RS-16-075, Supplemental Information Regarding License Amendment Request to Revise Technical Specification Section 5.5.13, Primary Containment Leakage Rate Testing Program, for Permanent Extension of Type a and Type C Leak Rate Test
| ML16091A077 | |
| Person / Time | |
|---|---|
| Site: | Clinton  |
| Issue date: | 03/31/2016 |
| From: | Simpson P Exelon Generation Co |
| To: | Document Control Desk, Office of Nuclear Reactor Regulation |
| References | |
| CAC MF7290, RS-16-075 | |
| Download: ML16091A077 (30) | |
Text
RS-16-075 10 CFR 50.90 March 31, 2016 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555-0001 Clinton Power Station, Unit 1 Facility Operating License No. NPF-62 NRG Docket Nos. 50-461
Subject:
Supplemental Information Regarding License Amendment Request to Revise Technical Specification Section 5.5.13, "Primary Containment Leakage Rate Testing Program," for Permanent Extension of Type A and Type C Leak Rate Test Frequencies
References:
- 1. Letter from Patrick R. Simpson (Exelon Generation Company, LLC) to U. S.
NRG, "License Amendment Request to Revise Technical Specification Section 5.5.13, "Primary Containment Leakage Rate Testing Program," for Permanent Extension of Type A and Type C Leak Rate Test Frequencies,"
dated January 25, 2016 (RS-16-015)
- 2. Letter from U. S. NRG to Bryan Hanson (Exelon Generation Company, LLC),
"Clinton Power Station, Unit No. 1 - Supplemental Information Needed For Acceptance Of Requested Licensing Action Concerning Extension Of Leak Rate Testing (CAC No. MF7290)(RS-16-015)," dated March 17, 2016 In Reference 1, Exelon Generation Company, LLC (EGG) requested an amendment to Facility Operating License No. NPF-62 for Clinton Power Station (CPS), Unit 1. The proposed change is a request to revise TS 5.5.13, "Primary Containment Leakage Rate Testing Program" to allow for the permanent extension of the Type A Integrated Leak Rate Testing (ILRT) and Type C Leak Rate Testing frequencies.
In Reference 2, the NRG provided the results of the acceptance review of the license amendment request. The NRG concluded that additional information is needed to enable the NRG to make an independent assessment regarding the acceptability of the proposed amendment in terms of regulatory requirements and the protection of public health and safety and the environment. As documented in Reference 2, the NRG requested that EGG supplement the application to address the information requested by April 4, 2016. In response to this request, EGG is providing the attached information.
March 31, 2016 U.S. Nuclear Regulatory Commission Page2 EGC has reviewed the information supporting a finding of no significant hazards consideration, and the environmental consideration, that were previously provided to the NRC in Attachment 1 of Reference 1. The additional information provided in this submittal does not affect the bases for concluding that the proposed license amendment does not involve a significant hazards consideration. In addition, the additional information provided in this submittal does not affect the bases for concluding that neither an environmental impact statement nor an environmental assessment needs to be prepared in connection with the proposed amendment.
There are no regulatory commitments contained in this letter. Should you have any questions concerning this letter, please contact Mr. Timothy A. Byam at (630) 657-2818.
I declare under penalty of perjury that the foregoing is true and correct. Executed on the 31 51 day of March 2016.
Patrick R. Simpson Manager - Licensing Exelon Generation Company, LLC
Attachment:
Response to Request for Supplemental Information cc:
NRC Regional Administrator, Region Ill NRC Senior Resident Inspector-Clinton Power Station Illinois Emergency Management Agency - Division of Nuclear Safety
ATTACHMENT Response to Request for Supplemental Information
RM DOCUMENTATION NO: CL-LAR-07 STATION: Clinton Power Station (CPS)
UNIT(S) AFFECTED: 1 REV: 1 PAGENO. 1 TITLE: Supplemental Information for Risk Assessment for CPS Regarding the ILRT {Type A) and DWBT Permanent Extension Request
SUMMARY
- CPS Is pursuing a License Amendment Request (LAR) to permanently extend the Integrated Leak Rate Test (ILRT) and Drywell Byp21ss Test (DWBT) to 15 years.
The purpose of this document is to provide supplemental information to the NRC regarding the F&Os from the 2009 Clinton Peer Review. This document supplements CL-LAR-07 Rev. 0, which remains In effect.
This is a Category 1 Risk Management Document in accordance with ER-AA-600-1012, which requires independent review and approval.
[ X] Internal RM Documentation
[ J External RM Documentation Electronlc Calculation Data Flies:
Mftbod of Review:
[ X 1 Detailed [ 1 Alternate ( ] Review of External Document This RM documentation supersedes:
NIA in Its entirety.
Prepared by:
Garrett Snedeker I
k~vtJte 2-~~c~-,
3-30-16 Name Signature Date Reviewed by:
Anthon~ Hable I
I 03~ 3Q**1' Name Date Reviewed by:
John Steinmetz I
I 3-30-16 Name Date lndependentiy ZJEV~
5-50-J~
Reviewed by:
Don Vanover I
Name
~/
Date Approved by:
Gene Kell~
I 0..Jo-1/,
Name Date 1g ature 1of27
Purpose This document provides supplementary information related to the Clinton Full Power Internal Events (FPIE) PRA Model Quality review documented in Appendix A of CL-LAR-07 Rev. 0.
