ML18270A323

From kanterella
Jump to navigation Jump to search
Attachment 2: Probabilistic Risk Assessment Technical Adequacy
ML18270A323
Person / Time
Site: Palisades Entergy icon.png
Issue date: 09/27/2018
From:
Entergy Nuclear Operations
To:
Office of Nuclear Reactor Regulation
Shared Package
ML18270A320 List:
References
PNP 2018-025
Download: ML18270A323 (58)
v • d • e


Text

ATTACHMENT 2 PNP 2018-025 Palisades Nuclear Plant Probabilistic Risk Assessment Technical Adequacy 57 pages follow

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Palisades Nuclear Power Station PRA Technical Adequacy to Support PNPS Relocation of Technical Specification Surveillance Requirements to an Owner Controlled Program (TSTF 425)

Page 1 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy TABLE OF CONTENTS

1. PURPOSE ....................................................................................................................... 3
2. SCOPE ............................................................................................................................ 3 2.1. Surveillance Frequency Change Process ........................................................................ 4 2.2. Technical Adequacy of a PRA ......................................................................................... 5
3. PNPS PRA TECHNICAL ADEQUACy ................. ........................................................... 6 3.1. Discussion ....................................................................................................................... 6 3.2. PNPS Internal Events and Internal Flooding PRA Model ................................................ 7 3.2.1. Plant Changes Not Vet Incorporated ....................................................................... 7 3.2.2. Peer Review Facts and Observations (F&Os) ......................................................... 7 3.2.3. Consistency with Applicable PRA Standards .......................................................... 8 3.3. PNPS Fire PRA ModeL .................................................................................................... 8 3.3.1. Plant Changes Not Vet Incorporated ....................................................................... 8 3.3.2. Peer Review Facts and Observations ..................................................................... 8 3.3.3. Consistency with Applicable PRA Standards .......................................................... 9 3.4. Identification of Key Assumptions .................................................................................. 10 3.5. External Events Considerations ..................................................................................... 10
4. CONCLUSiONS ............................................................................................................ 11
5. REFERENCES .............................................................................................................. 11 LIST OF TABLES Table 1. Summary of Plant Changes Not Vet Incorporated in the PNPS PRA ......................... 7 Table 2. List of Finding F&Os against the PNPS Internal Events and Flooding Models ......... 12 Table 3. List of SRs Assessed as CC-I in the PNPS Internal Events PRA Model .................. 23 Table 4. List of Finding F&Os Against the PNPS Fire PRA Model. ......................................... 27 Table 5. List of SRs Assessed as CC-I in the PNPS Fire PRA Model .... ................................ 56 Page 2 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy

1. PURPOSE The purpose of this report is to document the technical adequacy of the Palisades Nuclear Power Station (PNPS) Probabilistic Risk Assessment (PRA) model to support the implementation of the Surveillance Frequency Control Program (SFCP), also referred to as Technical Specifications Initiative 5b (Reference 1). PNPS intends to follow the guidance provided in NEI 04-10, Revision 1 (Reference 2), in evaluating proposed surveillance test interval (STI) changes (also referred as "surveillance frequency" changes).
2. SCOPE As explained in NEI 04-10, the Technical Specifications Initiative 5b uses a risk-informed, performance-based approach for establishment of the surveillance frequencies, where PRA methods are used to determine the risk impact of the revised intervals. The PRA technical adequacy is addressed through NRC Regulatory Guide (RG) 1.200, Revision 2 (Reference 3),

which references the ASME/ANS PRA standard, RA-Sa-2009 (Reference 4), for internal events at power. Risk impacts associated with fire, seismic, external events and shutdown activities may be considered quantitatively or qualitatively.

NEI 04-10 guidance includes the five key safety principles described in RG 1.174 (Reference 5),

which are followed as part of this risk-informed Technical Specification Interval change program.

The five key safety principles are:

1. Change meets current regulations unless it is explicitly related to a requested exemption or rule change
2. Change is consistent with defense-in-depth philosophy
3. Maintain sufficient safety margins
4. Proposed increases in core damage frequency (CDF) or risk are small and consistent with the Commission's Safety Goal Policy Statement
5. Use performance-measurement strategies to monitor the change The internal events PRA model Revision 3.3.0 (PSAr3.3.0) is the current model of record for PNPS. The previous revision (PSAr3) of the PNPS internal events PRA is the basis for the PNPS fire PRA model and the PNPS internal flooding PRA model.

The PNPS PRA models and technical content were constructed and documented to meet the ASME/ANS PRA standard (Reference 4). The PNPS fire PRA model was also constructed to meet requirements of NUREG/CR-6850 (Reference 6). The PRA model quantification methodology used at Entergy Operations, Inc, (Entergy) nuclear sites, including Palisades, is recognized within the industry.

Entergy's approach for maintaining, updating and documenting the PRA models at all Entergy nuclear sites is controlled in the fleet procedures. These procedures are consistent with the guidance of the ASME/ANS PRA standard (Reference 4). The procedural process is comprehensive and detailed, which in turn provides the basis for establishing and maintaining the technical adequacy of the models, as well as ensuring the models reflect the as-built, as-operated plant configuration of the sites. In addition, self-assessments and independent peer Page 3 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy reviews are also utilized by Entergy, which reassures the confidence in the approach and overall adequacy of the models against the recognized industry standards and methodologies.

Sections 2.1 and 2.2 describe the general change process and PRA adequacy requirements, respectively, required to support the Initiative 5b. Section 3 documents the technical adequacy of the PNPS PRA model specifically.

2.1. Surveillance Frequency Change Process NEI 04-10 describes the required steps to be followed to adjust a STI. A summary is presented below.

  • Once the STI requiring adjustment is selected, NRC regulatory commitments are collected and reviewed. If any prohibitive commitments are identified, such are examined to determine if the commitment can be changed. If there are no prohibitive commitments, or the commitments may be changed using a commitment change process based on NRC endorsed guidance, then evaluation of the STI revision proceeds. If a regulatory commitment exists and the commitment change process does not permit the change, then the STI revision is not implemented (NEI 04-10, Steps 0-4 (Reference 2)).
  • The PRA technical adequacy is evaluated using guidance from RG 1.200 (Reference 3).

The RG addresses the need to evaluate important assumptions that relate to key modeling uncertainties (such as reactor coolant pump seal models, common cause failure methods, success path determinations, human reliability assumptions, etc.).

Further, the RG addresses the need to evaluate parameter uncertainties and demonstrate that calculated risk metrics (i.e., CDF and large early release frequency (LERF)) represent mean values. The identified "gaps" to Capability Category (CC) II requirements from the endorsed PRA standards in the RG and the identified key sources of uncertainty serve as inputs to identifying appropriate sensitivity cases (NEI 04-10, Step 5 (Reference 2)).

  • Select the revised STI value and revise any changes to the test strategy (NEI 04-10, Step 6 (Reference 2)).
  • Qualitative considerations or qualitative analyses are developed for the STI revision.

Qualitative considerations include surveillance test and performance history, past industry and plant-specific experience, impact on defense-in-depth protection, among other considerations (NEI 04-10, Step 7 (Reference 2)).

  • Perform quantitative and/or qualitative PRA assessments. Steps 8 through 12, and 14, in NEI 04-10 provide details regarding the use of PRA for evaluating the STI. The use of the PRA includes: determining if the structures, systems and components (SSCs) in question are modeled in the PRA, whether the SSCs or operator actions can be modeled (and make changes to the model if possible) or not, performing qualitative assessments as needed, evaluating total and cumulative effect on CDF and LERF, and performing sensitivity studies as needed.
  • The results and proposed STI changes are documented and summarized for consideration by the Integrated Decision-making Panel (IDP). The IDP is usually comprised of the site Maintenance Rule expert panel, a surveillance test coordinator, and a subject matter expert. The IDP approves or rejects the STI changes (with the possibility of adjustments if applicable). If the IDP approves the STI changes, these are documented and implemented. The IDP is also responsible for reviewing the performance monitoring results and providing feedback, if the STI changes, once Page 4 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy implemented, results in unsatisfactory performance (NEI 04-10, Steps 16-20 (Reference 2)).

