.2. Pickard, Lowe and Garrick, Inc., "South Texas Project Probabilistic Safety Assessment," PLG-0675, prepared for the Houston Lighting & Power Company, May 1989.3. Houston Lighting & Power Company with Assistance from PLG, Inc., "South Texas Project Electric Generating Station Level 2 Probabilistic Safety Assessment and Individual Plant Examination," August 1992.4. PLG, Inc., "Fire Analysis Update for the South Texas Project Electrical Generating Station PSA for Selected Fire Zones," PLG-1 015, December 1994.5. STP Nuclear Operating Company, "Risk-Managed Technical Specifications (RMTS)Implementation Fire Scenario Risk Sensitivity Case Studies," PRA Analysis/Assessment Report PRA-06-018, March 2007.
Attachment 2 to Enclosure NOC-AE-1 1002773 Attachment 2 Seismic Impact Assessment Attachment 2 to Enclosure NOC-AE-1 1002773 Page 1 of 29 Seismic Impact Assessment Note: The references for this assessment report are provided on page 4 of this attachment.
Discussion The model of record (STPREV6) was modified to evaluate the effects of assuming the 2008 U.S. Geological Survey (USGS) and Lawrence Livermore National Laboratory (LLNL) hazard curves on the seismic contribution to core damage frequency (CDF) and release category frequencies.
The 1x1012 per year truncation limits used in the model of record were lowered to lx104 per year for this entire analysis to ensure that the contribution from seismic events is converged even with the higher seismic hazard curve frequencies.
The sum of the 4 seismic initiator contributors to CDF in the model of record (STP REV6) when quantified using lx10-14 per year as the truncation value is 7.53x1 0 8/year. Seismic initiators contribute just 1.1 percent to the total CDF of 6.46x10.6 per year.Prior to evaluating the impact of the USGS and LLNL hazard curves on CDF, the seismic model of record (STPREV6) was modified.
The modifications included: 1) Increasing the number of seismic initiators from 4 to 6 so as to extend the range of accelerations beyond 0.8g for comparison with the USGS and LLNL hazard curves (the USGS curve extends to 2. lg), and 2) The elimination of a sequence-specific recovery term for manually starting the turbine-driven auxiliary feedwater (AFW) pump that was being non-conservatively applied for seismic initiating events only in the STPREV6 model. This modification was conservative for the SAMA assessment.
The revised contribution of seismic initiators assuming the same Electric Power Research Institute (EPRI) hazard curve (Reference 1), but now extrapolated to 2.1g, is 8.80x10-8/year; or approximately 1.3E-08 higher than the seismic CDF in the original analyses.The 2008 USGS hazard curve data for STP was extracted from the collection of Peak Ground Acceleration (PGA) data posted by the USGS webpage titled "2008 NSHM Gridded Data". The following URL points to the web page from which the PGA data file "2008.US.pga.txt.gz" were downloaded:
http://earthquake.ussis.qov/hazards/products/conterminous/2008/data/#fileformat.
The "2008.US.pga.txt.gz" file contains a series of peak ground acceleration values paired with frequency of exceedance organized by longitude and latitude in 0.050 increments for the conterminous 48 States. The coordinates for STP are latitude 28.79560 and longitude-96.04890.
The two closest coordinates found in the PGA data file were for latitude 28.80 longitude
-96.050, and latitude 28.750 longitude
-96.00. The values for these two hazard exceedance curves are very close. The final seismic hazard curve was generated by a linear interpolation between the frequency of exceedance values for these two sets of coordinates.
Figure 1 below shows that the USGS 2008 hazard curve exceedance frequency extends to 2.13g and is still above 10-7 per year. By contrast, the EPRI curves from 1989 terminated at Attachment 2 to Enclosure NOC-AE-1 1002773 Page 2 of 29 0.8g with an exceedance frequency of only 108 per year. Meanwhile the arithmetic mean of the LLNL hazard curve for STP from 1989 (Reference
: 2) showed an exceedance frequency of 1.1x10-6 at 1.0g. The LLNL hazard curve is higher than the 2008 USGS curve at the lower acceleration levels but this relationship changes as the magnitude increases.
At 1.0g, the exceedance frequency on the 2008 USGS Curve (1.5xl0 6/year) is actually marginally higher than that on the LLNL curve (1.1x10 6/year).Table 1 compares the effect on seismically contributed CDF and on total CDF using the original component fragility curves but varied the selection of the hazard curve. The effects on the frequencies for the 4 major release categories used in the SAMA analysis are also listed. The results for the model of record are also provided.
All percentage increases shown are by comparison to the model of record results. When either the 2008 USGS or LLNL hazard curves are assumed, the resulting seismic contribution to CDF is increased from that when assuming the EPRI hazard curve. For the 2008 USGS hazard curve, the seismic CDF contribution is 4.65x10 6 per year which is about a 71% increase compared to the model of record in the total CDF from all initiators.
For the LLNL hazard curve, the seismic CDF contribution is then 1.48x1 05 per year, resulting in a 228% increase in the total CDF from all initiators.
The LLNL hazard curve leads to a higher CDF than the 2008 USGS curve primarily from the seismic acceleration range of 0.3 to 0.8g, as expected.
The large early release frequencies also increase for both the USGS and LLNL hazard curves, but by smaller percentages than does the total CDF.Table 2 provides a comparison with the model of record after incorporating revised seismic fragility curves for selected components.
These changes to the fragility curves resulted from a recent review of the calculations used to substantiate the original fragilities and by updating them to the extent possible.
The calculations were updated in accordance with EPRI guidance (References 3 through 6). A plant walkdown was also conducted as part of this recent fragility update to observe the short list of components whose fragility curves were revised. The recently evaluated fragility curves are documented in Table 5.It was observed that when evaluating the fragility curves to be applied to all three trains of identical equipment, that the model of record evaluated the fragility curves based on the component mounted highest in the building.
For the STP three train plant, this was found to be excessively conservative because the identical, redundant components were represented by a single fragility curve. The components represented by the fragility curve were then assumed to fail all three trains of the components, even the redundant trains are often located at different elevations within the same building.Therefore for this study, a change was made to the way in which these bounding fragility curves were evaluated and then applied in the risk model. Instead of for the identical, redundant components mounted at the highest elevation, the fragility curve for three train systems was evaluated for the "B" train component which is at the middle of the three elevations.
The seismic failure probabilities of the "A" and "B" trains were then modeled as failing jointly according to the new fragility curves. This is still conservative with respect to the failure of the lower elevation "A" train. During all seismic events, no credit was then taken for any of the highest elevation train "C" equipment (i.e. "C" train equipment was assumed failed for all seismic initiators).
The fragility curves for the inverters and battery chargers, batteries and racks, essential cooling water (ECW) pumps, and 4kv switchgear were evaluated for the next highest train (train B) in this way. All other fragilities included in the model impact only a single Attachment 2 to Enclosure NOC-AE-1 1002773 Page 3 of 29 train of equipment or at most two trains of equipment.
The assumption of evaluating the components at the higher elevation of the two trains of redundant equipment was retained for the two train systems.Table 2 compares the effect on seismically contributed CDF and on total CDF using the revised component fragility curves; including the failed train "C" approach, together with varying the hazard curve selected.