Specifically, resolution for those peer review findings related to the 2009 ASME/ANS PRA Standard Requirements (SRs) not meeting Category I requirements is provided.
Background
NRC Reg. Guide 1.200 An Approach for Determining the Technical Adequacy of Probabilistic Risk Assessments Results for Risk-Informed Applications, Revision 2, March 2009 provides guidance for documentation to support a regulatory submittal. Section 4 of the Reg. Guide recommends "A discussion of the resolution of the peer review (or self-assessment, for peer reviews performed using the criteria in NEI 00-02) findings and observations that are applicable to the parts of the PRA required for the application."
In a letter from the U. S. NRC to Bryan Hanson (Exelon Generation Company, LLC), "Clinton Power Station, Unit No. 1 - Supplemental Information Needed for Acceptance of Requested Licensing Action Concerning Extension of Leak ate Testing (CAC No. MF7290)(RS-16-015),"
dated March 17, 2016, the results of the NRC acceptance review of the Exelon Generation Company, LLC (EGC) amendment request to revise the Clinton Power Station (CPS) technical specifications to reflect a permanent extension of the Type A Integrated Leak Rate Testing and Type C Leak Rate Testing were documented. The NRC indicated they had completed their review of the EGC application and in order for the application to be complete, the NRC requested that EGC supplement the application as follows:
NRC Request:
EGC is using a risk-informed approach to support the test frequency extension consistent with Regulatory Guide (RG) 1.200, "An Approach for Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk-Informed Activities," Revision 2 (ADAMS Accession No. ML090410014). Therefore, consistent with the Section 3.2.4.1 of RG 1.200 and in Regulatory Issue Summary 2007-06, "Regulatory Guide 1.200 Implementation" (ADAMS Accession No ML070650428), EGC is requested to provide a list of all the facts and observations (F&Os) from the 2009 peer-review for which the probabilistic risk assessment (PRA) did not meet the 2009 American Society of Mechanical Engineers/American Nuclear Society PRA Standard Capability Category (CC) I supporting requirements. For each F&O, include details of its disposition, and if open, an explanation of why not meeting the corresponding CC I requirement has no impact on the request.
Methodology As part of the PRA Quality review, all 2009 Peer Review 'findings' associated with SRs not meeting Capability Category I are identified as having the potential to impact the ILRT LAR risk assessment.
Resolution documented in the Clinton PRA 'URE' database is reviewed and where necessary, the PRA model and documentation changes are also reviewed to verify
'findings' have been appropriately addressed.
If an F&O associated with a 'not met' SR has not been addressed, an explanation of why not meeting the corresponding CC I requirement has no impact on the request is provided.
Page 2 of 27
Results/Conclusions Peer Review findings, applicable SRs and resolution are found in Table 1 below.
All findings related to SRs 'not met' have been appropriately closed.
This document supplements the PRA Quality Section of the Clinton ILRT Risk Assessment CL-LAR-07 Rev. 0.
No changes to the PRA related quantitative results, other insights or conclusions are necessary.
Page 3 of 27
TABLE 1 RESOLUTION OF PEER REVIEW F&Os F&O SR Not Met F&O Description Resolution The SR requires that an evaluation of the potential for each A tabular system by system discussion of system [to assess the possibility of an initiating event caused by potential initiators is included in the IE F&O 1-2 IE-AS the system] be performed. While some systems have been Notebook. The additional system by system evaluated, as documented in Section 2, there is no evidence that review did not result in new initiators.
each system at CPS has been evaluated.
The basis for excluding loss of electrical area ventilation is not Engineering Changes (ECs) 402194 and supported with room heat-up calculations. Evaluate the room 402193 provide details of room cooling heat-up profile for each electrical room and compare the calculations for the switchgear rooms. The temperature profile with the capability of equipment in the room calculations show that without Switchgear to withstand conditions expected to result from a loss of Heat Removal System (VX) cooling the F&O 1-3 IE-AS ventilation. An explicit evaluation of room heat-up could result in switchgear rooms will not exceed the 122 new initiating events.
degree F equipment limits if operators open the doors as directed per procedures. Even without operator action the time to heatup to 122F is nearly 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> and it is not considered a credible initiating event.
No systematic evaluation of multiple equipment failures, A tabular system by system discussion of including common cause and routine system alignments, was potential initiators is included in the IE evident in the documentation. (This F&O originated from SR IE-Notebook. Common Cause Failure (CCF) is AG) Perform an evaluation of multiple equipment failures and considered within the System IE fault trees.
F&O 1-4 IE-AG routine system alignments.
Routine alignments based on plant operating history are modeled in each of Consideration of common cause events and routine alignments is the System IE fault trees. Multiple required by the SR and RG 1.200. Such an evaluation could result equipment failures are considered and in additional initiating events being identified.
modeled in the System IE fault trees.
Page 4 of 27
F&O SR Not Met F&O Description Resolution The Summary Notebook includes information that attempts to The Quantification Notebook includes the identify the key sources of uncertainty in the various analyses.
most recent CPS PSA modeling uncertainty IE-D3 However, with the changes to eliminate 'key' from the SR discussions including many quantitative AS-C3 definition, this SR cannot be considered met. Appendix B of the sensitivity studies. It should be noted that SC-C3 CPS-PSA-013 notebook discusses the industry 'key sources of the ILRT risk assessment included the SY-C3 uncertainty' per EPRI TR-1009652. However, the current analysis incorporation of several sensitivity studies HR-113 does not fully meet the requirement of RG 1.200, which requires and factored in the potential impacts from DA-E3 a discussion of sources of model uncertainty and related external events in a bounding fashion.