2.2. Technical Adequacy of a PRA As previously discussed, NEI 04-10 (Reference 2) references the guidance of the NRC Regulatory Guide 1.200 (Reference 3) for the PRA technical adequacy determination. For the purposes of this report, Section 4.2 of RG 1.200, Rev. 2 is used in support of Initiative 5b license amendment request (LAR) submittals. It is important to note that the scope of the Initiative 5b application is broad, and PRA assessments needed for each proposed STI change vary from case to case. The following requirements are noted in Section 4.2 as necessary to demonstrate that the technical adequacy of the PRA is of sufficient quality to support the Initiative 5b LAR submittal:

1. To address the need for the PRA model to represent the as-designed or as-built, as-operated plant,
2. Identification of permanent plant changes (such as design or operational practices) that have an impact on those SSCs modeled in the PRA but have not been incorporated in the baseline PRA model. If a plant change has not been incorporated in the PRA, the licensee provides a justification of why the change does not impact the PRA results used to support the application. This justification should be in the form of a sensitivity study that demonstrates the accident sequences or contributors significant to the application decision were not adversely impacted (remained the same).
3. Documentation that the parts of the PRA required to produce the results used in the decision are performed consistently with the standard as endorsed in the appendices of the RG. If a requirement of the standard (as endorsed in the appendix to the RG) has not been met, the licensee is to provide a justification of why it is acceptable that the requirement has not been met. This justification should be in the form of a sensitivity study that demonstrates the accident sequences or contributors significant to the application were not impacted (remained the same).
4. A summary of the risk assessment methodology used to assess the risk of the application, including how the base PRA model was modified to appropriately model the risk impact of the application and results (note that this is the same as that required in the application-specific regulatory guides).
5. Identification of the key assumptions and approximations relevant to the results used in the decision-making process. Also, include the peer reviewers' assessment of those assumptions. These assessments provide information to the NRC staff in their determination of whether the use of these assumptions and approximations is appropriate for the application, or whether sensitivity studies performed to support the decision are appropriate.
6. A discussion of the resolution of the peer review (or self-assessment, for peer reviews performed using the criteria in NEI 00-02) facts and observations that are applicable to the parts of the PRA required for the application. This discussion should take the following forms:
  • a discussion of how the PRA model has been changed, Page 5 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy

  • a justification in the form of a sensitivity study that demonstrates the accident sequences or contributors significant to the application decision were not adversely impacted (remained the same) by the particular issue.
7. The standards or peer review process documents may recognize different capability categories or grades that are related to level of detail, degree of plant specificity, and degree of realism. The licensee's documentation is to identify the use of the parts of the PRA that conform to capability categories or grades lower than deemed required for the given application (Section 1-3 of ASME/ANS RA-Sa-2009).

This PRA technical adequacy report addresses the quality of the PRA to support relocation of STI frequencies to a licensee-controlled document. There are no STI changes proposed for this Initiative 5b LAR submittal. Items 3 and 4, above, are addressed when preparing an STI change request and are, therefore, not covered in this report. The rest of the items are discussed in Section 3.

3. PNPS PRA TECHNICAL ADEQUACY 3.1. Discussion The PNPS PRA models are controlled in accordance with Entergy procedures consistent with the requirements provided in the RA-Sa-2009 PRA Standard (Reference 4), as previously stated in Section 2. Entergy procedures define the process to be followed to implement scheduled and interim PRA model updates and to control the PRA model files. In addition, the procedure also defines the process for identifying, tracking, and implementing model changes, and for identifying and tracking model improvements or potential issues that may affect the model. Model changes that are identified are tracked via model change requests (MCRs), which are entered in the PNPS MCR database.

Periodic PRA model updates are typically performed at least once every four years, with the option of extending the frequency for up to two years, such that the total update period does not exceed six years. Extensions are justified showing that the PRA model continues to adequately represent the as-built, as-operated plant, and must be approved by management.

The PNPS internal events model PSAr3.3.0 approved in 2017, and the internal fire and internal flooding PRA models approved in 2014 and 2013, respectively, are the models of record. The internal events and internal flooding PRA models of record were used for the 2018 peer review finding closure independent assessment. No changes have been made to the PNPS internal events and internal flood PRA models of record since the completion of the independent closure assessment. A peer review finding closure independent assessment has not been conducted for the internal fire PRA model.

The internal fire PRA will be updated in 2018, and the internal flooding PRA will be updated in 2019. Both are anticipated to be PRA maintenance updates and not involve upgrades or the use of new methods. Therefore, these updates are not expected to impact the TSTF-425 LAR submittal.

Section 3.2.2 discusses the 2018 independent closure assessment performed for the internal events and internal flooding peer review findings. Section 3.3 discusses the independent closure assessment for the fire PRA peer review findings.

Page 6 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy 3.2. PNPS Internal Events and Internal Flooding PRA Model 3.2.1. Plant Changes Not Yet Incorporated As discussed in Section 3.1, an MCR database tracks PRA issues or improvements identified by PRA personnel. The MCR database includes the identification of plant changes that could impact the PRA model.

As part of the PRA evaluation for each STI change request, sensitivity cases are expected to be explored for areas of uncertainty associated with unresolved items (peer review Findings for ASME/ANS PRA Standard CC-II or plant changes) that would impact the results of the STI change evaluation, prior to presenting the results of the risk analysis to the lOP.

There is currently only one plant change not yet implemented which may potentially impact the PNPS PRA. This is listed in Table 1, along with the EC and corresponding MCR number, as well as the potential impact on the PRA model and the SFCP application.

Table 1.

Summary of Plant Changes Not Yet Incorporated in the PNPS PRA MeR, .' .* '*g~P.*f" . ".>'

Nomber " Procedure ," ~~$(;ript,ion9f.,S~'~'tl~~,,;;' ....:> "lInpPrtilnce t6,AppUcatiQn Model EC Re-configuration of the plant instrument The risk impact of this modification is Issue # 63597/64193 air system to use the feedwater purity expected to result in a slight decrease in 630 building air compressors as the primary CDF. The potential impact will be source of air, with the three instrument air addressed by STI change evaluations compressors in standby as back-up. performed in accordance with the SFCP.

3.2.2. Peer Review Facts and Observations (F&Os)

The PNPS internal events and internal flooding PRA models have undergone several peer reviews and self-assessments which document the model quality and identify any areas with potential for improvement. The following assessment for PRA quality has been performed and documented for the PNPS model:

  • In October 2009, an industry peer review of version 3 of the internal events PRA model (PNPS PSAr3), including internal flooding, was performed and documented in a Peer Review Report. This peer review documented eighty (80) new F&Os including fifty-two (52) Findings, twenty-six (26) Suggestions, and two (2) Best Practices. The conclusion of the review was that the PNPS PRA substantially met the ASME PRA standard at CC-II, as endorsed by RG 1.200, Rev. 2, and could be used to support risk-informed applications.

The PNPS internal events model PSAr3.3.0 was approved in 2017 and the internal flood model based on PSAr3 was approved in 2013. These are the current PRA models as stated in Section 2, and address the findings from the 2009 peer review. The 2009 peer review findings and the associated resolutions are documented in a resolution summary report.

The peer review F&Os from 2009 and associated resolutions were reviewed by an independent assessment conducted in May 2018. The closure assessment was conducted in accordance Page 7 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy with Appendix X to NEI 05-04 (Reference 7) utilizing the conditions of acceptance stated in an NRC letter to the Nuclear Energy Institute dated May 3, 2017 (Reference 8). The closure assessment evaluated how the F&Os that were classified as "findings" or "suggestions" from the PRA model full scope peer review were addressed. The closure assessment was performed by a team of eight independent PRA experts. In addition to assessing the closure status, the changes made to the PNPS PRA to address the F&Os were also evaluated to determine whether the changes constituted a "PRA Upgrade" or if new PRA methods were introduced. The independent assessment is documented in a closure report and concluded that none of the changes made to the PNPS PRA constituted an upgrade, while one change implemented in response to a suggestion was considered a new PRA method.