The effects on the frequencies for the 4 major release categories used in the SAMA analysis are also listed. Compared to the model of record (i.e. STPREV6), the revised model assuming the EPRI hazard curves, shows that the total CDF and total release category frequencies are slightly lower because of the changed fragility curves and modeling changes. When either the 2008 USGS or LLNL hazard curves are assumed, the resulting seismic CDF is increased from that when assuming the EPRI hazard curve. For the 2008 USGS hazard curve, the seismic CDF contribution is 3.01x10 6 which results in about a 46%increase in the total CDF from all initiators.
For the LLNL hazard curve, the seismic CDF contribution is 7.59x10-6 per year, or a 116% increase in the total CDF from all initiators as compared to the model of record. The large early release frequencies also increase for both the USGS and LLNL hazard curves but by smaller percentages than does the total CDF.While updating the selected seismic fragility curves, it was observed that the original fragility curves involved rather large Beta-R and Beta-U values and that most of this uncertainty comes from the response spectra shape and soil-structure interaction.
It is suspected that some double counting of the uncertainty associated with the soil stiffness variation in developing the spectral shape factor and the factor for soil-structure interaction (SSI) may have played a role in the calculation of high Beta factors. If a new state of the art building response analysis were to be performed, we would expect to see much smaller Beta values resulting for the components analyzed.
EPRI 1002988 (Reference
: 5) states that "For older plants on soil sites it is especially important that the structural response be recomputed." This is due to the fact that the earlier SSI analysis used simple soil spring models that were not capable of capturing the complex behavior of structural response on layered soil sites. Therefore, two sensitivity cases were performed to study the effects of lower Beta values for the fragility curves on the total CDF.Table 3 shows the results comparable to the results in Table 2 but with the Beta values set to 0.3 for all components other than for the fragility curve representing the loss of power from the offsite grid. Beta values of around 0.3 are more typical of values for Beta-r and Beta-u found in modern studies for seismic risk analysis.
Compared to the model of record (i.e. STPREV6), which assumes the EPRI hazard curves limited to 0.8g, the total CDF and total release category frequencies for the extended EPRI hazard curve to 2. lg are again slightly lower because of the changed fragility curves and modeling changes. When either the 2008 USGS or LLNL hazard curves are assumed, the resulting seismic CDF increased from that when assuming the EPRI hazard curve. For the 2008 USGS hazard curve, the seismic CDF contribution is 2.10x10-6 which results in about a 31 percent increase in the total CDF from all initiators.
For the LLNL hazard curve, the seismic CDF contribution is 3.95x106 per year, or a 60 percent increase in the total CDF from all initiators.
These increases in the total CDF are about two-thirds of that found for the revised fragility curves in Table 2. The large early release frequencies (LERF) also increase for both the USGS and LLNL hazard curves but by smaller percentages than those presented in Table 2. The conclusion from this sensitivity case is that by assuming all beta values are 0.30, the CDF and release category frequency increases are about two thirds the increases reported in Table 2 for the revised fragilities.
Attachment 2 to Enclosure NOC-AE-1 1002773 Page 4 of 29 Table 4 repeats the sensitivity case described in Table 3, but assumes Beta values changed to 0.35. Compared to the model of record (i.e. STPREV6), which assumed the EPRI hazard curves were limited to 0.8g, the total CDF and total release category frequencies assumed the EPRI curves extend to 2.1g and are slightly lower because of the changed fragility curves and modeling changes. When either the 2008 USGS or LLNL hazard curves are assumed, the resulting seismic CDF is also increased from that when assuming the EPRI hazard curve. For the 2008 USGS hazard curve, the seismic CDF contribution is 2.41x10-6 which results in about a 36 percent increase in the total CDF from all initiators.
For the LLNL hazard curve, the seismic CDF contribution is 5.06x10 6 per year or a 77 percent increase in the total CDF from all initiators.
These increases in the total CDF are about four fifths for the USGS curves and two-thirds for the LLNL curves as compared to the increases found for the revised fragility curves in Table 2. The large early release frequencies also increase for both the USGS and LLNL hazard curves but by a smaller percentage for the USGS curve than does the total CDF and about the same percentage increase as for total CDF for the LLNL curves.References
: 1. EPRI NP-6395-D, "Probabilistic Seismic Hazard Evaluations at Nuclear Plant Sites in the Central and Eastern United States: Resolution of the Charleston Earthquake Issue," April 1989 2. Lawrence Livermore National Laboratory, "Seismic Hazard Characterization of 69 Nuclear Plant Sites East of the Rocky Mountains," NUREG/CR-5250, Volume 5.3. EPRI NP-6041, "A Methodology for Assessment of Nuclear Power Plant Seismic Margin," Electric Power Research Institute, October 1988.4. EPRI TR-103959, "Methodology for Developing Seismic Fragilities," June 1994.5. EPRI Report 1002988, "Seismic Fragility Application Guide," December 2002.6. EPRI Report 1019200, "Seismic Fragility Application Guide Update," December 2009.7. Diablo Canyon Power Plant Environmental Report; Operating License Renewal Stage, Appendix E, Attachment F -Severe Accident Mitigation Alternatives, November 2009.