IFPP-B3 assumptions. Also, there may be some plant-specific assumptions None of the sensitivity studies or bounding IFSO-B3 made that may not be fully captured by the generic list of analysis indicated any source of uncertainty F&O 1-13 IFSN-B3 potential sources of uncertainty.
or modeling assumption that would have IFEV-B3 resulted in exceeding the acceptance IFQU-B3 guidelines. Since the accepted process QU-El utilizes a bounding analysis approach which QU-F4 is mostly driven by that CDF contribution LE-El which does not already lead to LERF, there LE-F3 are no identified key assumptions or LE-G4 sources of uncertainty for this application (i.e. those which would change the conclusions from the risk assessment results).
Page 5 of27
F&O SR Not Met F&O Description Resolution Spray-induced and submergence induced failures appear to have The CPS PSA Internal Flooding analysis been addressed in the analyses. No documentation of a addresses spray/jet impacts on equipment systematic assessment of the effects of jet impingement, pipe (refer to Appendix B.5 of the CPS PSA whip, humidity, temperature, etc., on SSCs could be identified Internal Flooding Notebook, Vol. 1).
and Section 2.2.5 states that these effects are beyond the scope Humidity, temperature and pipe whip are of the analysis. No evaluation of the specific equipment also considered (refer to Section 2.2.5 of evaluated in the PRA compared to equipment considered in the the CPS Internal Flooding Notebook). The F&O 1-14 IFSN-A6 design analyses, e.g., EQ lists, was documented Since the PRA internal flooding analysis covers low can credit non-safety-related equipment, relying on design basis pressure low temperature fluids. High evaluations to dismiss these dynamic effects may credit Energy Line Break (HELB) related issues are equipment that cannot withstand the effects considered in the considered in the Interfacing System Loss of design analysis. In addition, failure in a system containing high Coolant Accident (ISLOCA) and Break temperature fluid can actuate fire systems and impact additional Outside Containment (BOC) analysis.
equipment. Also, the PRA models may evaluate breaks beyond those of the design basis.
Recovery events are included in the WS and TBCCW initiating The Service Water (WS) and Turbine event fault trees. For example, the WS fault tree includes failure Building Closed Cooling Water (TBCCW) of actions to perform a manual backwash of service water recovery events were added to Table 5.1-3 F&O 1-17 IE-C3 strainers and TBCCW include actions to 'recover initial failure.'
of the HRA notebook with a reference to However, no basis for these actions was documented. Document the respective system notebooks. The basis the basis for recovery actions included as recovery events.
for crediting recoveries using conservative probabilities is described in the system notebooks.
Page 6 of27
F&O SR Not Met F&O Description Resolution Nearly all event trees model [probabilistic] success of ECCS pump The bases for Emergency Core Cooling operation following a loss of suppression pool cooling if System (ECCS) pump failure/success containment venting is successful. No evaluation of the ability of following successful containment venting is ECCS pumps to operate at post-containment-venting documented in section titled 'Successful temperatures was provided. If ECCS pump operation for such RPV Injection and Successful Vent' in the ET F&O 1-21 AS-83 conditions cannot be supported, then additional core damage Notebook and the failure probabilities of sequences could result.
the associated ECCS failure modes (e.g.,
basic event lSY--STEAMBOUND-, "Vent Causes Steam Binding in ECCS Suction") are further discussed in AppendixG.13 of the Component Data Notebook.
Page 7 of27
F&O SR Not Met F&O Description Resolution An evaluation of the effects of conditions created by a break ISLOCA and break outside containment outside containment on equipment operation is not documented.
(BOC) accidents play a role in determining For example, a break in the RCIC steam line could cause harsh impact on risk. The operation of ECCS and conditions in plant buildings where equipment needed to support credited non-safety related system during accident mitigation is located, thereby precluding use of that these conditions is documented in the equipment in preventing core damage. If no equipment needed Event Tree and Component Data to mitigate the accident sequence is located in the areas where a notebooks.
break could occur, then such evaluations need to be made in the PRA for each accident sequence (or initiating event) in order to Additional information related to the meet the SR. Some documentation of considering such potential example provided in the F&O: A Reactor effects needs to be presented, even if the conclusion is that there Core Isolation Cooling (RCIC) steam line are no environmental effects by any initiating event. Breaks break would occur in containment (which F&O 1-22 AS-B3 outside containment and ISLOCA events that require long-term contains little equipment modeled in the injection (function XT-CRD) to be successful after an initial period PRA) or in the Auxiliary Building. The High of operation by HPCS or LPCS. However, the CRD model credits Pressure Core Spray (HPCS) pump is in a operation of condensate transfer as a suction source for CRD and separate building, the Fuel Building. The condensate transfer could be failed by the conditions created by Make-up Condensate (MC) and Cycled the BOC or ISLOCA initiators.
Condensate (CY) pumps are also in a separate building (the Turbine Building basement). A Control Rod Drive (CRD) pump is normally running, and it too is located in the Turbine Building. Credited systems are protected from the conditions created by the BOC and ISLOCA initiators by separation as noted above.
