ML12061A017

Supplemental Response to Request for Additional Information Regarding Relocation of Specific Surveillance Frequency Requirements to a Licensee Controlled Program

ML12061A017
Person / Time
Site:	Millstone
Issue date:	02/22/2012
From:	Hartz L Dominion, Dominion Nuclear Connecticut
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
11-696A, TAC ME6695
Download: ML12061A017 (20)

Text

Dominion Nuclear Connecticut, Inc.

5000 Dominion Boulevard, Glen Allen, Virginia 23060 *Dominion Web Address: www.dom.com February 22, 2012 U. S. Nuclear Regulatory Commission Serial No. 11-696A Attention: Document Control Desk NSSL/WDC RO Washington, DC 20555 Docket No. 50-423 License No. NPF-49 DOMINION NUCLEAR CONNECTICUT, INC.

MILLSTONE POWER STATION UNIT 3 SUPPLEMENTAL RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION REGARDING RELOCATION OF SPECIFIC SURVEILLANCE FREQUENCY REQUIREMENTS TO A LICENSEE CONTROLLED PROGRAM (TAC NO. ME6695)

By letter dated July 5, 2011, as supplemented by letter dated September 12, 2011, Dominion Nuclear Connecticut, Inc. (DNC) submitted a license amendment request (LAR) for Millstone Power Station Unit 3 (MPS3). The proposed amendment would relocate certain technical specification (TS) surveillance frequencies to a licensee controlled program by adopting Technical Specification Task Force (TSTF)-425, Revision 3, "Relocate Surveillance Frequencies to Licensee Control - RITSTF [Risk-Informed Technical Specification Task Force Initiative] 5b." The proposed change would also add a new program, the Surveillance Frequency Control Program, to the TSs, in accordance with TSTF-425. TSTF-425 is approved for use by the Nuclear Regulatory Commission (NRC). In a letter dated December 8, 2011, the NRC transmitted a request for additional information (RAI) to DNC related to the LAR. DNC responded to the RAI in letter dated January 19, 2012. In a telephone conference with the NRC on February 2, 2012, DNC agreed to supplement the response to the RAI by February 23, 2012. provides DNC's supplemental response.

If you have any questions regarding this submittal, please contact Wanda Craft at (804) 273-4687.

Sincerely, L. N. Hartz Vice President - Nuclear Support Services

Attachment:

1. Supplemental Response to Request for Additional Information Regarding Relocation of Specific Surveillance Frequency Requirements to a Licensee Controlled Program A4-c)!

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Page 2 of 2 Commitments made in this letter:

1. None cc: U.S. Nuclear Regulatory Commission Region I 475 Allendale Road King of Prussia, PA 19406-1415 C. J. Sanders Project Manager - Millstone Power Station U.S. Nuclear Regulatory Commission One White Flint North 11555 Rockville Pike Mail Stop 08-B3 Rockville, MD 20852-2738 NRC Senior Resident Inspector Millstone Power Station

Serial No. 11-696A Docket No. 50-423 ATTACHMENT 1 SUPPLEMENTAL RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION REGARDING RELOCATION OF SPECIFIC SURVEILLANCE FREQUENCY REQUIREMENTS TO A LICENSEE CONTROLLED PROGRAM DOMINION NUCLEAR CONNECTICUT, INC.

MILLSTONE POWER STATION UNIT 3

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 1 of 17 By letter dated July 5, 2011, as supplemented by letter dated September 12, 2011, Dominion Nuclear Connecticut, Inc. (DNC) submitted a license amendment request (LAR) for Millstone Power Station Unit 3 (MPS3). The proposed amendment would relocate certain technical specification (TS) surveillance frequencies to a licensee controlled program by adopting Technical Specification Task Force (TSTF)-425, Revision 3, "Relocate Surveillance Frequencies of Licensee Control - RITSTF [Risk-Informed Technical Specification Task Force Initiative] 5b." The proposed change would also add a new program, the Surveillance Frequency Control Program (SFCP), to the TSs, in accordance with TSTF-425. TSTF-425 is approved for use by the Nuclear Regulatory Commission (NRC). In a letter dated December 8, 2011, the NRC transmitted a request for additional information (RAI) to DNC related to the LAR. DNC responded to the RAI by letter dated January 19, 2012. In a telephone conference with the NRC on February 2, 2012, DNC agreed to supplement the response to the RAI.

This attachment provides DNC's supplemental response.