Attachment 2 to Enclosure NOC-AE-1 1002773 Page 5 of 29 Figure 1 Comparison of Seismic Hazard Exceedance Curves for STP Site (EPRI, 2008 USGS, and LLNL)Co parisono Ha ardCu vesfor TP 1.000E-03 4 ._1.0002-04
--_, _ = _ USG5 , 1.000E-05 Exceedance Frequency 1.00-0 --_ ......*.. .__(per year)1.OOOE-0 I " 000E-09 I 1.OOOE-10 1 -0 0 0 2r-1 1 1- _ _1.OOOE-12 PGA in g 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.5 1.8 2.0 2.2 Attachment 2 to Enclosure NOC-AE-1 1002773 Page 6 of 29 Table I -Comparison of Seismic Core Damage Frequencies and Seismic Release Category Frequencies (events per reactor calendar year) Assuming Revised Hazard Curves and Orininal Fraailitias CDF REL. CAT. I REL. CAT. REL. CAT. REL. CAT.(Total) (Large II III (Late IV Early (Small Release) (No Release) Early Release)Release)STPREV6 7.53E-08 3.95E-09 3.08E-08 1.61 E-08 2.37E-08 Seismic Contribution STPREV6 6.46E-06 5.13E-07 1.20E-06 1.54E-06 3.18E-06 Total Frequency EPRI Hazard Curves To 0.8g, Extrapolated to 2.1g Seismic 8.80E-08 4.62E-09 3.72E-08 1.84E-08 2.68E-08 Frequency/
New Total 6.47E-06 5.14E-07 1.21 E-06 1.54E-06 3.18E-06 Frequency/
% Increase in 0.20% 0.13% 0.53% 0.15% 0.10%Total USGS Hazard Curves; To 2.1g, Seismic 4.65E-06 2.25E-07 2.56E-06 7.83E-07 1.07E-06 Frequency/
New Total 1.10E-05 7.35E-07 3.73E-06 2.31 E-06 4.23E-06 Frequency/
% Increase in 70.80% 43.12% 209.92% 49.75% 33.05%Total LLNL Hazard Curves To 1.0g, Extrapolated to 2.1g Seismic 1.48E-05 8.02E-07 6.79E-06 2.99E-06 4.23E-06 Frequency/
New Total 2.12E-05 1.31E-06 7.96E-06 4.51E-06 7.38E-06 Frequency/
% Increase in 228.25% 155.33% 560.76% 192.76% 132.31%Total Attachment 2 to Enclosure NOC-AE-1 1002773 Page 7 of 29 Table 2 -Comparison of Seismic Core Damage Frequencies and Seismic Release Category Frequencies (per calendar year) Assuming Revised Hazard Curves and Updated Fragilities CDF REL. CAT. I REL. CAT. II REL. CAT. III REL. CAT. IV (Total) (Large Early (Small Early (Late Release) (No Release)Release) Release)STPREV6 7.53E-08 3.95E-09 3.08E-08 1.61 E-08 2.37E-08 Seismic Contribution STPREV6 6.46E-06 5.13E-07 1.20E-06 1.54E-06 3.18E-06 Total Frequency EPRI Hazard Curves To 0.8g, Extrapolated to 2.1g Seismic 4.77E-08 3.65E-09 1.62E-08 1.12E-08 1.58E-08 Frequency/
New Total 6.43E-06 5.13E-07 1.19E-06 1.54E-06 3.17E-06 Frequency/
% Increase in -0.43% -0.06% -1.21% -0.32% -0.25%Total USGS Hazard Curves; To 2.1g, Seismic 3.01 E-06 1.74E-07 1.52E-06 5.56E-07 7.55E-07 Frequency/
New Total 9.40E-06 6.83E-07 2.70E-06 2.08E-06 3.91 E-06 Frequency/
% Increase in 45.51% 33.09% 124.04% 35.00% 23.02%Total LLNL Hazard Curves To 1.0g, Extrapolated to 2.1g Seismic 7.59E-06 5.34E-07 3.OOE-06 1.69E-06 2.36E-06 Frequency/
New Total 1.40E-05 1.04E-06 4.17E-06 3.22E-06 5.51 E-06 Frequency/
% Increase in 116.34% 103.17% 246.29% 108.75% 73.41%Total Attachment 2 to Enclosure NOC-AE-1 1002773 Page 8 of 29 Table 3 -Comparison of Seismic Core Damage Frequencies and Seismic Release Category Frequencies (per calendar year) Assuming Revised Hazard Curves and Updated Fragilities (Sensitivitv Case with Beta's set to 0.31 CDF REL. CAT. I REL. CAT. II REL. CAT. III REL. CAT. IV (Total) (Large Early (Small Early (Late Release) (No Release)Release) Release)STPREV6 7.53E-08 3.95E-09 3.08E-08 1.61 E-08 2.37E-08 Seismic Contribution STPREV6 6.46E-06 5.13E-07 1.20E-06 1.54E-06 3.18E-06 Total Frequency EPRI Hazard Curves To 0.8g, Extrapolated to 2.1g Seismic 2.69E-08 2.53E-09 7.35E-09 7.08E-09 9.53E-09 Frequency/
New Total 6.41E-06 5.12E-07 1.18E-06 1.53E-06 3.16E-06 Frequency/
% Increase In -0.75% -0.28% -1.95% -0.59% -0.45%Total USGS Hazard Curves; To 2.1g, Seismic 2.1 OE-06 1.33E-07 1.02E-06 4.01 E-07 5.38E-07 Frequency/
New Total 8.48E-06 6.42E-07 2.20E-06 1.93E-06 3.69E-06 Frequency/
% Increase in 31.34% 25.04% 82.53% 24.97% 16.18%Total LLNL Hazard Curves To 1.0g, Extrapolated to 2.1g Seismic 3.95E-06 3.35E-07 1.34E-06 9.66E-07 1.30E-06 Frequency/
New Total 1.03E-05 8.44E-07 2.52E-06 2.49E-06 4.45E-06 Frequency/
% Increase in 59.97% 64.41% 108.89% 61.65% 40.14%Total Attachment 2 to Enclosure NOC-AE-1 1002773 Page 9 of 29 Table 4 -Comparison of Seismic Core Damage Frequencies and Seismic Release Category Frequencies (per calendar year) Assuming Revised Hazard Curves and Updated Fragilities ISensitivtv Case with Beta's set tn 0.351 CDF REL. CAT. I REL. CAT. II REL. CAT. Ill REL. CAT. IV (Total) (Large Early (Small Early (Late Release) (No Release)Release) Release)STPREV6 7.53E-08 3.95E-09 3.08E-08 1.61 E-08 2.37E-08 Seismic Contribution STPREV6 6.46E-06 5.13E-07 1.20E-06 1.54E-06 3.18E-06 Total Frequency EPRI Hazard Curves To 0.8g, Extrapolated to 2.lg Seismic 3.29E-08 2.97E-09 9.55E-09 8.43E-09 1.14E-08 Frequency/
New Total 6.41 E-06 5.13E-07 1.18E-06 1.53E-06 3.17E-06 Frequency/
% Increase in -0.66% -0.19% -1.76% -0.50% -0.39%Total USGS Hazard Curves; To 2.1g, Seismic 2.41E-06 1.51E-07 1.17E-06 4.60E-07 6.18E-07 Frequency/
New Total 8.79E-06 6.60E-07 2.35E-06 1.99E-06 3.77E-06 Frequency/
% Increase in 36.09% 28.61% 94.90% 28.79% 18.70%Total LLNL Hazard Curves To 1.0g, Extrapolated to 2.1g Seismic 5.06E-06 4.16E-07 1.78E-06 1.22E-06 1.64E-06 Frequency/
New Total 1.14E-05 9.25E-07 2.95E-06 2.74E-06 4.80E-06 Frequency/
% Increase in 77.18% 80.16% 145.01% 77.82% 51.01%Total Attachment 2 to Enclosure NOC-AE-1 1002773 Page 10 of 29 Table 5 -Comparison of Original and Revised Fragility Curves for STP Component BldglEl. Original Update V) (wloriginal structural Comments Name Betas Am beta-RI beta-U HCLPF Am I beta-R I beta-U HCLPF Inverters and Battery Chargers EAB/1 0'1.31 0.41 0.52 0.28 1.9 0.37 0.51 0.44 Inverters
& Battery EAB/35' 1.31 0.41 0.52 0.28 1.8 0.37 0.51 0.42 Chargers Inverters
& Battery Chargers EAB/60'1.31 0.41 0.52 0.28 0.98 0.37 0.51 0.23 Walkdown and the associated anchorage design calculations verified that the anchorages of these cabinets are not anticipated to govern the fragility.
The revised fragilities shown to the left, which are based on a functionality test of the inverters, are reasonable estimates at this time. The revised fragilities utilized the methodology in EPRI TR-103959.