Page 8 of27
F&O SR Not Met F&O Description Resolution From a review of system notebooks, it appears that no concerted The resolution of this F&O found the issue attempt was made to identify and model accident conditions that to be documentation issue only. The RCIC could cause system failures. For example, in the RCIC notebook, documentation was corrected to be Assumption 6 states the water leg pump is only needed if AC consistent on the water leg pump power is available. Assumption 17 says that the water leg pump requirement. The high steam flow and high is not needed and the fault tree gate UGATE06 included water leg temperature isolation have been added to pump failure as a failure of the RCIC system. Also, the high steam the documentation. The gland seal flow isolation is not addressed in the RCIC notebook nor is compressor discussion was enhanced for isolation based on area temperature sensors. The RCIC gland seal clarity. The Circulating Water {CW) pump compressor is included as a failure mode for RCIC, but it is not trip on high level was added to the CW F&O 1-24 SY-A18 clear from the documentation why that failure mode is included.
system notebook.
Another example is the circulating water system. No mention is made of the circulating water pump trip in high level in the Other system notebook corrections and turbine building. Because of the inconsistent documentation and enhancements have also been made since above mentioned omissions from the documentation, and further the Peer Review to better document the because no evidence of a review for such conditions was model.
identified in the documentation, it appears that this condition is not an isolated error but a complete omission from the analysis.
Because of these issues this SR is assessed as not met.
Page 9 of27
F&O SR Not Met F&O Description Resolution No calculations supporting room cooling requirements are Calculations are now documented in the documented. No evaluation of control room HVAC was Dependency Matrix Notebook. A reference identified. Better documentation of analysis to demonstrate that to room cooling discussion in Dependency rated or design capabilities are not exceeded is needed within the Matrix Notebook has been added to all of system notebooks. Document the calculations used to exclude the System Notebooks in Section 2.5.
room cooling for each system.
Room cooling calculations for the Switchgear Rooms support the mission time for electrical equipment. (See F&O 1-3 resolution above). A calculation was performed for the 2011 PRA update (Axioma Report 2009-0048) that confirms the success of opening the RHR B F&O 1-26 SY-A22 room doors for SX failures. This calculation is used as the basis of allowing credit for operators opening room doors as a success of room cooling. A Human Error Probability (HEP) is included in the model for operator action to open the room doors. (See F&O 6-8 below). The heat load in the MCR compared with its relatively large size and the ability to open doors, should the need arise, are judged as making it no more limiting than some of the areas where heatup effects have been evaluated and discounted.
Page 10 of27
F&O SR Not Met F&O Description Resolution The CPS PRA models one instance where credit is given for repair The RHR recovery terms represent the only of hardware failures, repair of RHR. Per the requirements of RG recoveries applicable to this F&O. They are 1.200, plant-specific data to quantify the probability of repair based on applicable industry experience as should be collected and analyzed to support any credit given. The allowed by SR DA-C15. Fifteen pages of analysis for CPS uses generic industry data. Therefore, this SR is documentation describing industry data, an SY-A24 assessed as not met. Collect plant-specific repair data for RHR EPRI study, alternatives to modeling and F&O 1-27 DA-C15 system failures and include the data in the repair probability conclusions are presented in Appendix G.14 LE-El calculations.
of the Component Data Notebook. The recovery values are based on an exponential repair model and the average MTTR is judged reasonable given the long time for RHR recovery to avoid containment failure.
Page 11 of27
F&O SR Not Met F&O Description Resolution Flow diversion criterion or discussion otherwise is provided in the The Clinton flow diversion modeling was "Assumption" sections ofthe individual system notebooks. All compared with the MSO expert panel flow flow diversion pathways that divert less than ten-percent of diversion analysis and conclusions. The system flow are categorically excluded without consideration of current 2014 Clinton flow diversion the flow actually required.
modeling agrees with the Expert Panel discussion. The current site MSO resolutions are now supported by Engineering Change (EC) Documentation.
The EC documentation provides additional detail. It appears that the present model is conservative for those scenarios where the F&O 1-30 SY-Al3 valves have been placed in the closed position and the circuit breaker placed on the off position (Ref. MSO 2i for RHR/LPCI flow diversion).
Note that the PRA model does not rely solely on MSO resolutions. For example, a LPCS test return valve F012 spurious opening is OK in MSO space, as LPCS is not a credited system and water is returned to the suppression pool. However, the PRA model fails LPCS ifthe F012 valve spuriously opens.
Page 12 of27
F&O SR Not Met F&O Description Resolution Table 5.1-4 summarizes the post-initiator HEPs in tabular form A new appendix was added to the Clinton and compares the results to results from previous versions of the 2011 HRA notebook to document the CPS PRA models. No evidence of a check for consistency between reasonableness of the HEPs. The validation operator actions was identified. Compare the final HEP values of the HEPs is performed primarily through against each other to check their reasonableness and document the following mechanisms:
the comparison.
- Independent review of the detailed HEP quantifications and results by both PRA and operations personnel
- Consistent use of HRA procedures that are designed to meet the requirements of F&O 1-31 HR-G6 the ASME/ANS PRA Standard. This consistently is facilitated by using the EPRI HRA Calculator.
In addition to the above, the appendix documents an additional HEP consistency review based on comparison of the final HEPs and associated characteristics. The length of time available to perform an action and the level of stress are the classic influential performance shaping factors (PSFs}.