Of the 52 peer review findings and 26 suggestions reviewed during the independent assessment, 38 findings and 16 suggestions were determined by the team to be closed. The 14 peer review findings remaining open are presented in Table 2 along with their dispositionlresolution, and the impact on the SFCP application.

3.2.3. Consistency with Applicable PRA Standards The 2009 peer review assessed the PNPS internal events and internal flooding PRA models to meet the ASME/ANS PRA standard (Reference 4) CC-II of the Supporting Requirements (SRs),

except where noted in Table 3. The F&O independent assessment reviewed the resolution of associated findings to determine if the issues identified in each F&O were addressed to meet the applicable CC-II SRs.

3.3. PNPS Fire PRA Model 3.3.1. Plant Changes Not Yet Incorporated Similar to the internal events model, as part of the fire PRA evaluation for each STI change request, sensitivity cases are expected to be explored for areas of uncertainty associated with open items (peer review Findings for ASME/ANS PRA Standard CC-II or plant changes) that would impact the results of the STI change evaluation, prior to presenting the results of the risk analysis to the lOP. As noted in Section 3.2.1, there is currently only one plant change not yet implemented with may potentially impact the PNPS PRA. As stated above, this item is expected to be reviewed and assessed during the specific STI request.

3.3.2. Peer Review Facts and Observations The PNPS fire PRA model has undergone several peer reviews, including a full scope and two in-process peer reviews. These reviews document the model quality and identify any areas with potential for improvement. The following assessment has been performed and documented for the PNPS fire PRA model:

  • The PNPS fire PRA peer review was conducted in March 2011, following two in-process peer reviews held in January 2010 and August 2010. The full-scope peer review produced a total of seventy-six (76) F&Os including sixty (60) Findings, fifteen (15)

Suggestions, and one (1) Best Practice.

The current PNPS fire PRA model was approved in 2014 and is based on internal events model PSAr3 that was approved in 2012. This is the current fire PRA model as stated in Section 2, and addresses the findings from the 2011 peer review. The 2011 fire PRA peer review findings and the associated resolutions are documented in a resolution summary report.

Page 8 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy The internal fire PRA will be updated in 2018. This is anticipated to be PRA maintenance update and not involve upgrades or the use of new methods. Therefore, this update is not expected to impact the TSTF-425 LAR submittal.

3.3.3. Consistency with Applicable PRA Standards As discussed in Section 3.1, the PNPS Fire PRA model was updated in 2012. Per Entergy procedures, all Entergy PRA models are required to meet current industry standards for PRA model development and documentation. Specifically, the Entergy PRA guidelines were developed to attempt to meet the ASME/ANS PRA standard (Reference 4) CC-II of all SRs.

NUREG/CR-6850 guidance was the primary methodology used for the development of the fire PRA. The updated fire PRA in some cases used methodologies that extend beyond the guidance of NUREG/CR-6850. These methods used in the PNPS Fire PRA are considered extensions of the NUREG/CR-6850 methods and are documented via reference to approved NEI 04-02 frequently asked questions (FAQs) or other NUREGs. These references are:

  • NUREG/CR-7150, Vol 2, "Joint Assessment of Cable Damage and Quantification of Effects from Fire." (JACQUE-FIRE)
  • NUREG-1921, Rev. 0, "EPRI/NRC-RES Fire Human Reliability Analysis Guidelines-Final Report."
  • FAQ 14-0009, Rev. 1, "Treatment of Well-Sealed MCC Electrical Panels Greater than 440V."

The full-scope peer review for PNPS fire PRA model was conducted in March 2011 using RG 1.200, Revision 2. The NRC's review and acceptance of Palisades internal fire PRA is documented in the NRC safety evaluation included in Palisades License Amendment dated February 27,2015 titled, "Palisades Nuclear Plant -Issuance of Amendment Regarding Transition to a Risk-Informed, Performance-Based Fire Protection Program In Accordance with 10 CFR 50.48(c) (TAC No. MF0382), Adams Accession Number ML15007A191. Since then, a model revision was completed which addressed the findings from the peer review. Table 4 provides a listing of the open finding-level F&Os related to the fire PRA and the acceptability of the finding-level F&Os in relation to this application. Of the 60 open findings, all but 13 have been resolved in the current fire PRA model. Table 5 lists SRs associated with the fire PRA which were not reviewed as the model element was not sufficiently complete or were assessed as CC-I only. Table 5 provides the disposition of CC-I acceptability for this application.

As part of the fire PRA evaluation for each STI change request, sensitivity cases would be expected to be explored for areas of uncertainty associated with open items (peer review Findings for ASME/ANS PRA Standard CC-II or plant changes) that would impact the results of the STI change evaluation, prior to presenting the results of the risk analysis to the lOP. At present, there are open items associated with the F&Os in Table 4 and the CC-I or unreviewed SRs in Table 5. Of the 13 unresolved findings in Table 4, seven pertain to detailed human reliability analysis (HRA) and human failure event (HRE) dependency analysis, two are related to cable selection, two to fire scenario selection, two to uncertainty analysis, and one to Page 9 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy seismic-fire interactions. Of the seven CC-I or unreviewed SRs in Table 5, four pertain to the plant response model and three pertain to the HRA. Sensitivity cases may need to be explored to assess the impact of using human error probabilities from the detailed HRA, additional cable selection criteria, incorporation of main control room abandonment scenarios, consideration of new initiating events or accident progressions, or undesired operator actions.

3.4. Identification of Key Assumptions The Initiative 5b is a risk-informed process which uses PRA model results to support a proposed STI change. The lOP uses the PRA results as an input to decide whether an STI change is warranted. The methodology recognizes that a key area of uncertainty for this application is the standby failure rate utilized in the determination of the STI extension impact. Therefore, the methodology requires the performance of selected sensitivity studies on the standby failure rate of the component(s) of interest for the STI assessment.

Any additional sensitivity studies identified for specific STI changes are also required per NEI 04-10, Revision 1. Therefore, results of the standby failure rate sensitivity study plus the results of any additional sensitivity studies identified during the performance of the reviews of gaps and open items as summarized in Sections 3.2 and 3.3 herein, will be documented and included in the results of the risk analysis submitted to the lOP.

3.5. External Events Considerations The NEI 04-10 methodology allows for STI change evaluations to be performed in the absence of quantifiable PRA models for all external hazards and shutdown. For those cases where the STI cannot be modeled in the plant PRA, or where a particular PRA model does not exist for a given hazard group, a qualitative or bounding analysis is performed to provide justification for the acceptability of the proposed test interval change.

External hazards were evaluated in the PNPS Individual Plant Examination of External Events (IPEEE) submittal in response to the NRC IPEEE Program (Reference 9). The IPEEE Program was a one-time review of external hazard risk and was limited in its purpose to the identification of potential plant vulnerabilities and the understanding of associated severe accident risks.

PNPS does not have a PRA model or applications associated with external hazards such as seismic, high wind, or external flooding, and quantitative results cannot be provided to support this STI effort. Therefore, a qualitative or bounding approach will be used to assess external event hazard risk at PNPS for STI changes.

Because PNPS does not have external hazards or shutdown PRA models, external hazards and shutdown screening evaluations are expected to be performed for STI changes in accordance with the guidance of NEI 04-10, Revision 1. When performing STI extension evaluations PNPS will assess the risk from external events hazards (seismic, winds and tornadoes, external flooding) by applying the screening evaluations generated in response to the IPEEE in accordance with the NEI-04-1 0 guidance. While it is recognized that the IPEEE assessments have remained static since they were completed, they do form the basis for an initial understanding of inSights from external hazards. These base insights would then be assessed to account for any updated information or attributes that may have changed since the IPEEE to better reflect the as-built, as-operated plant. Acceptability for the proposed surveillance requirement frequency change for that particular external hazard would then be determined and factored into the overall acceptability of the proposed change.

Page 10 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy The PNPS shutdown safety program developed to support implementation of NUMARC 91-06 (Reference 10) is used for the shutdown risk evaluation, or an application-specific shutdown analysis may be performed for STI changes in accordance with the guidance of NEI 04-10, Revision 1. The PNPS shutdown safety program includes input from a Defense-in-Depth shutdown Equipment Out Of Service (EOOS) PRA model.