Question 1 The licensee has identified five "gaps"in the MPS3 probabilisticrisk assessment (PRA) model's conformance to the American Society of MechanicalEngineers internalevents PRA standard. These gaps collectively cover eight deficiencies in meeting the supportingrequirements (SRs) of the PRA standard,CapabilityCategory II. The "description"and "currentstatus"columns in Table 1, Attachment 2, of the July 5 submittal, adequately describe these gaps. However, the "importanceto application" column only restates that the gap is a model logic issue and that sensitivity studies will be conducted. The significanceof these gaps to the overall PRA results and to the risk calculationssupporting changes to surveillance frequencies is not discussed, and must be addressedin order for the NRC staff to reach a conclusion on the significance of these gaps as related to this amendment request.

1. Provide a discussion of the significance of each of the five gaps identified in Table I of Attachment 2. Where applicable,provide a quantitative assessment of the sensitivity of core damage frequency (CDF) and large early release frequency (LERF) results. In addition, the following specific concerns should be addressedin your response:

Gap 1: The lack of modeling ventilation dependencies was specifically identified as a recommendationfrom the NRC staff's review of the MPS3 Individual Plant Examination (IPE). According to the model update timeline provided in the July 5, submittal, this was completed in December 1995. It is not clear how this same issue arisesin a 2007 self assessment. What is unique about these three specific ventilation subsystems which led to them not being modeled as a requiredsupport system following the 1995 IPE?

Gap 2: Changes to surveillance test intervals can directly impact an associatedpre-initiatorhuman errorprobabilityby reducing the opportunitiesfor

Serial No. 11-696A Docket No. 50-423, Supplemental RAI Response for TSTF-425 Attachment 1, Page 2 of 17 identification and restorationof a system misalignment. What is the scope of pre-initiatorhuman errors currently in the PRA model, and what is the basis for theirinclusion in the model, absent a review of plant proceduresand practices? Is resolution of this gap expected to add to the pre-initiatorscope significantly?

Gap 4: The NRC staff review of the internal flooding analysis for the IPE also identified the lack of discussion regardingmaintenance-inducedflooding.

What internal flooding sources are currently consideredand what is the basis for their frequencies? Is resolution of this gap expected to increase the contribution of internalfloods to CDF and LERF significantly? What is the schedule for addressingthese contributors?

DNC Supplemental Response In addition to the response provided by DNC in letter dated January 19, 2012, the following supplemental information is provided.

Table 1 of this supplement provides the gap status, including a discussion on the importance of the gap to this application. In addition, Table 1 provides the significance of these gaps to the overall PRA results analyzed as part of draft PRA model, M310A.

Modeling gaps are classified as either high risk significance or low risk significance (based on their Fussell-Vesely (FV) importance value and using a threshold value of 5E-3), or they are given a classification of no impact to the application.

In accordance with NEI 04-10, Revision 1, "Risk-Informed Technical Specifications Initiative 5b, Risk-Informed Method for Control of Surveillance Frequencies," if there are open gap analysis items when compared to the ASME Standard Capability Category II that would impact the results of the SFCP PRA application assessment, then sensitivity studies should be performed. The types of sensitivity studies that will be performed, following implementation of the SFCP, are identified in Table 1.

Question 2 The licensee identified an industry peer review being conducted in 1999, and stated that the 'A" and "B" findings from the review have been addressed. Are there other remaining open findings, from that review, which are relatedto different issues than those identified in the self-assessment conducted in 2007? Describe any such findings and their potentialimpact on this application.

DNC Supplemental Response In addition to the response provided by DNC in letter dated January 19, 2012, the following supplemental information is provided.

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 3 of 17 Table 2 of this supplement provides the description, current status and importance to this application for each of the level "C" facts and observations (F&Os). Since these are all documentation issues, there is no modeling significance associated with these F&Os.

Question 3 The self-assessment results identified 39 SRs, related to documentation and uncertainty evaluations, which do not meet PRA CapabilityCategory II. The details of these SRs are not discussed in the July 5, submittal. The characterizationof uncertaintiesis essential to assure that PRA results supporting changes in test frequencies are understood by the independent decision-making panel in order to adequatelyimplement TSTF-425. Further,properdocumentation is essential for maintainingthe PRA model and understandingthe results. Finally, the NRC staff is concerned about the number of SRs in these specific areas, and notes that its review of the MPS3 IPE submittal also identified several specific areas where documentation needed enhancement.

Describe the nature of the deficiencies in documentation and uncertaintyevaluations, including the currentstate of the PRA documentation and uncertainty analyses, and schedule for resolution of these items.

DNC Supplemental Response In addition to the response provided by DNC in letter dated January 19, 2012, the following supplemental information is provided.