The original capacity of the Fuel Oil tanks (FOT) was assumed to be the same as the calculated capacity for the CCW tank. The CCW tank is a horizontal tank on two steel saddles at El. 60' of MEAB, and the FOT is a flat bottom steel tank located at elevation 55' of the DGB. Therefore, assigning the capacity of surge tank to FOT is not a reasonable assumption.
The new estimated capacity (based on guidelines of EPRI NP-6041 and TR-103959) shown to the left is governed by the tank's available sloshing height.The tank HCLPF capacity based on tank shell buckling or tank sliding is estimated to be about 0.9g with median capacity of 3.73g, and therefore is not limiting.Diesel Fuel Oil Tank DGB/55'1.09 0.42 0.51 0.23 1.1 0.35 0.44 0.30 125v dc batteries 1 041 0.53 0.40 33 0.37 0.54 0.74 original capacity of the battery/rack and racks I ___ 0 1.91 0 __ 1 0.7 is based on dynamic testing of the Attachment 2 to Enclosure NOC-AE-1 1002773 Page 11 of 29 Table 5 -Comparison of Original and Revised Fragility Curves for STP Component BldgIEl. Original Update') (wloriginal structural Comments Name Betas Am beta-R beta-U HCLPF Am "beta-R beta-U HCLPF 125v dc batteries EAB/35' 1.91 0.41 0.53 0.40 3.3 0.37 0.54 0.74 batteries
& racks. The updated and racks fragilities utilize the guidelines of EPRI 125vdc batteries E1.TR-1 03959 for estimating capacities 125nd rat s EAB/60' 1.91 0.41 0.53 0.40 1.82 0.37 0.54 0.41 based on dynamic testing of andracks equipment.
The updated fragilities are based on 4KV SWGR EAB/10' 1.69 0.41 0.56 0.34 3.14 0.37 0.51 0.74 dynamic testing of the switchgears and utilize the guidelines provided in EPRI 4KV SWGR EAB/35' 1.69 0.41 0.56 0.34 3.14 0.37 0.51 0.74 TR-103959.
Simple hand calculations I_ confirmed that the anchorage of the 4KV SWGR EAB/60' 1.69 0,41 0.56 0.34 1.67 0.37 0.51 0.39 switchgears does not govern its seismic fragility.
Original capacity is based on Lube Oil Cooler Tank at El. 25', which included the effect of very large nozzle loads (allowable nozzle loads instead of actual nozzle loads were used in the original calculation).
Based on the walkdown, the attached piping is rigidly Emergency diesel DGB/25'& 1.62 0.40 0.48 0.38 4.85 0.36 0.50 1.17 supported and would not impart large generators 551 nozzle loads. Therefore, the lube oil cooler tank can be screened out as having capacity greater than the DG control panel located on the mezzanine floor above elevation 29';i.e. at El. 55'.The updated seismic fragility is based on functionality testing of the DG Icontrol panels.The original capacity of the AFW tank was assumed the same as the AFW storage tank YDB/29' 1.09 0.42 0.51 0.23 1.09 0.42 0.51 0.23 calculated capacity for the CCW tank.The CCW tank is a horizontal tank on two steel saddles at El. 60' of MEAB, Attachment 2 to Enclosure NOC-AE-1 1002773 Page 12 of 29 Table 5 -Comparison of Original and Revised Fragility Curves for STP Component BldglEl. Original Update (1) (wloriginal structural Comments Name Betas Am beta-R beta-U HCLPF Am beta-R beta-U HCLPF and the AFW tank is a cylindrical vertical concrete tank with 30" thick wall and roof including a 1/4" steel liner and is located in the yard at El. 29'.Since no detailed information on the AFW tank is available at this time, and the original capacity assigned to this tank seems reasonable, the original seismic capacity assigned to this tank is retained.
It is not anticipated that cracking of the tank shell or tank overturning would govern seismic fragility of this tank.The original capacity of the CCW surge tank was based on anchor bolt failure mode. The original analysis conservatively included very high nozzle loads, which corresponded to the nozzle allowable loads from the manufacturer.
The walkdown of this tank verified that the attached piping to CCW surge tank MAB/60' 1.09 0.42 0.51 0.23 2.54 0.39 0.52 0.57 this tank are all very well supported and limit the amount of imparted nozzle loads to insignificant levels. The revised fragility takes into consideration estimated realistic nozzle loads and uses the current EPRI TR-103959 guidelines to estimate the seismic fragility of this tank based on anchor bolt failure.
Attachment 2 to Enclosure NOC-AE-1 1002773 Page 13 of 29 Table 5 -Comparison of Original and Revised Fragility Curves for STP Component Bldg/El. Original Update (1) (w/original structural Comments Name Betas Am beta-R beta-U HCLPF Am beta-R beta-U HCLPF The original capacity of these panels was based on the capacity of the 10KV inverters, which were governed by ESFAS CABINET EAB/35' 1.31 0.41 0.52 0.28 1.8 0.37 0.51 0.42 dynamic testing. While, so far no specific dynamic test for these panels have been identified, the assigning of the inverter capacity to these panels is reasonable.
The updated capacities are elevation specific and are based on the inverter capacity.
Per the recent walkdown and review of the associated ESFAS CABINET EAB/10' 1.31 0.41 0.52 0.28 1.9 0.37 0.51 0.44 anchorage design calculation, the anchorage of these panels can be screened out as not being a limiting failure mode.The original capacity of these panels was based on the capacity of the 10KV inverters, which were governed by dynamic testing. While, so far no specific dynamic test report for these panels has been identified, the assigning of the inverter capacity to SSPS CABINET EAB/35' 1.31 0.41 0.52 0.28 1.8 0.37 0.51 0.42 these panels is reasonable.
The updated capacities are elevation specific and are based on the inverter capacity.
Per the recent walkdown and review of the associated anchorage design calculation, the anchorage of these panels can be screened out as not being a limiting failure mode.The original capacity of these panels NSSS CABINET EAB/35' 1.31 0.41 0.52 0.28 1.8 0.37 0.51 0.42 was based on the capacity of the 10KV inverters, which were governed by Attachment 2 to Enclosure NOC-AE-1 1002773 Page 14 of 29 Table 5 -Comparison of Original and Revised Fragility Curves for STP Component BldglEl. Original Update (1) (wloriginal structural Comments Name Betas Am beta-R beta-U HCLPF Am beta-R beta-U HCLPF dynamic testing. While, so far no specific dynamic test report for these panels has been identified, the assigning of the inverter capacity to these panels is reasonable.
The updated capacities are elevation specific and are based on the inverter capacity.
Per the recent walkdown and review of the associated anchorage design calculation, the anchorage of these panels can be screened out as not being a limiting failure mode.ECW PUMP ECW/32' 2.01 0.39 0.48 0.48 2.01 0.39 0.48 0.48 The credible failure mode of the ECW ECW PUMP EAB/10' 2.01 0.39 0.48 0.48 3.14 0.37 0.51 0.74 pumps was established in the original Breaker Cabinet calculation as being the seismic ECW PUMP restraint bolts for the support of the Breaker Cabinet EAB/35' 2.01 0.39 0.48 0.48 3.14 0.37 0.51 0.74 pump casing. While the original calculation uses the DGB's structural response factor and its variabilities and applies it to the Intake Structure, the original pump median capacities are reasonable and will be retained.