Page 13 of27
F&O SR Not Met F&O Description Resolution CPS-PSA-004 Section 5.2 discusses the use of screening values For the HRA dependency analysis used for HEPs in order to identify cutsets with dependent HEPs.
performed as part of the 2011 PRA update, However, only twelve of the over 100 basic events modeling post-the post-initiator HEPs are set to 0.1 or initiator operator actions are listed in Table 5.2-1 as using higher (i.e., if the HEP is already >0.1 it is screening values to identify dependency. Of these, six use a value left at that higher value) and the model re-of 1.0E-02 and one uses a value of 1.0E-03. The remaining five quantified during the PRA update HR-G7 use a value of 0.1. It appears that all other HEPs are quantified development process to support HR-H3 with their nominal values. Use of such low probability values is identification of important operator action F&O 1-32 QU-A5 likely to result in combinations of dependent HEPs being omitted combinations in cutsets. Those QU-Cl by truncation values. Use of a sufficiently high value for HEPs is combinations are then explicitly discussed, QU-C2 required by SR QU-Cl and not using a sufficiently high value dependent events created, and then LE-El would result in an inadequate assessment of dependent HEPs.
included in the quantification. The HEPs are then left at the value of 0.1 or higher for the base model run. The recovery file identifies the dependent groups and also resets the independent HEPs to their actual probabilities. The HRA Notebook has been updated to document this process.
Page 14 of27
F&O SR Not Met F&O Description Resolution Several operator actions, which are evaluated with their values For the HRA dependency analysis set to nominal probability, are evaluated. For example refer to performed as part of the 2011 PRA update, Sections 3.4, 3.5, and 3.6 which evaluate actions related to the HRA, post-initiator HEPs are set to 0.1 makeup condensate. The failure probability of these actions is so or higher (i.e., if the HEP is already >0.1 it is low that the associated basic events would not appear in cutsets left at that higher value) and the model re-for dependency analyses. Similarly, several events are developed quantified during the PRA update for actions to start the standby condensate booster pumps or the development process to support associated lube oil pumps. No discussion of the dependence of identification of important operator action HR-G7 these actions on the need to restart a feedwater pump was combinations in cutsets. Those HR-H3 found. Particularly, would failure to start a condensate booster combinations are then explicitly discussed, F&O 1-33 QU-AS pump result in a low suction pressure trip of the feedwater pump.
dependent events created, and then QU-Cl included in the quantification. The HEPs QU-C2 are then left at the value of 0.1 or higher LE-El for the base model run. The recovery file identifies the dependent groups and also resets the independent HEPs to their actual probabilities. The HRA Notebook has been updated to document this process. This process ensures cutsets will not be inappropriately truncated out and eliminate the concern of modeling too fine a level of detail.
Solving the PRA models with some HEPs at nominal can result in This F&O overlaps with F&Os 1-32 and 1-33 cutsets with multiple operator actions being truncated out or and therefore the resolutions to those HR-G7 with the combined probability of all operator actions much below F&Os also resolves this F&O.
HR-H3 the lE-6 or SE-7 floor that the HRA notebook says is used.
F&O 1-34 QU-AS The peer review team quantified the PRA model with post-QU-Cl QU-C2 initiator HEPs set to 0.1 and identified a significant number of LE-El cutsets containing combinations of basic events representing operator action failure. These combinations were reviewed and a large number of combinations identified in this review were not included in the CPS HRA dependency evaluation Page 15 of27
F&O SR Not Met F&O Description Resolution Common cause factors used only non-staggered CCF values (CPS-CPS testing practices are staggered (at least PSA-002 Volume 2 section 4.2). This assumption could impact for on-line work), testing is performed in results in a non-conservative basis and may not be true for the context of divisional work weeks; as outage tested systems. The use of staggered testing could such, the use of staggered CCF approach is F&O 1-42 LE-El provide non-conservative results and is not supported with plant appropriate and is used in the model. Key experience. Identify, document and correct basis for use selected PRA components are typically tested online.
applied CCF factors.
Section 4.2 of Vol. 2 of the Component Data Notebook states "The Clinton Data Analysis utilizes the staggered component testing scheme".
Page 16 of27
F&O SR Not Met F&O Description Resolution The top ten cutsets are reviewed in Section 6.3.1 of CPS-PSA-014.
In the latest CPS FPIE PRA model the top 10 The frequency of these cutsets total to 7. 7E-07 per year or 1.4%
cutsets total to 20% of CDF. The top 10 of the 5.57E-06 CDF. The top ten accident sequences are accident sequences total 81% of CDF.
reviewed in Section 6.3.2 of CPS-PSA-014. These sequences total Review of the top 10 cutsets and accident 1.7% of CDF.
sequences is documented. A listing of the top 200 cutsets for both CDF and LERF are A "significant cutset" is defined in the ASME/ANS standard. One provided in the Quantification notebook. A of the criteria is a cutset that contributes to the summed sampling of cutsets and sequences are percentage of 95% of CDF with other additional criteria specified.
reviewed during the PRA Model review Other criteria than specified in the standard for "significant" can prior to model approval. This is considered be used if justified, however, no other criteria are specified in the sufficient. Refer to PRA Standard Inquiry documentation reviewed. Although some cutsets were reviewed,06-609 which states "the lists in [SY, QU, F&O 1-43 QU-Dl the review did not constitute a review of "significant cutsets" as etc.] are provided as examples of defined in the standard.
documentation forms or types that may be used to meet the documentation A "significant accident sequence" is defined in the ASME/ANS requirements of the PRA Element. They standard. One of the criteria is an accident sequence that should not be interpreted as specific contributes to the summed percentage of 95% of CDF with other requirements for the documentation."
additional criteria specified. Other criteria than specified in the standard for "significant" can be used if justified, however, no other criteria are specified in the documentation reviewed.