Table 3 of this supplement provides the description, current status and schedule for resolution for each of the not-met supporting requirements identified as documentation issues.

Table 4 of this supplement provides the description, current status and schedule for resolution for each of the not-met supporting requirements identified as assumption and uncertainty documentation issues.

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 4 of 17 Table 1 Status of Identified Modeling Gaps to NEI 00-02 and Capability Category II of the ASME PRA Standard

.NEI Title Description'. Elemnent/ CurnSau I Comment impprtance'taA~ppicabon~~

ASMESR Gap #1 Engineering analysis is IE-B3 The following Heating, Low Risk Significance (based on RPCCW HVAC as discussed below):

required to screen SY-A19 Ventilation, and Air Conditioning support systems for SY-B6 (HVAC) dependencies are not Importance to application is as follows:

accident initiation or modeled:

• EDG Sequencer HVAC: No Impact to Application - Based on a room mitigation. 0 Emergency Diesel heatup calculation, no PRA modeling dependency is required for room Generator (EDG) cooling of the EDG sequencer.

Sequencer HVAC

" Normal Switchgear HVAC Following implementation of the SFCP, a sensitivity study will not be

" Reactor Plant Component required.

Cooling Water (RPCCW)

Pumps HVAC

• Normal Switchgear HVAC: No Impact to Application - There surveillance testing or requirement for the normal is no switchgear ventilation.

This 6.9kV bus supplies the reactor coolant pumps (RCP) and condensate pumps.

Following implementation of the SFCP, a sensitivity study will not be required.

• RPCCW HVAC: Low Risk Significance - There is a low risk significance impact to modeling HVAC dependency for the RPCCW pumps and charging pumps.

Following implementation of the SFCP and in accordance with PRA procedures, a sensitivity study will be performed using the HVAC dependency for the RPCCW and charging pumps for both accident initiation and mitigation.

Gap #2 The Type A human HR-Al As part of a 2007 model No Impact to Application:

reliability analysis revision, numerous pre-initiator (HRA) identification human error probabilities The 2007 Dominion self-assessment specifically states that surveillance employed an initial (HEPs) were added to the procedures and practices have been adequately reviewed for this supporting review of PRA system model after a systematic review requirement, therefore, the resolution of this gap is not considered impacting to P&IDs to identify of each system was performed. the SFCP application.

components potentially Each system notebook contains susceptible to Type A the reviews and the resulting Following implementation of the SFCP and in accordance with PRA procedures, realignment errors, pre-initiator HEPs that were a sensitivity study will not be required.

followed by a review of determined to be required. The surveillance HR.1 notebook documents the procedures to identify quantification of the HEPs. No

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 5 of 17 Table I Status of Identified Modeling Gaps to NEI 00-02 and Capability Category IIof the ASME PRA Standard Title Description Element/ Current Statusf CommentI 1 mportance to Applicatio:

_____________ASMESR______________ __________

those that require screening values were used.

realignment. The This process did not specifically documented look at activities other than methodology doesn't surveillances because only the appear to discuss surveillance activities are within realignment of the scope of this application.

equipment outside its normal operational or Resolution of this gap is not standby status for expected to add to the scope of activities other than pre-initiator HEPs because surveillance. when maintenance is performed that re-aligns a system, post-Include in the Type A maintenance testing (i.e., TS assessment surveillance) is performed to realignment of establish system operability. If equipment outside its the surveillance does not normal operational or introduce a potential pre-initiator standby status for HEP, then a successful test will activities other than prove that the system is aligned surveillance, i.e., properly. If the surveillance maintenance and does introduce a potential pre-inspection activities, or initiator HEP, then as stated document why these previously, the current PRA activities do not pose addresses surveillances that re-credible Type A align a modeled system.

failures.

Gap #3 Develop alignment DA-C8 The current PRA model, Low Risk Significance:

specific basic event M308A, meets Capability values based on actual Category I, in that it assumes As part of the PRA model update, the component/train alignment basic events plant operating an overall average distribution will be based on actual operating history. The risk significance of these experience (OE). of system alignments. The alignment basic events is low (i.e., FV < 5E-3).

estimates used are reasonable.

However, this approach does Following implementation of the SFCP and in accordance with PRA procedures, not meet Category II a sensitivity study will be performed using the alignment specific basic event requirements. values identified and documented in the updated PRA model, M31OA.