On the other hand, since the pumps are all on elevation 32' of the intake structure ECW PUMP and the breakers are at El. 10', 35' and Breaker Cabinet EAB/60 2.01 0.39 0.48 0.48 1.67 0.37 0.51 0.39 60' of the auxiliary building, one needs to establish the seismic fragility of the breakers.
These breakers are inside the 5kv switchgears and therefore the seismic capacity of the switchgears can be assigned to these breakers.
If Train C is not credited, then the fragility of the pumps will govern, otherwise the Attachment 2 to Enclosure NOC-AE-1 1002773 Table 5 -Comparison of Original and Revised Fragility Curves for STP Page 15 of 29 Component Bldg/El. Original Update 11) (w/original structural Comments Name Betas Am beta-R beta-U HCLPF Am beta-R beta-U HCLPF fragility of the breakers at El. 60' of EAB would govern. For the current study, no credit is taken for train C so that the pump fragility governs.(1) Reported estimated update capacities are based on guidelines in EPRI NP-6041 and TR-103959.
Original structural response factors and variabilities along with updated equipment specific variabilities per EPRI TR-103959 were used to calculate updated capacities.
Legend of acronyms used in Table 5: AFW -auxiliary feedwater CCW -component cooling water DG -diesel generator DGB -diesel generator building EAB -electrical auxiliary building ECW -essential cooling water El -elevation ESFAS -engineered safety features actuation system HCLPF -high confidence low probability of failure MEAB -mechanical/electrical auxiliary building NSSS -nuclear steam supply system SSPS -solid state protection system SWGR -switchgear Attachment 2 to Enclosure NOC-AE-1 1002773 Page 16 of 29 Impact on SAMA In order to determine how the updated seismic hazard curves would impact the SAMA analysis, it is necessary to identify which of the cases described above should be used in the evaluation.
For this response, the most appropriate results are considered to be those represented by the use of the 2008 USGS seismic hazard curves (the most recent of the hazard curves) and the updated fragility curves for the components listed in Table 5 (best estimate fragility data for selected components).
While indications are that the original Beta-R and Beta-U values yield overly conservative results, the original values are retained for this evaluation given that no detailed analysis has been performed to update them. These conditions correspond to the results provided in Table 2 for the "USGS Hazard Curves".The "Table 2" conditions identified above will be used in two distinct evaluations:
* Case 1: Evaluate how the use of the updated seismic data impacts the SAMA analysis when considered alone,* Case 2: Evaluate how the use of updated seismic and fire data (from the response to RAI 3.b provided in Attachment 1 to this Enclosure) impacts the SAMA analysis when applied together.These assessments were performed using the steps below, although it should be noted that the updated importance list review was performed only for Case 2 because it provides "bounding" results: " Update the STP Maximum Averted Cost-Risk (MACR) using the new results," Check the importance review threshold for the overall SAMA identification process,* Determine if any new, potentially cost beneficial SAMAs can be identified,* Update the Phase I analysis,* Update the Phase II analysis.Case 1: Seismic Only The updated STP MACR was calculated in the same manner as described in Section F.4 of the STP SAMA submittal using the total CDF and the release category frequencies from Table 2.The result is a single unit MACR of $396,000 and $792,000 for the site, which is $274,000 greater than the $518,000 site MACR used in the SAMA submittal.
Because the importance review threshold was artificially reduced for STP in order to provide a more robust review of the PRA results, the increase in the MACR would not result in a reduction of the review threshold below that which was used in the SAMA submittal.
The minimum SAMA implementation cost of $100,000 for the site correlates to a risk reduction worth (RRW) value of about 1.14, which is significantly higher than the RRW threshold of 1.022 that was applied in the SAMA analysis.
No new, non-seismic contributors are included in the importance list above the review threshold.
While unlikely, it is possible that some previously un-reviewed seismic split fractions could have RRW values greater than the new review threshold of 1.14; however, the importance list review is deferred to the examination of the combined fire and seismic results in Case 2.
Attachment 2 to Enclosure NOC-AE-1 1002773 Page 17 of 29 The Phase I analysis, which is dependent on the value of the site MACR, must be re-performed using the MACR that reflects the Table 2 seismic results. As identified above, the MACR increased from $518,000 to $792,000 for the base case. When the 95th percentile PRA results are applied, this value became $2,138,400 (using the multiplier of 2.7 from the response to RAI ld). Some of the SAMAs (i.e., 3b, 7,'8, 9, 11, and 16) that were screened in the Phase I analysis of the SAMA submittal can no longer be screened when the updated 95th percentile MACR is applied and are transferred to the Phase II screening process for further evaluation.
One method of screening SAMAs in the Phase II analysis is to perform a PRA model run to support a detailed cost benefit analysis for the SAMAs. However, PRA insights can also be used to demonstrate that the SAMAs could not be cost beneficial.
This type of screening was performed for these same SAMAs as part of the response to the follow up RAI on question 1.d (
 
==Reference:==
 
STPNOC response to NRC Requests for Additional Information for the South Texas Project License Renewal Application dated August 23, 2011[NOC-AE-1 1002711] [ML1 1250A067])
which considered a 9 5 th percentile MACR of $2.4 million.The process performed for SAMAs 7, 8, 9, 11, and 16 in Case 2 as part of the for the Case 2 response as the results of the Case 2 screening evaluation will bound the Case 1 screening evaluation.
Due to the difficulty associated with characterizing the potential cost benefit of SAMA 3b using only PRA insights, its averted cost-risk was determined using a PRA model quantification along with the original Phase II SAMAs from the ER (see Tables 5 and 6).In order to determine how the seismic updates impact the STP Phase II SAMA quantifications, each of the quantifications was updated to include the new hazard curves and fragility data.Table 5 documents the PRA results for the updated quantifications.
Table 5: Phase II Quantification Results Using the 2008 USGS Seismic Hazard Curves and Table 5 Fragility Data Updates Case Iese CDF RELI REL II REL III REL IV Identifier BASEUSGS 9.40E-06 6.830E-07 2.70E-06 2.08E-06 3.91 E-06 SAMA 3b 9.36E-06 6.830E-07 2.67E-06 2.08E-06 3.91 E-06 SAMA 4 9.27E-06 5.58E-07 2.70E-06 2.08E-06 3.91E-06 SAMA 10 9.40E-06 6.74E-07 2.71 E-06 1.93E-06 4.06E-06 SAMA 12 9.40E-06 6.80E-07 2.70E-06 2.08E-06 3.91E-06 SAMA 13 9.37E-06 6.82E-07 2.69E-06 2.08E-06 3.90E-06 SAMA 15 9.31 E-06 6.78E-07 2.68E-06 2.06E-06 3.87E-06 Using the 2.7 uncertainty multiplier from the response to RAI ld and the results from Table 5, the cost benefit analysis was updated to determine if the use of the updated seismic data would impact the conclusions of the SAMA analysis.