4. CONCLUSIONS The information presented herein demonstrate that the PNPS PRA technical adequacy and capability evaluations, as well as the maintenance and update processes conform to the ASME/ANS PRA Standard, which satisfies the guidance of RG 1.200, Revision 2. Therefore, the PNPS PRA support NEI 04-10 SFCP implementation at Palisades.
5. REFERENCES
1. TSTF-425, "Technical Specification Task Force - Relocate Surveillance Frequencies to Licensee Control- RITSTF Initiative 5b", Revision 3, March 2009 (ADAMS Accession Numbers are ML090850627, ML090850630, ML090850638).
2. NEI 04-10, "Risk-Informed Technical Specifications Initiative 5b. Risk-Informed Method for Control of Surveillance Frequencies", Revision 1, April 2007 (ADAMS Accession Number is ML071360456).
3. Regulatory Guide 1.. 200, "An Approach for Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk-Informed Activities", Revision 2, March 2009.
4. ASME RA-Sa-2009, "Standard for Level1/Large Early Release Frequency Probabilistic Risk Assessment for Nuclear Power Plant Applications", February 2009.
5. RG 1.174, "An Approach for Using Probabilistic Risk Assessment in Risk-Informed Decisions on Plant-Specific Changes to the Licensing Basis," Revision 2, May 2011.
6. NUREG/CR-6850 - EPRI-1011089, "Fire PRA Methodology for Nuclear Power Facilities",

August 2005.

7. NEI 05-04/07-12112-06 Appendix X, "Closeout of F&Os", March 2017 (NRC ADAMS Accession No. ML16158A035).
8. Letter from J. Giitter (NRC) & M. Ross-Lee (NRC) to G. Krueger (NEI), "U.S. Nuclear Regulatory Commission Acceptance on Nuclear Energy Institute Appendix X to Guidance 05-04,07-12, AND 12-13, Close-Out of Facts and Observations (F&Os)", May 3,2017 (NRC ADAMS Accession No. ML17079A427).
9. Generic Letter 88-20, "Individual Plant Examination of External Events (IPEEE) for Severe Accident Vulnerabilities - 10CFR 50.54(f), Supplement 4", June 1991.
10. NUMARC 91-06, "Guidelines for Industry Actions to Assess Shutdown Management,"

December 1991 (ADAMS Accession Number ML14365A203).

Page 11 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 2.

Open Finding F&Os against the PNPS Internal Events and Internal Flooding Models Justification for the exclusion of Added additional justification for the exclusion of data data before January 2003 used prior to January 2003 to the initiating events notebook documentation is to identify plant-specific initiating NB-PSA-IE. The start date of January 2003 was needed to close this events was not provided. maintained for this update to cover a broad base of plant finding. Justification of experience, but not too far in the past so as to avoid data collection periods Justification for the exclusion of plant experience that is not relevant to recent is not expected to data before January 2003 used maintenance and operating practices. impact STI change to identify plant-specific initiating evaluations performed events was not provided. Independent Assessment: in accordance with the Finding remains OPEN. Other reasons for trimming SFCP.

Provide the requested experience prior to 2003 need to be included. For justification. example, implementation of a regulatory rule, such as the SBO rule, or a plant change or modification that changed the plant performance. Optionally, a data trend of number of trips should be used that shows a measurable chanae in plant events.

In relation to IE-C6, Operator The basis for excluding control room HVAC from the full Additional (Partially actions are apparently credited power internal events model was strengthened to documentation is Resolved) for the exclusion of some events include other aspects in addition to operator actions and needed to close this (e.g., CRHVAC refer to earlier documented in NB-PSA-ETSC. The discussion of the finding. Justification of HVAC comments) without control room heat-up rate effects on the reactor initiating event justifying each such credit protective system (RPS) components concluded that a screening is not (operator training, procedures, loss of HVAC would not result in a significant increase in expected to impact STI etc.) the failure probability of the RPS. change evaluations performed in If component/system failures In addition, a comparison of sensitivity analyses accordance with the lead to an initiating event but are performed for 14 owner's group sites that modeled the SFCP.

screened from further analysis by contribution to CDF due to loss of control room HVAC.

crediting operator actions or The sensitivity studies found that the average CDF/yr equipment/systems to avert the was 1.61 E-07 with a median of 1.31 E-07/yr. Given transient, then quantify the total Palisades core damage frequency is on the order of E-initiating event frequency 05, the change in CDF due to loss of control room considering these events and HVAC would less than 1%.

apply criteria of IE-C6 to determine if screening criteria is For cable spreading room cooling, an analysis of the met. cable spreading room heat-up following a loss of Apply IE-C6 screening criteria ventilation was developed using the GOTHIC software and document as <>nnrnnri<>t., code. This analysis developed a conservative room Page 12 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 2.

Open Finding F&Os against the PNPS Internal Events and Internal Flooding Models Status . . ,. *. Applica~le*.'. Finding/Observation .'... . ...........

F&O Disposition II11 P0r'ta"c&to

.....,. '... ; .... I***.**,***** ........* SRs ...  ;. .... ;.... ,...... <. .....

Application*

heat-up profile based on actual test data and assuming operators take no action to either open doors or affix portable ventilation. Based on the evaluation of equipment qualification reports, and vendor data, it was concluded there is reasonable assurance of operability for all equipment in the room under these conditions.

The conclusions of these analyses demonstrate ventilation to the cable spreading and control room areas is not necessary to be explicitly modeled and the bases for these conclusions do not require operator action to mitigate elevated temperatures.

Independent Assessment:

Finding remains OPEN. NB-PSA-ETSC discussion does not explicitly address criterion "C" of IE-C6. The Loss of HVAC IE frequency has not been quantified for Palisades. However, a loss of HVAC does not result in an immediate reactor shutdown (see IE-C6, criterion c).

The owners' group CDF information does not explicitly address the requirements of the SR because support system requirements and component response to high temperatures are plant specific.

While CR HVAC is cited as an example in the finding, no other initiators are addressed in the response. The text needs a discussion of the IE-C6 criteria for all screened initiators, or a statement that no other initiators were screened via these criteria.

SY-B3-01 Open SY-B3 Common cause failures as a The modeling of common cause failures, as applied in Refined treatment of (Not whole are modeled correctly and the Palisades PRA, is based on, and consistent with, the certain common cause Resolved) consistently. However, the Multiple Greek Letter approach. This approach produces failures is needed to modeling of the HPI, LPI, and valid cut sets. close this finding, which common line check valves is may have a small producing non-minimal and A review of the updated model PSAR3 results, following impact on the PRA potentially nonvalid cutsets. update of common cause logic and data, do not show results. For those STls any cutsets containing random failure of a single check on which this finding is Because of the safety valve in the same cutset as a check valve failed due to determined to have a Page 13 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 2.

Open Finding F&Os against the PNPS Internal Events and Internal Flooding Models 1.StattJs iAp. Pl. *. i*.*. c. . . able SRs** I. * .

..*... *.* .**. *. FindingfObservation Disposition .II1'lP0r-ttln~eto

~ieati()n' significance of the LPI and HPI common cause for the HPI and LPI systems. potential impact, the systems, the non-minimal and effect is expected to be non-valid cutsets are Therefore, the concerns expressed by this finding do not assessed in the change overestimating the risk appear to be correct, and modeling or quantification evaluations for the associated with those failures. changes are not considered necessary. affected STls.

Review the common cause Independent Assessment:

modeling of components in the Finding remains OPEN. EA-PSA-RG1.200F&O-10-01 PRA model, especially of the describes two lines with two check valves each and a valves in series and revise the common cause group of 4. It further states that CCFs of model as appropriate. the two valves in series is a valid cutset, even though it Alternatively, non-valid is non-minimal to the independent fault of a single valve combinations can be added to in the line. This method of modeling redundant in-series the mutually exclusive file to pairs is non-conservative for two reasons. First, the remove the non-minimal and inner valves and the outer valves see different service, non-valid cutsets. one at high DP and the other at low DP. Second, the use of the single 4 group dilutes the failure probability by adding non-minimal pairs that use a fraction of the CC factor without actually failing the redundant trains. The standard convention for this configuration is to model two CC groups of 2.