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 6 of 17 Table I Status of Identified Modeling Gaps to NEI 00-02 and Capability Category II of the ASME PRA Standard

.NEI motnetAplatn Title Descripir Eeet Cuet-ASME SR Y ,~ <

Gap #4 Identify flooding IF-B2 Need to incorporate internal Low Risk Significance:

sources that may flooding frequencies associated result in fluid release with non-piping failures (e.g., The current PRA model includes maintenance-induced internal flooding initiating due to failure of other expansion joints, bellows, and events. Internal flooding contributes approximately 2% to overall core damage (non-piping) inadvertent sprinkler actuation). frequency (CDF). Resolution of Gap #4 is not anticipated to identify any new components, human internal flooding scenarios. The majority of non-piping components (e.g.,

error during pumps, valves, tanks) are identified and included in the internal flooding maintenance, or analysis. The remaining non-piping failures (expansion joints, bellows and inadvertent sprinkler The schedule to address this inadvertent actuation of fire protection system) are bounded by already analyzed actuation. gap will be performed in flow rates. Since the remaining non-piping failures make up a small percentage accordance with PRA policies of the overall system piping failures, any changes in the internal flooding and procedures. initiating event frequencies will not have a significant impact to the overall CDF/Large Early Release Frequency (LERF) or impact the SFCP.

Following implementation of the SFCP and in accordance with PRA procedures, a bounding sensitivity study will be performed by doubling the internal flooding initiating events.

Gap #5 Include realistic LE-C2a The SAMGs have not been High Risk Significance:

treatment of feasible LE-C6 completely incorporated into the operator actions MPS3 Level 2 analysis. Need As part of the PRA model update, M31 OA, a post-initiating event operator action following the onset of to credit SAMG operator actions has been developed to address RCS depressurization before an induced SGTR core damage for depressurization of the RCS event. Modeling of this operator action results in a decrease in the overall LERF consistent with to allow low pressure injection (i.e., removal of conservatism in the LERF value).

applicable procedures, to prevent induced steam e.g., Emergency generator tube rupture (SGTR). Following implementation of the SFCP and in accordance with PRA procedures, Operating Procedures a sensitivity study will be performed using the operator action for RCS (EOPs), Severe depressurization.

Accident Management Guidelines (SAMGs),

proceduralized actions, or Technical Center guidance.

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 7 of 17 Table 2 Status of Identified Gaps identified in the 1999 MPS3 PRA Peer Review

______ Description~ /jCurrent StatuJs I Commen~t Importance to Application AS-6 There are examples of safety functions and Table 6-2 (Plant-Specific Assumptions and Uncertainties) of No Impact to Application consequential events that either have been omitted, or the Accident Sequence notebook, AS.l, provides an impact for which the technical basis for exclusion is not well assessment for the following: Documentation issue only.

understood. Examples include:

Assumption/Uncertainty Impact Assessment a) whether the potential for pressurized thermal shock a) Pressurized thermal shock This assumption could induced failures of the reactor pressure vessel were events are not modeled, result in a small non-considered during severe overcooling transients, conservatism in the model, particularly in the b) whether pressure induced SGTR during secondary steam line break depressurization was considered, modeling.

c) Failure of letdown and RCP Exclusion of these c) whether the potential for consequential bypasses seal return line isolation is not bypass paths will result such as letdown isolation and seal return line isolation considered as a potential in a small were considered, and containment bypass event, underestimation of interfacing systems loss d) whether the need to isolate the accumulators of coolant accident following injection was considered in development of scenarios. However, the event tree logic. the probability of these bypass paths remaining unisolated should be much lower than those of other bypass scenarios.

d) Failure to isolate the RCS Exclusion of these accumulators following failures would not be discharge is not modeled, expected to have any significant impact on the CDF or LERF results.

Item b) on pressure induced SGTR has been addressed in the current MPS3 PRA model via the Modular Accident Analysis Program (MAAP) analysis, which accounts for induced SGTR during secondary depressurization.

This F&O will be closed in the M310A model.

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 8 of 17 Table 2 Status of Identified Gaps identified in the 1999 MPS3 PRA Peer Review F&O Description - - - - - - -- Current Status /Co~mment Importance to Application, DE-2 Some spatial dependencies are considered, but to Both the SW strainers (which are under gate SWX700, Loss of No Impact to Application varying levels of detail. The Service Water (SW) SW Due to SW Self-Cleaning Strainer Blockage) and system notebook discusses SW intake blockage circulating water traveling screens (which are under gate Documentation issue only.

(biofouling) but dismisses this as low probability given "LOSS-OF-ALL-CW') have been modeled. The common available time and operator intervention. However, the cause failure (CCF) has been modeled in each case.

PRA engineer discussed an event in which a component was improperly modified, resulting in the This F&O will be closed in the M31OA model.

intake of sea grass through the strainers, which ended up in numerous components served by SW. It was not clear that this was captured in the common cause SW failure probability.