As documented in Table 6 below, the conclusions of the analysis did not change:
Attachment 2 to Enclosure NOC-AE-1 1002773 Page 18 of 29 Table 6: Cost Benefit Analysis Results for Phase I1 SAMAs Supported by Detailed PRA Quantifications (2008 USGS Seismic Hazard Curves, Table 5 Fragility Data Updates, 9 5 th Percentile PRA Results)Cost of Total Averted Change in SAMA ID Net ValueStus Implementation Cost-Risk Status?SAMA 3b $796,677 $10,514 -$786,163 No SAMA 4 $100,000 $71,906 -$28,094 No SAMA 10 $100,000 $8,122 -$91,878 No SAMA 12 $100,000 $2,187 -$97,813 No SAMA 13 $100,000 $5,081 -$94,919 No SAMA 15 $100,000 $18,587 -$81,413 No Case 2: Inteqrated Seismic and Fire The updated STP MACR was calculated in the same manner as described in Section F.4 of the STP SAMA submittal using the CDF and release category frequencies resulting from the quantification of the STP PRA model after incorporation of both the updated seismic data described in Case 1, above, and the updated fire scenario initiating event frequencies from the response to follow up RAI 3.b (see Attachment 1 to this Enclosure).
Table 7 documents these frequencies.
Table 7: Integrated Seismic and Fire Model PRA Results Case Ieie CDF REL I REL II REL III REL IV Identifier BASEInteqrated 1.06E-05 7.29E-07 3.50E-06 2.22E-06 4.10E-06 The result is a single unit MACR of $451,000 and $902,000 for the site, which is $384,000 greater than the $518,000 site MACR used in the SAMA submittal.
Because the importance review threshold was artificially reduced for STP in order to provide a more robust review of the PRA results, the increase in the MACR would not result in a reduction of the review threshold below that which was used in the SAMA submittal.
The minimum SAMA implementation cost of $100,000 for the site correlates to a risk reduction worth (RRW) value of about 1.12, which is significantly higher than the RRW threshold of 1.022 that was applied in the SAMA analysis.
However, because the postulated fire and seismic changes have resulted in a significant increase in CDF and in the risk profile, the importance review has been re-performed.
In this case, however, no effort was expended to expand the review beyond the RRW review threshold of 1.12. Tables 8 and 9 document the results of the Level 1 and Level 2 importance list reviews, respectively.
Note: the Level 2 importance list was generated in the same manner as described in Section F.5.1.2 of the South Texas Project License Renewal Application Environmental Report.
Attachment 2 to Enclosure NOC-AE-1 1002773 Page 19 of 29 Table 8 Level I Importance Review Event Probability Risk Description Potential SAMAs Name Reduction Worth TMEBBC 2.67E-01 1.33E+00 No Planned Maintenance This split fraction was addressed in the importance list review performed in Trains B, C Running the ER and as described there, it only denotes the configuration of the plant systems at the time of the initiating event. However, the top sequences that include this split fraction were reviewed to identify new fire and seismic contributors.
For this plant configuration, the SEIS6 (lg to 2.1g), SEIS5 (0.8g to lg) and SEIS4 (0.5g to 0.8g) initiators are top contributors that are often paired with seismically induced failure of all EDGs (seismically induced SBO). These event combinations are an indication of a seismically induced SBO scenario; however, the STP units are only designed for a Safe Shutdown Earthquake of 0.1g and have an Operating Basis Earthquake of 0.05g.Therefore, for the seismic events characterized by these initiators, taking steps to ensure the availability of emergency 4KV power would address only a subset of the total failures.
For these scenarios, not even the RCS is expected to remain intact. Even if the RCS is assumed to be intact, the development of a "seismic-safe" system (SAMA $1) for heat removal would be extremely expensive.
Diablo Canyon Power Plant estimated the cost of a 480V generator and a 480V AC, air cooled positive displacement pump for RCS injection to be $6.4 million dollars (Reference 7). That cost was for a non-seismically qualified system, but it was used as a lower bound estimate for the seismically qualified system. Similarly, STP developed an implementation cost for a portable 480V AC generator for AFW support and an alternate means of providing power to the PDP for primary side injection (SAMA 1). While SAMA 1 is designed to protect the TSC DIG in high wind events, protecting the TSC D/G for the seismic initiators identified here would be significantly more expensive.
For this analysis, the implementation cost for SAMA 1 ($3,457,400) is considered to be a lower bound cost for the seismic safe system (SAMA 51) for cases with an intact RCS. Again, in most cases, the RCS would not be intact and the capability Attachment 2 to Enclosure NOC-AE-1 1002773 Page 20 of 29 Table 8 Level I Importance Review Event Probability Risk Description Potential SAMAs Name Reduction Worth of SAMA S1 would have to be significantly increased to ensure adequate core cooling (e.g., higher pump flow rate, a core sparger system).The same fire scenarios that were addressed in the ER also were included in combinations with TMEBBC (fire scenarios 2 and 3). Some sequences included fire scenario 4, but as described in the response to the follow up question on RAI 3.b, fire scenario 4 has a PACR of only $24,000 and no SAMAs are suggested.
The remaining new sequences have frequencies below 1 E-8/yr and SAMAs are not considered to be required to address them.TMEBCA 2.67E-01 1.32E+00 No Planned Maintenance, This split fraction was addressed in the importance list review performed in Trains C, A Running the ER and as described there, it only denotes the configuration of the plant systems at the time of the initiating event. However, the top sequences that include this split fraction were reviewed to identify new fire and seismic contributors.
Generally, TMEBCA included the same types of sequences as TMEBBC and the same conclusions are applicable.
For some of the lower contributors, the seismic sequences include other unique combinations of split fractions, but the overriding issue of widespread plant damage makes their inclusion inconsequential.
No additional SAMAs required.TMEBAB 2.67E-01 1.32E+00 No Planned Maintenance, This split fraction was addressed in the importance list review performed in Trains A, B Running the ER and as described there, it only denotes the configuration of the plant systems at the time of the initiating event. However, the top sequences that include this split fraction were reviewed to identify new fire and seismic contributors.
Generally, TMEBAB included the same types of sequences as TMEBBC and the same conclusions are applicable.
For some of the lower contributors, the seismic sequences include other unique combinations of split fractions, but the overriding issue of widespread plant damage makes their inclusion inconsequential.
Attachment 2 to Enclosure NOC-AE-1 1002773 Page 21 of 29 Table 8 Level I Importance Review Event Probability Risk Description Potential SAMAs Name Reduction Worth No additional SAMAs required.GAA 6.21 E-02 1 .1OE+00 DG 11 FAILS -ALL This split fraction was addressed in the importance list review performed in SUPPORT AVAILABLE the ER and even with the updated fire and seismic data, those initiators are small contributors to the importance of the GAA split fraction.
No additional SAMAs are required.OGRB 6.00E-01 1.08E+00 NON-RECOVERY OF 345 This split fraction was addressed in the importance list review performed in AND 138KV the ER and no additional review is required (fire and seismic events are non-contributors to sequences including this event).