SY-85-02 Open SY-85 Potentially risk-significant manual In the most recent NUREG/CR-692B parameter Additional (Partially valves were excluded from the estimation update (2010) the probability of manual valve documentation is Resolved) model without explanation. Their spurious operation is B.42E-OB/hr. It can be presumed needed to close this exclusion should be based on SR that locked valves would likely have a spurious finding. Justification for SY-A15 screening criteria. For operation failure probability that is significantly lower screening of manual example, manual valves in the than BE-B. valve failure modes is Containment Spray system flow not expected to impact paths were not modeled. The potential impact of failure to remain in position of STI change evaluations locked open I closed manual valves could result in loss performed in It was noted that some of these of pump suction, pump deadheading, or fluid flow accordance with the manual valves are actually diversion. In the Palisades risk significant stand-by SFCP.

depicted on the simplified system systems, all of these effects are equivalent to pump fail drawings, but they are not to start. The generic pump fails to start failure rates from labeled. To avoid confusion, it is NUREG/CR- 692B have a minimum failure probability of suggested that all components in on the order of 1E-3. Therefore, the criterion described these drawings be labeled. Note: in SY-A15 for excluding manual valves is met.

site practice is to include all Page 14 of 57

ATTACHMENT 2 PNP 2018*025 . Probabilistic Risk Assessment Technical Adequacy Table 2.

Open Finding F&Os against the PNPS Internal Events and Internal Flooding Models mechanical components on the Independent Assessment:

simplified PRA schematics and to Finding remains OPEN. For the containment spray label only those components header manual valves in question, these valves are not specifically included. This easily tested. If they have not been functionally tested provides a quick indication of for the life of the plant, e.g., 40 years and the failure rate what components are physically is 8.4E-8 per hour, the standby failure probability is present but not explicitly 1.5E-2. This number is likely not less than 1% of the modeled. highest failure probability. The intent of the F&O is that manual valves be evaluated and screened on a system Excluded manual valves may be level showing an evaluation of their potential impact on risk significant. the systems. This issue should be addressed at the system level for all systems, not just Containment Spray.

Provide explanation for the excluded valves based on SY-A 15 or include them in model.

SY-B12-01 Open SY-B12 Palisades did not model HVAC The basis for excluding control room HVAC from the full Additional (Partially for the control room or the cable power internal events model was strengthened to documentation is Resolved) spreading room based on include other aspects in addition to operator actions. needed to close this operator actions to implement This evaluation was fully documented in NB-PSA-ETSC. finding. Justification of alternate cooling strategies such The conclusion states that control room cooling in the initiating event as opening doors or using a Palisades internal events PRA is not considered an screening is not proposed portable exhaust fans. issue based on high design temperature limits of the expected to impact STI (See pages 17 and 24 of major control room components, conservative modeling change evaluations attachment 8 to NB-PSA-ETSC assumptions, operator philosophy with respect to performed in r01). However, the operator remaining in the control room during such an event, and accordance with the actions to implement the the relative un-importance of HVAC failure in a variety of SFCP.

alternate actions were not plant PRA studies. Therefore, it is considered included in the models. There unnecessary to model either loss of HVAC as an initiator was never an intent to model the or as a support system for the internal events model.

operator actions given that past analyses have shown that both For cable spreading room cooling, an analysis of the rooms can survive a loss of cable spreading room heat-up following a loss of HVAC. It is recognized that the ventilation was developed. Based on the conclusions of analyses require updating and the analysis, ventilation to the cable spreading area is that the documentation requires not explicitly modeled as failure to re-establish updating. ventilation does not result in equipment failure prior to the PRA 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> mission time. NB-PSA-ETSC has been Palisades should either to reflect these conclusions.

Page 15 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 2.

Open Finding F&Os against the PNPS Internal Events and Internal Flooding Models additional justification for not Independent Assessment:

modeling the HVAC systems for Finding remains OPEN. NB-PSA-ETSC addresses the cable spreading room and control room HVAC. It includes a sensitivity study to control room, or model the evaluate cases with and without the setup of portable operator actions to implement emergency ventilation. Requirement SY-B12 explicitly alternate cooling strategies or prohibits the screening of systems on the basis of model HVAC for these two recovery actions:

rooms. ".. .it is not acceptable to not model a system such as HVAC or CCW on the basis that there are procedures for dealing with losses of these systems."

The Palisades model omits control room HVAC, and specifically credits operator recovery actions as the basis. However, this SR is very specific and is directly applicable to this finding. Several improvements need to be made for the MCR HVAC screening analysis. If this cannot be accomplished, then the HVAC needs to be modeled and, if proceduralized, recovery actions can be credited.

Although an ultimate failure temperature is not indicated in the analysis, sufficient evidence has been provided for not modelina room coolina to the CSR.

-B12-02 Open SY-B12 Detailed analysis of See disposition for F&O SY-B12-01 regarding control Additional (Partia"y systems/component Dependency room and cable spreading room HVAC. documentation is Resolved) on HVAC/ventilation should be needed to close this provided in the individual See disposition for F&O SY-BS-01 regarding SWS pump finding. Justification for systems and/or Dependency failure to start. not modeling Tables. Palisades needs to dependencies is not provide better documentation of Independent Assessment: expected to impact STI the basis for not modeling the Finding remains OPEN. The F&O has not been change evaluations HVAC within the system addressed, as it recommends that a systematic review performed in notebooks for the control room of dependencies be performed, while the resolution accordance with the and cable spreading room. focused on HVAC and SWS. The finding requires that SFCP.

individual system notebooks discuss room cooling SWS pump failures to start are requirements. Also see disposition of SY-B5-01 and valid contributors to EDG failure. SY-B12-01.

A review of the affected cutsets Page 16 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 2.

Open Finding F&Os against the PNPS Internal Events and Internal Flooding Models iif!;' *f".I::~~~~IJ~'";;#%1~~~~J~:~'~01.~~I~~~~9'Ob!rerVation ".....

still has a cutset with a diesel generator run failure in the same cutset as the SWS pump failure to start. Given failure of the SWS pump to start, the diesel generator fails to run should be 1.0. The model needs to account for these and similar dependencies.

These specific failures should be incorporated into the fault tree model. And, given the similarity of this finding with Finding SY-85-02, it is recommended that a systematic review of other potentially risk important dependencies be performed.

HR-G7-01 Open HR-G7 Palisades has not completed As part of the latest FPIE update, a human error Several assumptions (Partially their HFE Dependency dependency analysis was completed as documented in seem to be in conflict Resolved) Evaluation for their updated EA-PSA-FPIE-16-03. The methodology for evaluating with NUREG-1842 and HRA. This is specifically noted in human error dependency was developed as described common HRA Section 5.2 of PLP-HRA. in HRA Notebook N8-PSA-HR. This approach evaluates practices; further the dependency between the multiple operator actions justification for the Failure to meet explicit that occur in the accident sequences of the Palisades conflicts is needed. For requirement of the standard. PSA. those STls on which this finding is After the HRA is complete, redo To address the human action dependency issue with determined to have a and document the dependency respect to CDF, Palisades developed a systematic potential impact, the evaluation. approach that investigates a sufficient number of human effect is expected to be actions to merit confidence that the impact of these assessed in the change dependencies have been thoroughly assessed and evaluations for the adequately represented in the PSA models. The affected STls.

approach is iterative and methodical.

Independent Assessment:

Finding remains OPEN. The HRA dependence is erformed usino the HRA Calculator tool and the latest Page 17 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 2.

Open Finding F&Os against the PNPS Internal Events and Internal Flooding Models While the process for HRA combinations was described in the HRA document at the time of the peer review, no HRA combinations were available to indicate how the method was applied and how adjustments in the HRA combinations were made based on the cutsets where the combination occurred. Based on the Fire PRA Peer Review report, the Review Team considered that the method had been reviewed.