L2-3 The containment pressure capacity analysis, although The Level 2 severe accident transition to MAAP 4 analysis is No Impact to Application more than 16 years old, is still viewed as state of the art documented in PRA99NQA-01694S3 Rev. 0. There was no and should be useful in support of updates to the Level update of the containment structural analysis. Documentation issue only.

2 probabilistic safety assessment. As most aspects of this evaluation are structural in nature, the potential for The MPS3 Probabilistic Safety Study (August 1983) structural continued applicability is very good. When the updated failure at the 95th percentile is 132 psi. A 1999 MAAP input L2 severe accident analysis is updated with MAAP, the file has the containment structural integrity challenged (i.e.

assumed initial conditions and thermal hydraulic loads failed) at 126 psi.

assumed in this analysis should be reviewed as well as possible plant modifications for continued applicability. The recommendation from the Westinghouse Owner's Group This work could then be incorporated into the updated PRA Peer Review was to update MPS3's older containment Level 2. structural failure analysis. However, there is no issue with using an older analysis, which is still considered valid and not disputed as incorrect.

The MAAP model has been updated over the years to address new software versions and updating of thermal hydraulic analysis (i.e., success criteria). These updates include a review of initial conditions and thermal hydraulic loading. This work has been performed by both Fauske and Associates, Inc and Dominion PRA engineers.

This F&O will be closed in the M31 OA model.

MU-5 The PRA model control process does include a review A model freeze date is a good idea but is not required by the No Impact to Application of design changes, but there is no record of the date ASME PRA Standard. This F&O will be closed inthe M310A through which plant changes and operating experience model. Documentation issue only.

data have been incorporated into the model.

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 9 of 17 Table 2 Status of Identified Gaps identified in the 1999 MPS3 PRA Peer Review F&O Description. K CurrentStatus I Comment SY-6 Although system dependencies on HVAC have been EQ limits are conservative. SR SY-A20 Category II requires No Impact to Application considered, the basis for these dependencies may be taking credit for system or component operability if an analysis overly conservative. In most cases, the room heatup exists to demonstrate that rated or design capabilities are not Documentation issue only.

calculations had established a Maximum Abnormal exceeded. Using the EQ limits meets this supporting Excursion (MAE) limit of 120°F for equipment requirement.

survivability. This came from the Environmental Qualification (EQ) program, which allows for such an This F&O will be closed in the M310A model.

excursion once in the life of the equipment, yet the equipment is still considered qualified for the remainder of life of the plant. Thus, establishing this temperature as a failure limit may be conservative.

During a plant walkdown, it was observed that some of the rooms containing the affected equipment are reasonably large, and may have considerable ability to disperse the heat generated by the running equipment.

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 10 of 17 Table 3 Status of Identified Documentation Gaps to NEI 00-02 and Capability Category IIof the ASME PRA Standard ASME! Description  : Current Status Comment Schedule for Resolution

~SR DA-D4 The Data Analysis Notebook, DA.2, documentation This supporting requirement is met using method (e). Closed should be expanded to discuss the other consistency/applicability tests on the Bayesian-updated events as defined in this supporting requirement.

DA-D6 Provide documentation in the Data Analysis Notebook, The PRA Guidance and Reference Document (GaRD), Will be closed with release of DA.3, to demonstrate that the generic factors were ProbabilisticRisk Assessment Proceduresand Methods: PRA model M310A.

reviewed against plant-specific OE to ensure that the PRA Data Analysis-Common Cause Modeling (NF-AA-generic factors are appropriate. PRA-101-2062) provides instruction for generic CCF data.

Specifically, the generic CCF data should be examined by the analyst to ensure that it is appropriate for the Dominion model and to ensure that the component boundaries assumed for the CCF failures are consistent with those used for the independent events (i.e., plant-specific OE).

HR-B1 1) Generate and document in HR.1 a list of rules Pre-initiator HEP screen criteria are contained in GaRD, Will be closed with release of established for screening individual activities from Human Reliability Identification (NF-AA-PRA-101-2051). PRA model M310A.

further consideration. 2) For activities that have no or The PRA model notebook, HR.1, for PRA model M310A, insignificant impact on PRA results, provide quantitative documents the criteria used when screening activities from bases for screening activities in terms of CDF/LERF further consideration based on a negligible potential to impact. cause a restoration or calibration error. SR HR-R1 does not require a "quantitative" basis for negligible potential pre-initiator HEP due to restoration or calibration error.