Attachment 2 to Enclosure NOC-AE-1 1002773 Page 22 of 29 Table 9 Level 2 Importance Review Event Probability Risk Description Potential SAMAs'Name Reduction Worth C4A 3.70E-01 1.36E+00 NO CONTAINMENT HEAT This split fraction was addressed in the importance list review performed in REMOVAL the ER. In the current model, the non-fire/non/seismic scenarios are still the dominant sequences including the C4A split fraction, but some seismic sequences include C4A. The contributors include. combinations of the seismic initiating events with the seismically induced failures of the EDGs, which are addressed in the Level 1 importance list (Table 12) for split fraction TMEBBC.This split fraction was included in fire scenarios, but they are the same ones addressed in the ER (fire scenarios 2 and 3).TMEBBC 2.67E-01 1.33E+00 No Planned Maintenance This split fraction is addressed in Table 8.Trains B, C Running TMEBCA 2.67E-01 1.32E+00 No Planned Maintenance This split fraction is addressed in Table 8.Trains C, A Running TMEBAB 2.67E-01 1.32E+00 No Planned Maintenance This split fraction is addressed in Table 8.Trains A, B Running 1 Level 2 versions of the split fractions are identified with "(2)" designators.
Attachment 2 to Enclosure NOC-AE-1 1002773 Page 23 of 29 Table 9 Level 2 Importance Review Event Probability Risk Description Potential SAMAs'Name Reduction Worth RPT 5.00E-01 1.20E+00 -200<RCS PRESS<600 The main contributors for the sequences including RPT are the same as in PSIA @VB GIVEN A STUCK the ER and the treatment would be the same. Fire scenario 2 is a larger OPEN PORV @UTAF contributor in the updated model, but the treatment proposed in the ER is still applicable here and no new SAMAs are required.Some sequences included fire scenario 4 (FS4), but as described in the response to the follow up question on RAI 3b, FS4 has a PACR of only$24,000 and no SAMAs are suggested.
Seismic initiators SEIS6(2) and SEIS4(2) are also combined with RPT, but the scenarios and conditions are the same as for TMEBBC, TMEBCA, and TMEBAB in Table 8.RPS 5.00E-01 1.20E+00 -RCS PRESS<200 PSIA The main contributors for the sequences including RPS are the same as in@VB GIVEN STUCK OPEN the ER and the treatment would be the same. Fire scenario 2 is a larger PORV @UTAF contributor in the updated model, but the treatment proposed in the ER is still applicable here and no new SAMAs are required.Some sequences included FS4, but as described in the response to the follow up question on RAI 3b, FS4 has a PACR of only $24,000 and no SAMAs are suggested.
Seismic initiators SEIS6(2) and SEIS4(2) are also combined with RPS, but the scenarios and conditions are the same as for TMEBBC, TMEBCA, and TMEBAB in Table 8.
Attachment 2 to Enclosure NOC-AE-1 1002773 Page 24 of 29 Table 9 Level 2 Importance Review Event Probability Risk Description Potential SAMAs'Name Reduction Worth LSB 8.00E-01 1.14E+00 INDUCED PORV FAILURE As was the case in the ER, most of the contribution from sequences WEN RCS PRESS > 2000 including LSB is from fire scenario 2. The contribution from these scenarios PSIA OR OP OPENS P is increased for the updated model, but the treatment is the same and no new SAMAs are required.Some sequences included FS4, but as described in the response to the follow up question on RAI 3b, FS4 has a PACR of only $24,000 and no SAMAs are suggested.
LSB is also often combined with the SEIS6 initiating event, which is addressed in Table 8 for the plant configuration split fractions (e.g., TMEBBC).Other new contributors are low frequency and do not require SAMAs.SOG6 9.94E-01 1.10E+00 SEIS6, Hazard Levels: 1 to These seismic initiating events, which would cause extreme and 2.1 widespread plant damage, are addressed in Table 8 for the plant configuration split fractions (e.g. TMEBBC).
Attachment 2 to Enclosure NOC-AE-1 1002773 Page 25 of 29 The importance list review indicates that incorporation of the new fire and seismic data resulted in the generation of one new seismic SAMA, but no new fire SAMAs. The following is a summary of the new SAMA and its Phase I dispositions:
SAMA ID Description Cost of Implementation Phase I Disposition SAMA S1 Install a "Seismic Safe" System $3,457,400 (lower bound The cost of surrogate from SAMA 1 implementation exceeds in the ER) the $902,000 MACR.Screened from further review.As shown above, the new SAMA was able to be screened in the Phase I analysis.The Phase I analysis for the original SAMAs was re-performed using the MACR that reflects the updated model. As identified above, the MACR increased from $518,000 to $902,000 for the base case. When the 9 5 th percentile PRA results are applied, this value becomes $2,435,400 (using the multiplier of 2.7 from the response to RAI 1d). Some of the SAMAs (i.e., 3b, 6, 7, 8, 9, 11, and 16) that were screened in the Phase I analysis of the SAMA submittal can no longer be screened when the updated 9 5 th percentile MACR is applied and are transferred to the Phase II screening process for further evaluation.
One method of screening SAMAs in the Phase II analysis is to perform a PRA model run to support a detailed cost benefit analysis for the SAMAs. However, PRA insights can also be used to demonstrate that the SAMAs could not be cost beneficial.
This type of screening was performed for these same SAMAs as part of the response to the follow up RAI on question 1 .d (
 
==Reference:==
 
STPNOC response to NRC Requests for Additional Information for the South Texas Project License Renewal Application dated August 23, 2011 [NOC-AE-1 1002711][ML1 1250A067])
which considered a 95th percentile MACR of $2.4 million. Table 10 documents a similar Phase II screening process using insights from the updated model. Due to the difficulty associated with characterizing the potential cost benefit of SAMA 3b using only PRA insights, its averted cost-risk was determined using a PRA model quantification along with the original Phase II SAMAs from the Environmental Report (see Tables 11 and 12).
Attachment 2 to Enclosure NOC-AE-1 1002773 Page 26 of 29 Table 10: Phase II Screening for SAMAs 6, 7, 8, 9, 11, and 16 (2.7 uncertainty multiplier, updated fire and seismic data)SAMA ID Description Cost of Disposition Implementation SAMA 6 Install an Additional
$2,301,000 The cost of implementation (COI) is 95% of the 95"' percentile MACR. An Diverse, High Head additional high pressure primary side injection system dependent on AC Safety Injection (HHSI) power will not mitigate 95% of plant risk and would clearly not be cost Pump beneficial.
Screened from further analysis.SAMA 7 Provide Portable Fans $950,000 The importance of the EAB HVAC system was estimated by reviewing the and Ductwork for loss of EAB HVAC initiating event contributions to Level 1 and Level 2 in Alternate Electrical conjunction with a review of the importance of the system split fractions.
Auxiliary Building (EAB) The importances were then correlated to frequencies and assessed to Room Cooling determine if they could result in an averted cost-risk greater than the COI.The absolute contribution from the loss of EAB HVAC scenarios would not be impacted by changes to the fire and seismic initiating event frequency updates. Therefore, as noted in the follow-up response to RAI 1d, the CDF contribution from loss of EAB HVAC is 4% of the REV6 model CDF (0.04-6.39E-06=
2.5E-07).
Similarly, contribution to the Group I release category is 3.1% of the Rev6 frequency (0.031*5.01E-07=1.5E-08) and 5.4% of the Group III frequency (0.054*1.48E-06=8.0E-8).