IFSO-A4-01 Open IFSO-A4 Palisades did not explicitly In the latest update, human induced flood events were Additional (Partially identify and characterize human characterized for each flood area initiating event as part documentation is Resolved) induced flooding events for each of the maintenance induced flooding frequency needed to close this flood area. Instead, Palisades development. Maintenance induced flood frequency in finding.

chose to characterize the human- each flood area is system specific to characterize the Characterization of induced flooding events by flood mechanism. human-induced setting a generic element and flooding mechanisms is then back-calculating a The approach to this is described in reports EA-PSA- not expected to impact frequency without actually FLOOD-IE-13-02 and EA-PSA-INTFLOOD-13-06. STI change evaluations delineating what the human performed in induced event was. Independent Assessment: accordance with the Finding remains OPEN. While human-induced flooding SFCP.

Without a reasonable frequencies are now quantified, the events are still not characterization of the specific characterized per the requirements of SR IFSO-A4. The human induced flooding events it SR requires the identification of flooding mechanisms is difficult to understand their full that would result in the release of water or steam and to impact on the results or address include human-induced mechanisms that could lead to them should they be found to be overfilling tanks or diverting flow through openings significant contributors. created during maintenance activities.

Palisades should either more fully characterize the human induced flooding events or they should be explicitly called out as assumptions so that they can be assessed for applications

",ff",,..tinn internal f1nnrlinn Page 18 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 2.

Open Finding F&Os against the PNPS Internal Events and Internal Flooding Models Those automatic or operator Palisades has developed a flood mitigation abnormal responses that have the ability to operating procedure AOP- 39 which defines operator documentation is terminate or contain the flood actions for flood mitigation in all 11 PRA defined flood needed to close this propagation for each defined areas. Detailed human error probabilities have been finding. Discussion of flood area and flood source were developed and incorporated into the model for risk automatic and operator not identified. significant actions based on this procedure. responses to contain flood propagation is not Required by SR. Independent Assessment: expected to impact STI Finding remains OPEN. Although human error change evaluations Identify and document the probability values were developed for floods in the 1D performed in automatic and operator switchgear room, no other automatic plant or system accordance with the responses that do have the responses or operator actions for other rooms were SFCP.

ability to terminate or contain the detailed. Because some action, plant, or system flood propagation for each response would be needed to respond to nearly all defined flood area and source. flooding events and none have been identified outside the 1D switchaear room. this findina is considered IFQU-A9-01 Open IFQU-A9 A specific discussion of jet walkdown documentation and Additional (Partially impingement and pipe whips was was added to EA-PSA-INTFLOOD-13-06 to documentation is Resolved) not identified. demonstrate additional walkdowns were performed both needed to close this before and after the 2008 walkdown in which the escort finding. Discussion of Consideration of jet impingement had limited time. other flood failure and pipe whips (as appropriate) mechanisms (humidity, are a requirement of the standard This includes how pipe whip and jet impingement were condensation, for this element. evaluated during the lSI walkdowns; as the RI-ISI report temperature, etc.) is not provides the basis for the indirect effects on equipment expected to impact STI Provide a discussion of how jet for each flood or spray initiator. Detailed walkdowns of change evaluations impingements and pipe whips all flood areas were documented in the RI-ISI indirect performed in were considered and handled. effects report. accordance with the The Internal Flooding Analysis SFCP.

Report referenced walkdowns Independent Assessment:

performed for the IPE. The scope Finding remains OPEN. Regulatory Guide 1.200 of these walkdown was limited as requires for CC 11111111 that humidity, condensation, a result of time constraints temperature concerns, and any other identified failure placed on the walkdown team by modes also be addressed. Temperature effects would the authorized team escort. be expected to be fire system actuation as a result of a Palisades indicated that they had HELB. Additionally, high humidity or elevated performed a more recent temperature effects on equipment reliability could be a

,.."mnl"t" walkdown. but that concern. No evidence was seen that any of the reauired Page 19 of 57

ATTACHMENT 2 PNP 2018*025* Probabilistic Risk Assessment Technical Adequacy Table 2.

Open Finding F&Os against the PNPS Internal Events and Internal Flooding Models walkdown was not referenced in the Internal Flooding Analysis Report. The consideration of jet impingement and pipe whip is qualitatively and semi-quantitatively discussed in the walkdown notes for the more recent walkdown. If Palisades wants to credit the more recent walkdown, they need to reference it in the Internal Flood U-A3-01 Open I QU-A3 The mean ISLOCA The updated Palisades PRA employs a fault tree (Partially frequency does not account for to establish ISLOCA IE frequency. Component failure the treatment of the Resolved) the state-of-knowledge basic events that represent parallel path identical SOKC is needed to correlation (SOKC). Per SR QU- components utilize the same uncertainty correlation close this finding, which A3, the effect of the SOKC has class when model uncertainty is quantified. Therefore, may have a small been found to be significant in the ISLOCA IE fault tree accounts for the SOKC when impact on the PRA cutsets contributing to ISLOCA simulations are performed using random sampling results. For those STls frequency. methods to evaluate uncertainty. on which this finding is determined to have a Explicitly required in Note 1 of In addition, two examples were selected for evaluation potential impact, the the SR. using a Monte Carlo simulation using as input the failure effect is expected to be rate and distributions from the PSAR3 model. Based on assessed in the change Update the ISLOCA frequencies these simulations, a correction factor was applied as a evaluations for the with SOKC. recovery event to the ISLOCA cut sets generated by affected STls.

CAFTA. The result is an increase in the initiating event frequency by a factor of 3.

This increase in initiating event frequency is not a significant deviation and results in a negligible impact on overall core damage frequency. Therefore, the event tree rules and basic events developed to account for the SOKC will only be incorporated into the model for specific applications that examine ISLOCA events.

Independent Assessment:

Findina remains OPEN. The ISLOCA IE freauencv has Page 20 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 2.

Open Finding F&Os against the PNPS Internal Events and Internal Flooding Models been estimated in EA-PSA-FPIE-16-03 as 6.152E-10/yr.

An estimate of the ISLOCA IE frequency accounting for the SOKC effect in EA-PSA-RG1.200F&O-1 0-01 is 6.13E-9/yr. This suggests that the effect of the SOKC correlation is a factor of nearly 10. Such a large increase in the ISLOCA IE frequency seems to be significant for this event. Since the SOKC effect was investigated and a decision was made to not include the effects of the SOKC in the base model, this finding has not been addressed to be consistent with QU-A3 CC-II.

QU-C1-01 Open QU-C1 Conditional HEPs were The complete detailed methodology for evaluating As described above for (Partially developed by Palisades for human error dependency is described in HRA Notebook HR-G7-01, several Resolved) several HFEs and incorporated in NB-PSA-HR. assumptions seem to the fault tree models. Some be in conflict with accident sequences revealed To address the human action dependency issue with NUREG-1842 and HFE combinations for which respect to COF, Palisades developed a systematic common HRA dependency between the HFEs approach that investigated a sufficient number of human practices. For those has not been assessed and actions to merit confidence that the impact of these STls on which this documented. dependencies have been thoroughly assessed and finding is determined to adequately represented in the PSA models. The have a potential impact, While the Palisades model has approach is iterative and methodical. the effect is expected to been quantified and cut sets for be assessed in the accident sequences have been Independent Assessment: change evaluations for identified, the review and update Finding remains OPEN. Closure of this Finding should the affected STls.

of those sequences with respect be linked to Closure of Finding HR-G7-01. Based on the to combinations of HFEs is not Fire PRA Peer Review report, the Review Team complete. considered that the method had been reviewed ..

However, the implementation of the method did not Complete review and update of meet the standard requirements for SR HR-G7, so that accident sequence cut sets Finding remains open, and, as a result, this Finding also to combinations of HFEs. remains OPEN ..

01 Open QU-01 model review has not The documentation of the final model of record Additional (Partially been completed and (PSAR3), including final reviews, is complete and documentation is Resolved) documented. documented in EA-PSA-FPIE-16-03. needed to close this finding. Comparison of The final review of accident Independent Assessment: PNPS results to other sequence results has not been Finding remains OPEN. QU-04 is met at CC-I because similar plants is not comoleted and documented so no discussion of how Palisades' results comoare with STI Page 21 of 57

ATTACHMENT 2 PNP 2018*025* Probabilistic Risk Assessment Technical Adequacy Table 2.