HR-B2 Document in the appropriate SY.3 notebooks Each system notebook (SY.3) contains identification of pre- Will be closed with release of considerations of activities that could simultaneously initiator HEPs. The pre-initiator HEP analysis is PRA model M310A.

have an impact on multiple trains of redundant or documented in the Human Reliability notebook, HR.1. Any diverse systems. If any such activities exist, also pre-initiator HEP that could simultaneously impact multiple document that they have been retained and modeled by trains of redundant or diverse systems are documented, the HRA. within the SY.3 and HR.1 notebooks.

HR-C2 Include consideration of modes of unavailability Upon completion of updating the pre-initiator HEPs Will be closed with release of resulting from failure to restore: 1) automatic (Reference Gap 2 in Table 1), the process will be PRA model M310A.

realignment or 2) electrical power. documented in the HR.1 notebook in accordance with GaRD NF-AA-PRA-101-2051, which will address failure to restore either automatic realignment or electrical power.

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 11 of 17 Table 3 Status of Identified Documentation Gaps to NEI 00-02 and Capability Category II of the ASME PRA Standard ASME Description Cu.rrent*StatuslComment, -Schedule for Resolution:

HR-D3 As part of the pre-initiator HEP evaluations, document The PRA GaRD, Human ErrorProbabilityAssessment Will be closed with release of the evaluation process of the following plant-specific (NF-AA-PRA-101-2052), provides technical guidance to PRA model M310A.

relevant information: a) the quality (including format, specifically address supporting requirement HR-D3. Upon logical structure, case of use, clarity, and completion of the PRA model update, M31 OA, the process comprehensiveness) of written procedures and the for evaluating pre-initiator HEPs (Reference Gap 2 in Table quality (eg., configuration control process, technical 1) will be documented in the HR.1 notebook.

review process, training process, and management emphasis on adherence to procedures) of administrative controls (for independent review); (b) the quality (e.g., adherence to human factors guidelines and results of any quantitative evaluations of performance per functional requirements) of the human-machine interface, including both the equipment configuration, and instrumentation and control layout.

HR-El Provide documentation of the process used to identify The PRA GaRD, Recovery Actions (NF-AA-PRA-101- Will be closed with release of MPS3 key human response actions (i.e., the Type C 2053), provides technical guidance for the identification PRA model M310A.

human reliability events), and modeling of recovery actions. This will be documented in PRA model, M310A.

HR-G6 Document a review of the Human Failure Events and This supporting requirement is met through the Will be closed with release of their final HEPs relative to each other to confirm their dependency evaluation performed in accordance with PRA model M310A.

reasonableness given the scenario context, plant GaRD, NF-AA-PRA-101-2053 and documented in model history, procedures, operational practices, and notebook, HR.4, HEP Dependency Analysis.

experience.

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 12 of 17 Table 3 Status of Identified Documentation Gaps to NEI 00-02 and Capability Category 11of the ASME PRA Standard ASMVE Description Current Status / Comment ~Schedule for Resolution IE-A6 Interview plant personnel to determine if potential MPS3 There is currently no documentation of previous informal All issues potentially affecting the initiating events have been overlooked, discussion between plant personnel and the PRA PRA model are captured in the engineers on modeled initiating events in the MPS3 PRA. PRA Configuration Control The PRA staff has worked closely over the years with plant Program, NF-AA-PRA-410, along personnel in support of risk-informed activities (e.g., with their estimated impact on Maintenance Rule (a)(4), Mitigating Systems Performance model results. This program Indicators, Notice of Enforcement Discretion, Significance prioritizes implementation of each Determination Process). A process has been developed to issue as: asap (as soon as document discussions with plant personnel going forward. possible), next model update, or voluntary (i.e., not required for procedure or regulatory compliance). This unmet supporting requirement, which has been assigned a low priority based on its low risk significance, will be addressed in a future model update.

IE-A7 Include in the Initiating Event (IE) evaluation a review of There is no documented evidence of a review for plant- All issues potentially affecting the plant-specific OE for initiating event precursors. specific OE for IE precursors that may have occurred at the PRA model are captured in the plant. The MPS3 PRA model includes a comprehensive PRA Configuration Control list of initiating events that are similar in comparison to Program, NF-AA-PRA-410, along other Dominion Fleet PRA model initiating events. This is with their estimated impact on a documentation issue rather than a modeling issue. model results. This program prioritizes implementation of each issue as: asap (as soon as possible), next model update, or voluntary (i.e., not required for procedure or regulatory compliance). This unmet supporting requirement, which has been assigned a low priority based on its low risk significance, will be addressed in a future model update.