Although some cutsets were reviewed, the review did not constitute a review of "significant accident sequences" as defined in the standard. Perform and document a review of "significant" cutsets and accident sequences.
Page 17 of27
F&O SR Not Met F&O Description Resolution No review of nonsignificant cutsets or accident sequences was Exelon Risk management ER-AA-600-1015 identified. Perform and document a review of nonsignificant FPIE PRA Model Update guidance was cutsets and accident sequences.
followed which requires the following, "Review a sampling of the dominant and F&O 1-46 QU-DS non-dominant cutsets from the integrated results, based on the changes that have been made to the model." This review is performed and documented during PRA model updates.
No definition of significant accident progression sequence was The definition of significant accident found. This F&O originated from SR LE-G6. A definition is sequence progression is defined in F&O 1-48 LE-G6 required per this SR. Document that the definition from the PRA Appendix B (refer to pp. B-95 and B-122) of standard was used.
document CPS-PSA-000; the definition in Section 2 of the PRA Standard is adopted.
As described in the PRA standard, convergence can be considered The truncation study has been revised for sufficient when successive reductions in truncation value of one the 2014 PRA update for both CDF and decade result in decreasing changes in CDF or LERF, and the final LERF. The truncation limit was lowered for change is less than 5%. The truncation study in Section 3.1.2 of CDF from lE-11 to SE-13 to demonstrate CPS-PSA-013 clearly demonstrates that this convergence has not convergence. Lowering the truncation to occurred. For example, in Table 3.1-4 of CPS-PSA-013, there is a SE-14 will produce a <5% increase in CDF.
F&O 1-49 QU-B3 20.5% difference in CDF between lE-11 and lE-12 truncation The truncation for LERF was lowered to SE-limits. Similarly, convergence of LERF was not demonstrated with 14 from SE-13. Lowering the truncation a 2go,;6 increase from a truncation of 1.E-11 to lE-12. Complete further was not possible due to computer and document a convergence study to the criteria specified in the memory limitations. The truncation level is PRA standard.
6 orders of magnitude lower than the LERF result and is considered appropriate.
Page 18 of 27
F&O SR Not Met F&O Description Resolution The truncation study shown in Section 3.1 of CPS-PSA-013 shows This F&O overlaps with F&O 1-49 and the a 20% change in CDF from a truncation value of 1.0E-11 (used for same resolution applies.
the CPS PRA) to l.OE-012. The requirement of this SR is to use a truncation value that will not eliminate dependencies of significant cutsets or accident sequences. Since the definition of F&O 2-3 QU-B2 "significant" given in the PRA standard is the top 95% of cutsets (or accident sequences), the "significant" cutsets will change by increasing CDF by 20%. No basis for why the truncation value of 1.0E-11 maintains the dependencies of significant cutsets was identified in the documentation. Therefore, this SR is considered not met.
(This F&O originated from SR QU-B2)
The ISLOCA initiating event frequency values appear to be much The Clinton ISLOCA evaluation was lower than others in the industry and are developed based on old performed in accordance with NSAC-154 data. Furthermore, the frequency calculations do not appear to and is representative of the latest consider correlated data which is recommended in recent technology. The 2011 PRA updated the industry documents such as NUREG/CR-5744 and NU REG/CR-ISLOCA analysis to remove credit for F&O 2-7 IE-C14 5124. Also, recovery check valve failures for the valves that have isolation ofthe MOVs at ISLOCA pressures.
LE-04 failed to cause the initiating event is credited in the analysis. No The increase in frequency produced a <1%
basis for the credit is provided.
change in both CDF and LERF. Currently the Clinton total ISLOCA falls between the ISLOCA values for Quad Cities and Dresden as documented in Table 4-3 in the Initiating Event Notebook.
The pre-initiator HRA in the CPS [PRA Notebook] did not review The 2011 PRA project updated the pre-plant procedures or work practices specific to the plant. This F&O initiator identification and quantification originated from SR HR-A3. Identification of work practices that process. The PRA update included pre-HR-A3 involve a mechanism that simultaneously affects equipment in initiators that defeat multiple systems or F&O 2-14 LE-El either different trains of a redundant system or diverse systems is trains. The CPS HRA uses the system required by the SR.
analysis identification process which involves reviewing In Service Testing (IST) procedures and general post-maintenance test procedures.
Page 19 of27
F&O SR Not Met F&O Description Resolution Review of HRA notebook did not reveal rule that discusses not The CPS 2011 PRA update included pre-screening out activities that could impact multiple trains of a initiators that defeat multiple systems or redundant SSC or diverse SSCs. Furthermore, no evidence could trains. As part of the pre-IE HEP F&O 2-16 HR-B2 be found that that the process established by Exelon RM Best development IST procedures were LE-El Practice BP-005 was followed.
reviewed to find any restoration errors that could affect multiple systems or trains. The HRA documentation has been updated to better discuss these details.