Open Finding F&Os against the PNPS Internal Events and Internal Flooding Models the reasonableness change evaluations results can be verified. Palisades performed in indicated that this review is accordance with the required but not complete. This SFCP.

finding is being written against all of the QU-O supporting requirements as well as some QU-F requirements.

Palisades needs to complete the formal review of accident sequence quantification results and make modifications as needed to address issues found in that review. The final results should then be documented in the correspondina notebooks.

LE-G5-01 Open I LE-G5 The Palisades PSA Level 2 Given the Palisades two source term models, PAL-L2 Additional (Partially Notebook does not explicitly and PWROG-L2, it is considered that sufficient detail documentation is Resolved) discuss any limitations in the exists such that this requirement is met. needed to close this LERF analysis that might impact finding. Discussion of applications. However, additional documentation, as presented in LERF model limitations Appendix C of this analysis, presents the guidance that is not expected to It is expected that the limitations has been added to the next pending issuance of the impact STI change will be similar to those discussed Level 2 notebook. evaluations performed for the level 1 analyses, but the in accordance with the level 1 discussion does not Independent Assessment: SFCP.

explicitly cover LERF so their Finding remains OPEN. While the appendix provides analysis does not comply with information about the MAAP models and SAMA the SR. evaluation, it does not discuss limitations of the LE model that could impact applications. Such a discussion Palisades should develop such a should identify any assumptions or uncertain discussion similar to that phenomenological parameters that could limit the developed for the level 1 model's insights regarding applications.

analyses or revise the level 1 discussion to include LERF.

Page 22 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 3.

List of SRs Assessed as CC-I in the PNPS Internal Events PRA Model

~'m~~~~,~~(~g};~~~M~~I,~~l~J:,'

Finding IE-A9-01 A documented review of all maintenance rule and work order This issue was resolved such No documentation provided for review for failures was added to the initiating events notebook NB-PSA-IE that it conforms to SR CC-II precursors. Only mentioned in Special initiator to determine if they are potential precursor events. Component and therefore is not expected discussion as "Special initiating events or the failures were obtained from the review of failures documented in to impact STI change potential for such events (e.g., precursors) the data notebook NB-PSA-DA and individually evaluated as to evaluations performed in were considered during the PRA teams' their potential as a precursor event. No new initiating events accordance with the SFCP.

review of the MR database and Maintenance were developed as a result of the evaluation. However, the Work Orders (MWO) in support of the data exercise did confirm several existing transient initiator events effort." were appropriately modeled in the PRA.

Independent Assessment:

Finding determined to be CLOSED. This resolution meets the IE-A9 CC-II requirement. The resolution does not involve a new PRA method or a PRA uoarade.

AS-A10 Finding AS-A 10-01 NB-PSA-HR, Vol. 1, Rev. 4, Appendix I provides discussions for This issue was resolved such Event trees are used to model the a number of HEPs that are used in multiple Event Trees. HEPs that it conforms to SR CC-II progression of each accident sequence, were assigned to the Operations Department Operating Crews and therefore is not expected including the applicable success criteria for for review. Their reviews included ensuring indications, to impact STI change each node of the event trees. Although the procedure selection and use, and activity performance man- evaluations performed in event trees include operator actions required power and timing are correct. Training personnel reviews accordance with the SFCP.

for success of key safety functions, the included ensuring procedure selection and use were consistent documented actions do not include with current training expectations, and the training type and verification that the operator actions, as frequency are accurate.

evaluated, are "bounding" for all event tree nodes where the operator action is applied. Independent Assessment:

The CC-II requirement to capture and provide Finding determined to be CLOSED. This resolution meets the sufficient detail for significant differences in AS-A 10 CC-II requirement. The resolution does not involve a requirements associated with systems and/or new PRA method or a PRA upgrade.

ooerator resoonses is not oerformed.

HR-C2 Finding HR-C2-01 A review of plant history was conducted for plant specific This issue was resolved such No evidence that an operating experience operating experience. The result of the review was that while that it conforms to SR CC-II review [of plant-specific events] was there were instances noted of conditions that would be and therefore is not expected performed [for pre-initiator HFEs]. considered pre-initiators, the examples noted were either already to impact STI change covered by a pre-initiator event identified during the evaluations performed in implementation of the revised methodology or were related to accordance with the SFCP.

equipment not credited in the PRA. The plant operating experience review is documented in HRA notebook.

Indeoendent Assessment:

Page 23 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 3.

List of SRs Assessed as CC-I in the PNPS Internal Events PRA Model

!;,status)* .

Finding determined to be CLOSED. This resolution meets the HR-C2 CC-III11I requirement. The resolution does not involve a new PRA method or a PRA uoorade.

DA-C7 I Finding DA-C7-01 The revision to the data notebook extended the data window. This issue was resolved such Palisades used actual plant procedures and Component starts, stops, and run time data for this period was that it conforms to SR CC-II experience to count surveillance tests. based on recorded data collected from the Maintenance Rule and therefore is not expected Planned maintenance activities are estimated program availability database, plant operating history for the to impact STI change rather than being based on maintenance period, the PI database (from the plant process computer) and evaluations performed in plans. control room operating logs. Plant experience applied to the data accordance with the SFCP.

update was not estimated based on surveillance frequencies.

Independent Assessment:

Finding determined to be CLOSED. This resolution meets the DA-C7 CC-"'"1 requirement. The resolution does not involve a new PRA method or a PRA uoorade.

DA-D1 I Finding DA-D1-01 Prior distributions obtained from NUREG'CR-6928, deemed This issue was resolved such Bayesian updates of all plant specific consistent with plant data, represent the variability in the industry that it conforms to SR CC-II calculations used the industry average data and meet the requirements of standard SR DA-D1. and therefore is not expected distributions. For Category II, it is necessary Similarly, data obtained from NUREG'CR-7037 was developed to impact STI change to update significant basic events using a using distributions to account for variation between plants and evaluations performed in non-informative prior or a prior that represents checked for consistency. accordance with the SFCP.

the variability in industry data.

Independent Assessment:

Finding determined to be CLOSED. This resolution meets the DA-D1 CC-II requirement. The resolution does not involve a new PRA method or a PRA uoorade.

IFEV-A6 I Finding IFSN-A17-01 The current methodology does not rely on the RI-ISI derived pipe This issue was resolved such Generic pipe failure rate data is used if the failure frequency data. All pipe failure frequencies were that it conforms to SR CC-II length of pipe is known. Plant-specific RI-ISI developed in accordance with latest EPRI methodology. All pipe and therefore is not expected data is used if the length of pipe is not readily lengths were obtained from plant isometric drawings. to impact STI change known. This approach appears to meet the evaluations performed in requirements of the Cat I criteria but does not Independent Assessment: accordance with the SFCP.

appear to meet the requirements for Cat II. Finding determined to be CLOSED. This resolution meets the IFSN-A17 and IFEV-A6 CC-II"'"1 requirements. The resolution does not involve a new PRA method or a PRA uoorade.

QU-D4 I Finding QU-D1-01 The documentation of the final model of record (PSAR3), Additional documentation is The final model review has not been including final reviews, is complete and documented in EA-PSA- needed to close this finding.

completed and FPIE-16-03. Comparison of PNPS results documented. [The final review of accident to other similar olants is not Page 24 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 3.

List of SRs Assessed as CC-I in the PNPS Internal Events PRA Model

Statos. ****..*i'li'!i'. illi1!~;~~~~~ig!~~~g~~8!i!~~i sequence results has not been completed Independent Assessment: expected to impact STI change and documented so that the reasonableness Finding remains OPEN. QU-D4 is met at CC-I because no evaluations performed in of the results can be verified. Palisades needs discussion of how Palisades' results compare with those from accordance with the SFCP.

to complete the formal review of accident similar plants was found.

sequence quantification results and make modifications as needed to address issues found in that review. The final results should then be documented in the corresponding notebooks.

LE-C2 HRA values from WCAP-16341-P, which are No model change has been made as generic HRA values which For those STls on which use of generic treatments, were used in the LERF meet CC-I are considered adequate for most PRA applications. generic HRA for the LERF model. No plant-specific HRA evaluation was model is determined to have a performed on Level 2 HRA events. potential impact, the effect is expected to be assessed in the change evaluations for the affected STls.