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 13 of 17 Table 3 Status of Identified Documentation Gaps to NEI 00-02 and Capability Category IIof the ASME PRA Standard

..ASME f.Descri.ti6n Current Status /.Comment Schedule for Resolution IE-C10 Perform a reasonableness check of the expansion joint Internal flood contributes only approximately 2% to the All issues potentially affecting the rupture frequencies modeled in the PRA. overall CDF. Any change in the expansion joint rupture PRA model are captured in the frequencies may result in very small impact to the overall PRA Configuration Control CDF. See response to Gap #4 in Table 1 for additional Program, NF-AA-PRA-410, along information on internal flooding modeling and impact. with their estimated impact on model results. This program prioritizes implementation of each issue as: asap (as soon as possible), next model update, or voluntary (i.e., not required for procedure or regulatory compliance). This unmet supporting requirement, which has been assigned a low priority based on its low risk significance, will be addressed in a future model update.

IF-C3b Update the propagation analysis and flood models to Internal Flooding notebook, IF.2, Rev. 3, does not consider All issues potentially affecting the consider the potential for barrier unavailability. Provide the potential for barrier unavailability. For example, PRA model are captured in the a summary discussion of the locations of floor drain "...Flood Compartment CSW-3. The room is equipped with PRA Configuration Control check valves and the considerations for their potential a water-tight door that is assumed to remain intact, and Program, NF-AA-PRA-410, along failure. thus propagationto othercompartments is not postulated." with their estimated impact on In addition, no mention of floor drain check valves are model results. This program included in the Internal Flooding notebooks. See response prioritizes implementation of each to Gap #4 in Table 1 for additional information on internal issue as: asap (as soon as flooding modeling and impact. possible), next model update, or voluntary (i.e., not required for procedure or regulatory compliance). This unmet supporting requirement, which has been assigned a low priority based on its low risk significance, will be addressed in a future model update.

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 14 of 17 Table 3 Status of Identified Documentation Gaps to NEI 00-02 and Capability Category IIof the ASME PRA Standard

.ASME.

Descriptndrti P7on :i Current Status! Comment . Schedule for Resolution IF-C4a Document in more detail that no credible flood It is possible for water to propagate from the Condensate All issues potentially affecting the propagation paths between the plants exist via the Polishing Facility (CPF) to the turbine building. Potential PRA model are captured in the Condensate Polishing Facility. water propagation is bounded by the already analyzed PRA Configuration Control internal flooding events in the turbine building. Specifically, Program, NF-AA-PRA-410, along the amount of water generated during a circulating water with their estimated impact on pipe break is far greater than the amount of water possibly model results. This program generated in the CPF. See response to Gap #4 in Table 1 prioritizes implementation of each for additional information on internal flooding modeling and issue as: asap (as soon as impact. possible), next model update, or voluntary (i.e., not required for procedure or regulatory compliance). This unmet supporting requirement, which has been assigned a low priority based on its low risk significance, will be addressed in a future model update.

IF-D5 Provide clarification in the Internal Flooding PRA In the current PRA model, M308A, the internal flooding All issues potentially affecting the documentation that the initiating event frequencies are initiating events are in "per calendar year'. This is PRA model are captured in the calculated in units of reactor years. Document the basis conservative since the internal flooding frequency has not PRA Configuration Control for the availability factor used to convert initiating event been multiplied by capacity factor. Additionally, internal Program, NF-AA-PRA-410, along frequencies to events per reactor year, or provide a flooding only contributes approximately 2% to the overall with their estimated impact on cross reference. CDF value. model results. This program prioritizes implementation of each issue as: asap (as soon as possible), next model update, or voluntary (i.e., not required for procedure or regulatory compliance). This unmet supporting requirement, which has been assigned a low priority based on its low risk significance, will be addressed in a future model update.

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 15 of 17 Table 3 Status of Identified Documentation Gaps to NEI 00-02 and Capability Category II of the ASME PRA Standard ASMVE~ Description~ Current Status / Comment Schedule for Resolution SR LE-C9b Review the dominant LERF sequences for possible This item is listed as a documentation recommendation as Will be closed with release of credit for containment systems and/or operator actions it is not expected that the review would identify any PRA model M310A.

after containment failure that could potentially reduce modeling enhancements that could adversely impact the LERF. overall LERF value.

QU-D3 The current version Model Quantification Results This documentation issue (i.e., comparison of CDF Will be closed with release of Notebook (QU.2) provides a list of plant features that contributors with plants outside Dominion) will be PRA model M310A.

influence risk, but no comparison of results with similar completed with the updated MPS3 PRA model, M31OA.

plants. The previous version of the notebook included a comparison of the CDF contributors with two other Dominion Westinghouse PWRs, some possible reasons for the differences, and the list of plant features that influence risk. In some cases, only the difference is noted, and not the cause for the difference. Identify and document possible causes for significant differences.