Per Section 2.3.2.1.1 of CPS-PSA-004, the pre-initiators were The 2011 PRA project updated the pre-identified by'... it was judged that the best use of resources is to initiator identification and quantification identify the leading candidates for risk significant pre initiators process. The CPS HRA uses the system HR-Al and not to expend substantial effort to quantify other preinitiator analysis identification process which F&O 3-13 HR-A2 HEPs for which no data are available and for which the involves reviewing IST procedures and LE-El contribution can be expected to be small.' The section states a general post-maintenance test procedures.
methodology 'when resources become available', the identification should be performed in a way that would meet the SR.
Limitations in the LERF analysis that would impact applications In subsequent PRA updates documentation were not identified.
was enhanced to explicitly indicate that there are no limitations in the LERF analysis unique to Clinton that would affect applications. Uncertainties with regard to F&O 3-22 LE-GS severe accident phenomena and plant response (e.g., equipment survivability and crew actions) are addressed in the uncertainty evaluation performed for the 2014 Clinton model in Appendix B of the Summary Notebook.
Page 20 of27
F&O SR Not Met F&O Description Resolution The PRA documentation uses Maintenance Rule data to quantify The peer reviewer found documentation component unavailability data. No comparison of the was lacking to substantiate the use of Maintenance Rule function definitions to the PRA component Maintenance Rule and MSPI data for PRA boundaries or of the Maintenance Rule definition of available updates. Exelon document ER-AA-310-compared to the PRA was identified in the documentation.
1004, Maintenance Rule - Performance Therefore, it is not possible to determine if the unavailability Monitoring provides details on DA-Cll values used in the PRA data analysis are correct.
Maintenance Rule unavailability definitions.
F&O 4-2:
LE-El These rules are found to lead to unavailability tracking consistent with PRA unavailability criteria. Maintenance Rule unavailability is only used where Maintenance Rule functions are mapped appropriately to the PRA train or component. The PRA documentation has been updated to reflect this.
The PSA Deterministic Calculations Notebook, CPS-PSA-007 does The Success Criteria have been reviewed.
not have a comparison of results with those of the same analyses Comparisons of the resulting success performed for similar plants.
criteria were reviewed and compared with other similar plants using the MAAP F&O 4-5:
SC-BS computer code as a basis, with NU REG-1150 BWR/6 plant results, and with NEDO 24708A (GE report on core damage prevention success criteria). No anomalies are identified in the Clinton success criteria.
Page 21of27
F&O SR Not Met F&O Description Resolution Basic events in Table C.2-1 appear to be grouped by Type Code, For the CPS 2011 PRA Update, the data but not all Basic Events within some 'groups' have the same groups were reevaluated based on probability. As examples, the following are noted component similarities in design, safety (System/Component/ Failure Mode): AP/CB/K, AP/CB/U, categorization, normal operating state, and SP/AT/F, VY/FN/A water type. A new Table C.2-1 shows the F&O 4-11 LE-El results of this analysis. Table C.2-1 also identifies systems for which the type code applies. Groups were then updated with plant specific data (where available).
All basic events with the same type code are assigned the same probability.
CPS-PSA-010, Rev. 2 has no discussion of how grouping was done For the CPS 2011 PRA Update, the data to avoid including outliers in the definition of a group. Data groupings were reevaluated based on grouping process is not clear regarding how generic data was component similarities in design, safety updated by failure mode and grouped by mission type and by categorization, normal operating state, and system. Including outliers, either by including exposure time or water type. A new Table C.2-1 shows the failures, could skew the data results. Identify outliers that are results of this analysis. These groups were included in the data analysis and provide discussion of how then updated with plant specific data. A F&O 4-12 LE-El grouping was done and include discussion of how outliers were review of the exposure time and failures excluded from the definition of a group.
was performed to ensure applicability. If an outlier was found (e.g., the data reflects more than a factor of three unavailability above the average), then a specific assessment of that system or component was required separate from the other similar systems or components.
Page 22 of27
F&O SR Not Met F&O Description Resolution CPS-PSA-010, Vol. 1, App. G.8 and Table G.8-1 identify how For the 2011 PRA update, the coincident coincident corrective maintenance unavailability was included in maintenance was revised with plant specific the model. Assumption of 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> per 10 years for all data on coincident maintenance of multiple combinations of coincident corrective maintenance of selected PRA related systems. Only planned and DA-Cl4 components appears to be conservative when compared to the repetitive coincident maintenance events F&O 4-18 LE-El unavailability of the individual components but no justification is were considered consistent with the latest provided for selection of this value. Coincident unavailability that PRA Standard. The 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> in 10 years is no results from planned, repetitive activities should be based on longer used.
actual plant experience if available. Provide justification for selection of 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> in 10 years for assumed planned maintenance.
The contributors and sequences considered is much less than The definition of significant accident what would be required in order to meet the defined limits as sequence progression is defined in stated in the ASME standard. The use of the term "significant" in Appendix B (refer to pp. B-95 and B-122) of F&O 4-23 QU-F6 the CPS document CPS-PSA-000, the definition in Quantification notebook must be defined in accordance with the Section 2 of the PRA Standard is adopted.
ASME standard. The use of the term 'significant' is not defined in The CPS model.
Demand and run-time data is documented in Table C.2-3. The The 2011 CPS PRA update obtained plant data sources used were provided in the table. A footnote in the specific data for the data collection period.
table indicates that plant-specific data were not used. Some of This data is based on actual plant DA-CG the data is based on the MSPI basis document which should experience and includes demands based on F&O 5-2:
DA-ClO provide a basis for the period of the basis document. However, surveillance tests and maintenance events.