For contributions not associated with the loss of EAB HVAC initiating event, the sum of the Fussel-Vesely (F-V) values for all EAB HVAC split fractions was used to estimate percent contribution to CDF and Level 2 RCs.Because some split fractions are correlated (they are only used with other system split fractions), it should be noted that the system importance is over estimated using this method. For CDF, the F-V value is 3.7E-2, which corresponds to a frequency of about 3.9E-7/yr (0.037*1.06E-05=3.9E-07).
For Level 2, the F-V value of 3.8E-2 was generated from the composite file used to develop the Level 2 importance list in the ER and it corresponds to a frequency of only 2.4E-7/yr (6.45E-6*0.038).
For this analysis, it is assumed that it all maps to the Group I release category, which is conservative.
The total potential CDF reduction is, therefore, 6.4E-07 (2.5E-07+3.9E-07=6.4E-07).
The Level 2, the total potential frequency reduction in the significant release categories is 3.4E-07 (1.5E-08+8.0E-08+2.4E-07=3.4E-07).
If this frequency is assumed to be all binned to Group I, it would correlate to about 13% of Attachment 2 to Enclosure NOC-AE-1 1002773 Page 27 of 29 Table 10: Phase II Screening for SAMAs 6, 7, 8, 9, 11, and 16 (2.7 uncertainty multiplier, updated fire and seismic data)SAMA ID Description Cost of Disposition Implementation the total dose-risk and 27% of the total economic cost-risk.
Given the cost of implementation (COI) is about 40% of the 9 5 th percentile MACR, the contributions from EAB HVAC (6.0% CDF, 13% person-rem, 27% economic cost-risk) are low enough to ensure this SAMA would not be cost beneficial.
Screened from further analysis.SAMA 8 Enhance Fire Barriers $1,150,500 SAMA 8 addresses the risk from fire scenario 18 (FR18(2)).
A potential in Control Room averted cost risk for this fire scenario can be calculated using the Envelope (CRE) Panel frequencies documented in Tables 11 and 12 of the response to the RAI 3b 22/4 follow-up question (the seismic update do not impact the fire contributions).
Using the cost benefit methodology documented in section F.4 of the ER, the potential averted cost risk associated with this scenario is $64,800 when the 95 percentile PRA results are considered.
The $1,150,500 cost of implementation exceeds the PACR for this SAMA. Screened from further analysis.SAMA 9 Use AMSAC to Back $970,000 The COI is about 40% of the 95t' percentile MACR. The OTA split fraction Up the Existing SCRAM (operator manually trips reactor) is included in most ATWS sequences and Signal can be used to quickly estimate the contribution of ATWS to the CDF and the Level 2 sequences.
For this evaluation, it is assumed ATWS is double the OTA contribution.
For the model with the combined fire and seismic updates, the Fussel-Vesely value for OTA is about 0.024 for CDF and about 0.015 for release categories 1, 11, and II1. Even if these estimates are increased by a factor of two and a 5% reduction in CDF is assumed and 3% reduction in the release category 1, 11, and III frequency is mapped to release category I, the averted cost-risk would only be about $161,000 when the 95th PRA results are considered.
Screened from further analysis.SAMA 11 Modify Fire Protection
$849,600 This SAMA would only prevent containment failure and reduce source terms System to Supply (no CDF reduction).
Group I is dominated by bypass failures (and is Containment Spray assigned a bypass source term) and containment spray is assumed to have (CS) Headers no significant impact on this group. Group II is characterized by a pre-existing hole in containment, so the containment failure prevention aspect of this SAMA is not applicable.
With regard to source term reduction, the group is already characterized as a small release and no credit is assumed I_ I_ I_ for further reductions.
Group IV is a containment leakage case and CS Attachment 2 to Enclosure NOC-AE-1 1002773 Page 28 of 29 Table 10: Phase II Screening for SAMAs 6, 7, 8, 9, 11, and 16 (2.7 uncertainty multiplier, updated fire and seismic data)SAMA ID Description Cost of Disposition Implementation availability is assumed to have no impact on the release. Group III scenarios, however, would potentially be mitigated.
Even if this SAMA was assumed to eliminate all group III contributions (0.63 person-rem, $34 economic cost-risk), this SAMA would not be cost beneficial.
The averted cost-risk corresponding to this reduction is approximately
$105,000 when the 9 5 th percentile PRA results are considered.
This is significantly less than the COI. Screened from further analysis.SAMA 16 Portable, Engine Driven $1,227,200 Loss of instrument air initiating event contributes less than 1% the total CDF Instrument Air and to the Group 1, 11, and III release categories.
In addition, the only split Compressor fraction with an RRW value greater than 1.0 for CDF and those release categories is IAS14 at 1.02. These metrics imply a potential risk reduction of less than 3% for both CDF and the important level 2 release categories.
Given that the COI is over 50% of the 95 h percentile MACR, this SAMA is screened from further review.
Attachment 2 to Enclosure NOC-AE-1 1002773 Page 29 of 29 In order to determine how the fire and seismic updates would impact the STP Phase II SAMA model quantifications, each of the quantifications was updated. Table 11 documents the PRA results for the updated quantifications.
Table 11: Phase II Quantification Results (Updated Fire and Seismic Data)Case dese CDF REL I REL II REL III REL IV Identifier BASEUSGS 1.06E-05 7.29E-07 3.50E-06 2.22E-06 4.10E-06 SAMA 3b 1.05E-05 7.29E-07 3.47E-06 2.22E-06 4.10E-06 SAMA 4 1.04E-05 6.03E-07 3.50E-06 2.22E-06 4.10E-06 SAMA 10 1.06E-05 7.20E-07 3.51E-06 2.07E-06 4.25E-06 SAMA 12 1.06E-05 7.25E-07 3.50E-06 2.22E-06 4.1OE-06 SAMA 13 1.05E-05 7.27E-07 3.50E-06 2.22E-06 4.09E-06 SAMA 15 1.05E-05 7.23E-07 3.48E-06 2.19E-06 4.06E-06 Using the 2.7 uncertainty multiplier from the response to RAI ld and the results from Table 15, the cost benefit analysis was updated to determine if the use of the updated fire and seismic data would impact the conclusions of the SAMA analysis.
As documented in Table 12 below, the conclusions of the analysis did not change: Table 12: Cost Benefit Analysis Results for Phase II SAMAs Supported by Detailed PRA Quantifications (Updated Fire and Seismic Data, 95 Percentile PRA Results)Cost of Total Averted Change in SAMA ID Implementation Cost-Risk Net Value Status?SAMA 3b $796,677 $17,771 -$778,906 No SAMA 4 $100,000 $82,890 -$17,110 No SAMA 10 $100,000 $8,197 -$91,803 No SAMA 12 $100,000 $2,597 -$97,403 No SAMA 13 $100,000 $15,741 -$84,259 No SAMA 15 $100,000 $22,000 -$78,000 No}}

Latest revision as of 06:22, 30 April 2019

Redirect to:

Retrieved from "https://kanterella.com/w/index.php?title=ML12030A081&oldid=1554412"