Consider expanding the comparison to plants outside the Dominion fleet.

SY-A4 This documentation gap effects both SY-A4 and SY- The system notebooks have been updated to include All issues potentially affecting the SY-C2 C2. It is suggested a separate reference be provided additional information on plant walkdowns and plant PRA model are captured in the (e.g., walkdown checklists or interview notes) for the personnel interviews. However, the walkdowns and plant PRA Configuration Control items required by this supporting requirement. personnel interviews have not been completed. This is an Program, NF-AA-PRA-410, along on-going process. with their estimated impact on model results. This program prioritizes implementation of each issue as: asap (as soon as possible), next model update, or voluntary (i.e., not required for procedure or regulatory compliance). This unmet supporting requirement, which has been assigned a low priority based on its low risk significance, will be addressed in a future model update.

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 16 of 17 Table 3 Status of Identified Documentation Gaps to NEI 00-02 and Capability Category II of the ASME PRA Standard ASME Description Current Status / Comment Schedule for Resolution SR \

SY-B8 Resolve open items (SY-7, DE-1, DE-2, DE-3) to Three of the four open supporting requirements (SY-7, DE- Will be closed with release of perform spatial dependency analysis within each 1, and DE-3) have been addressed in previous model PRA model M310A.

system analysis. Approve and incorporate the updated revisions. The remaining supporting requirement, DE-2, internal flooding analysis into the model of record. will be closed as documented in Table 2. The internal flooding model was incorporated into the previous PRA model, M305Ac. This supporting requirement remains not met until the documentation issues relating to the spatial dependency are completely documented. MPS3 is highly compartmentalized, which is why a documented spatial dependency analysis would not yield any new insights. The results of the model are not expected to be impacted by these open items.

IE-C3 The IE.2 notebook does not appear to calculate This is a documentation issue since all the initiating events Closed initiating event frequencies on a reactor-year basis. are based on a reactor-year with the exception of internal flooding initiating events (see current/status discussion for IF-D5).

Serial No. 11-696A Docket No. 50-423 Supplemental RAI Response for TSTF-425 Attachment 1, Page 17 of 17 Table 4 Status of Identified Assumptions and Uncertainty Gaps to NEI 00-02 and Capability Category II of the ASME PRA Standard ASME*V Description CurrentStatus I omment Schedule for Resolution AS-C3 Supporting requirements Each of the MPS3 PRA element notebooks (Initiating Events, Accident Sequences, The uncertainty analyses will DA-E3 associated with identification Success Criteria, System Analysis, Data Analysis, Human Reliability, Large Early be resolved with issuance of IE-D3 and documentation of Release, and Quantification) has identified potential sources of model uncertainty. A PRA model M310A.

IF-E6 uncertainty analysis. While characterization of these sources of uncertainty and evaluation of the generic sources IF-F3 assumptions and sources of of uncertainty has not yet been completed for the MPS3 PRA model.

HR-13 uncertainty are documented to LE-F2 some degree in the PRA Though the evaluation of modeling uncertainties for the MPS3 PRA model has not LE-F3 notebooks, a systematic review been completed, the assessment of modeling uncertainties for Surry PRA model has LE-G2 of assumptions and sources of been completed. The Surry modeling Assumptions and Uncertainty (QU.4) notebook LE-G4 uncertainty as defined by the documents the identification and characterization of sources of uncertainty with QU-B1 Standard are not documented. significant potential to influence PRA results and applications. Non-parametric sources QU-E1 of uncertainty include "modeling" uncertainties and "scope and level of detail" QU-E2 uncertainties. Modeling uncertainties must be considered in both the base PRA and in QU-E4 specific risk-informed applications. Uncertainties associated with scope and level of QU-F2 detail should be documented in the PRA, but would be only considered for their impact QU-F4 on a specific application. In the Surry QU.4 notebook, modeling uncertainties are QU-F5 identified and characterized. Information contained in this notebook will be reviewed SC-C3 for any additional sensitivity studies required for this application specific (i.e., RITS-5b SY-C3 evaluation) process.

PRA models within the Dominion Fleet are generally based on the same modeling assumptions and uncertainties. This is true since the models have been maintained and updated by the Dominion PRA staff. Therefore, insights from the Surry QU.4 assessment can provide a technically justifiable basis for uncertainties in the MPS3 PRA model.