LE-El data was extrapolated beyond the period of time considered in The PRA documentation has been revised the basis document. No consideration of surveillance tests, and Appendix F of the Component Data maintenance acts, etc., was documented. There was no Notebook contains all of the plant specific accounting for the exclusion of post maintenance demands.
data obtained from the site.
Page 23 of27
F&O SR Not Met F&O Description Resolution Actual plant experience, practices and plant maintenance plans The 2011 CPS PRA update obtained plant were not used to base the number of surveillance tests and specific data for the data collection period.
maintenance acts.
This data is based on actual plant experience and includes demands based on F&O 5-3:
LE-El surveillance tests and maintenance events.
The PRA documentation has been revised and Appendix F of the Component Data Notebook contains all of the plant specific data obtained from the site.
Plant-specific records were not used to determine the time that Typically systems that have one or more components were configured in their standby status.
components in standby while others are running will rotate the equipment for even wear on all the components according to the system managers. For systems with a F&O 5-4:
LE-El preferred lineup (e.g., condensate), the standby probability is adjusted to reflect actual plant operation. Estimates are used for these probabilities. Plant specific records would have a negligible impact on results. SR DA-C8 allows the use of estimates for Capability Category I.
Median values were used for HEPs - need to use mean values.
The EPRI HRA Calculator is now used to quantify the probabilities of the HEPs in the Clinton model. The HRA calculator uses the mean values from the TH ERP tables for F&O 5-7:
HR-06 quantification. The transition to using the LE-El EPRI HRA Calculator did not represent a methodology change, but the results produced by the HRA Calculator are considered to be industry acceptable for use in the PRA.
Page 24 of 27
F&O SR Not Met F&O Description Resolution No evidence (other than the final HEP) of accounting for quality The Clinton PRA HRA Analysis now uses the of procedures and quality of the human machine interface of HFE EPRI HRA Calculator to calculate both the HR-04 was found - values are provided but input for assessment is not pre-and post-initiator HEPs. The HRA F&O 5-10 HR-05 provided. Documentation of process and a flow chart (section 4.3 Calculator takes into account the clarity and LE-El of CPS-PSA-004) and final pre-initiator HEP values - would expect simplicity of the procedures and the some justification (inputs) for final basis of the individual HEP.
required steps. The current HRA documentation shows the calculator inputs for all pre-and post-initiator HEPs.
Credit has been taken for the capability to close valves against In a subsequent PRA update, the credit for ISLOCA dynamic loads. The evaluation of closure capability only isolating the MOVs in ISLOCA was removed.
IE-C14 evaluated normal loads and delta pressures.
The ISLOCA and Initiating Event notebook F&O 5-12 LE-04 have been updated to reflect this change.
The CDF and LERF results were not significantly affected by this minor change (i.e., delta CDF and LERF < 1%).
System notebooks, specifically fault trees associated with the A review of plant specific coincident ECCS Systems (division 1 and division 2), do not contain divisional maintenance events was performed for the planned concurrent maintenance activities. For example RHR B CPS 2011 update. Coincident maintenance SY-A20 and RHR C.
events were updated based on plant F&O 6-7:
DA-C14 specific planned and repetitive coincidence maintenance activities during the data period. Several coincident maintenance events were added to the model (including RHR Band RHR C).
Page 25 of27
F&O SR Not Met F&O Description Resolution ECCS Room Cooling applications document the need for room Each ECCS room has a design basis cooling. Credit is provided for the opening of doors (LPCS pump calculation supporting system operation room for example). No formal engineering analyses were with room coolers in service and doors identified that supports this application.
shut. In addition a calculation was performed for the 2011 PRA update (Axioma Report 2009-0048) that confirms the success of opening the RHR B room doors for SX failures. This calculation is used as the basis of allowing credit for operators opening room doors as a success F&O 6-8:
SY-86 of room cooling. The analysis used a computational fluid dynamics (CFO) code (FLUENT), which models density driven airflows through open doorways, to demonstrate the viability of open doors as a room cooling method. Because of the similarity of ECCS rooms (e.g. similar large motor driven pumps, with similar doorways) the physical processes are believed to be similar and the door open cooling method was applied in the PRA model to the other ECCS rooms.
Page 26 of27
F&O SR Not Met F&O Description Resolution CPS-PSA-015 Section 7 and CPS-PSA-013 document the overall In subsequent PRA updates the Level 2 results of the Level 2 assessment. While it is assumed that results results have been reviewed for presentation involved the review for reasonableness, review of reasonableness and documented in the the associated documentation did not explicitly reveal a check for Summary Notebook. The CPS Level 2 PRA reasonableness. The model owner was asked about this results are plant specific and are strongly comparison and replied that no such comparison was performed.
influenced by the plant specific contributors No comparison was noted to have been made to similar plants.
to the risk profile from Level 1 accident sequences.
The available PRA Level 2 results are mostly limited to LERF analyses. The older NU REG-1150 analysis was distorted in its F&O 6-10 LE-F2 calculation of the Grand Gulf Generating Station (GGNS) by attributing 99% of the risk profile to Station Blackout. Therefore, the availability of one-to-one correspondence with other studies is not possible.
Nevertheless, the CPS results have been compared qualitatively with PRA analyses performed by other Mark Ill plants and similar PRA analyses performed by the NRC to ensure that the latest techniques, data, and deterministic results are properly incorporated into the CPS Level 2.
Page 27 of27