LE-C9 Suggestion LE-C9-01 No credit is taken for equipment operability or operator actions in This issue was resolved such No credit is taken for equipment survivability adverse environments or after containment failure. Palisades that it conforms to SR CC-II or human actions under adverse reviewed the LERF results for opportunities to take such credit and therefore is not expected environments. I think that equipment (as documented in the Level 2 Notebook) and justified the lack of to impact STI change survivability was reviewed but results did not credit. evaluations performed in justify the use of equipment survivability. Based on way the standard is written, the only way to earn a CC- accordance with the SFCP.

II categorization is to credit equipment operation in adverse environment (for LE-C9 and C-10) and after containment failure (for LE-C11 and C12).

From an equipment context, Palisades does credit equipment in containment in environments that are considered beyond the EEQ harsh environment for which the equipment is qualified in the design basis. Justifications for this credit is provided by engineering evaluations.

Independent Assessment:

Suggestion determined to be CLOSED. As stated above, the wording of SR LE-C9 in the ASME Standard is poor. CC-II/III states justification must be given for credit taken for equipment survivability beyond design. However, when the cutset review determines that taking such credit will not reduce LERF, the intent of the Standard should be interpreted as credit does not HAVE to be taken. Therefore, the iustification orovided in the Page 25 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 3.

List of SRs Assessed as CC-I in the PNPS Internal Events PRA Model

~R.<<i Topic .......................*.................. ........ ...... ... ....... Status.* . ..

/

, Intportanceto.Applicati0l'li.*'**

Level 2 notebook and in MAAP analyses referenced are sufficient to meets the LE-C9 CC-IIIIII requirements. The resolution does not involve a new PRA method or a PRA upgrade.

LE-C10 Suggestion LE-C9-01 See discussion for LE-C9 This issue was resolved such Since LE-C9 is CC-I then this SR also has to that it conforms to SR CC-II be a CC-I. and therefore is not expected to impact STI change evaluations performed in accordance with the SFCP.

LE-C11 Suggestion LE-C9-01 See discussion for LE-C9 This issue was resolved such No credit is taken for equipment survivability that it conforms to SR CC-II or human actions that could be impacted by and therefore is not expected containment failure. Equipment survivability to impact STI change was reviewed but results did not justify the evaluations performed in use of equipment survivability. accordance with the SFCP.

LE-C12 Suggestion LE-C9-01 See discussion for LE-C9 This issue was resolved such Since LE-C11 is CC-I then this SR also has to that it conforms to SR CC-II be a CC-I. and therefore is not expected to impact STI change evaluations performed in accordance with the SFCP.

Page 26 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model F&O .*Ii~~t<s Applicable I. Finding/Observation Disposition I~po~.n<?e .* to SR(s) .....*** AJ>.plication*

PP-A1-01 I Open PP-A 1 I Requirement PP-A01 includes Note PP- In the Palisades fire PRA there were two This finding was (Resolved) A 1-2 which clarifies that the intent of the PAUs qualitatively screened. The qualitative resolved and requirement is to include plant locations screening process and criteria are described therefore is not with no credited plant equipment that may in Section 2 of the Plant Partitioning and Fire expected to impact affect locations with credited plant Ignition Frequency Development Report STlchange equipment in multi compartment fire 0247-07-0005.02. evaluations scenarios. With respect to the multi performed in compartment analysis, the report 0247- accordance with the 07-0005.02 makes no mention on the SFCP.

treatment of qualitatively screened buildings or plant locations.

It is recommended that section 2.1.2.2 of the report clarifying that buildings connected to locations with credited equipment will be considered in the multi compartment fire evaluations.

PP-C2-01 Open PP-C2 It is not entirely clear how some excluded Section 2.1.2.2 of the Plant Partitioning and This finding was (Resolved) areas listed in Section 2.1.2.2 of Report Fire Ignition Frequency Development Report resolved and 0247-07-0005.02 satisfy the exclusion 0247-07-0005.02 has been updated therefore is not criteria, namely the Service Building and to satisfy the exclusion criteria of the Service expected to impact Administrative building. These buildings Building and Administrative Building. The STI change appear to share a common boundary with buildings common boundary with the evaluations the Auxiliary Building. For example, would Auxiliary Building has been detailed and the performed in not a major fire in the Service building be PAUs are retained for MCA analysis. accordance with the designated a challenging fire requiring a SFCP.

plant shutdown? The report states that fires within the Administration Building, Service Building, and Service Building Addition were not expected to propagate to the included physical analysis units, cause a plant transient, or require plant shutdown.

Are excluded buildings permanently excluded, or are they considered during multi-compartment evaluations?

Report 0247-07-0005.02 should indicate that the excluded building sharing a wall Page 27 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model lr~~,.;i?<<\'I:~~~~~: * ***.*.*<i. Ji$~~j~b'j ito/

Fin~ing/Observation Disposition . *1. . *. II1l P.C)rta.nee.

Application" with any included building should be included later in the mUlti-compartment analysis to ensure situations in which a fire in the excluded building may propagate to the adjacent included buildings are evaluated.

ES-A2-01 Open ES-A2 It is unclear at this point if all interlock I A complete review of Safety Injection Signal This finding was (Resolved) permissive circuits which may lead to (SIS), Containment High Pressure (CHP), resolved and specific consequential IEs have been Containment High Radiation (CHR), therefore is not properly captured for the functions being Containment Isolation Signal (CIS) and expected to impact credited in the PRA. This may be Recirculation Actuation Signal (RAS) logic STlchange particularly important where the function was performed to identify potential adverse evaluations credited in the Appendix R analysis is component actuations that could occur due performed in different from the Fire PRA, or auto to a spurious signal from any of these accordance with the actuation of the component is required in sources. SFCP.

the PRA but not in Appendix R.

Logic was added for 45 PRA components to Need to document process by which all consider spurious operation from any of the supporting equipment and interlocks have automatic actuation circuits. The Multiple been addressed. The PRA team appears Spurious Operation Report 0247-07-0005.04 to recognize this deficiency exists at was updated to reflect these changes.

present and have plans in places to rectify once all auto actuation modeling issues are resolved.

ES-A3-01 Open ES-A3 The review of initiating events considered Appendix B of the Model Development This finding was (Resolved) in the internal events analysis is Report 0247-07-0005.03 was resolved and described in Report 0247-07-000503 updated to provide additional detail as to therefore is not Appendix B. A rationale for re-examining how initiating events were screened as to expected to impact the screening process to identify new IE their applicability for fire scenarios. STlchange which may have been screened or evaluations subsumed in that analysis is discussed. The success criteria for consequential LOCA performed in No new initiating events or additional events and their associated pathways and accordance with the equipment were identified. However, the sizes were updated in Section 5.0 of the SFCP.

review process undertaken is not well Event Trees and Success Criteria Notebook documented. It is unclear, e.g., if multiple NB-PSAETSC. The updated notebook coincident pathways were addressed details consequential LOCA events that may when identifying the size of LOCA that result from fire. ConsequentiallSLOCA may be induced by fire and any potential events, potentially caused by fire, are success criteria conflicts which may arise specifically addressed in the XFR-ISLOCA Page 28 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model i:F,;lldi6Q1QbserVati.~rf in the mapping of the fire induced IE to the internal events IE. (e.g. very small LOCA, Small LOCA vs. medium LOCA). Additional detail was also added to the (Note the same concerns arise when Multiple Spurious Operations Report addressing the screening process for 0247-07-0005.04 for the PCP seal Containment isolation pathways where failures and chemical and volume control such pathways were screened on the size system (CVCS) pathways.

of a single pathway.)

The modeling approach for containment Expand the documentation of initiating isolation pathways was updated as event screening process to identify if fire described in Attachment D of the Internal initiating events should be expanded to Events and Fire PRA Model Update Report address issues identified above. EA-PSA-FIRE-12-04.

ES-A5-01 Open ES-A5 A review of the MSO report 0247 The MSO expert panel was reconvened on This finding was (Resolved) ES-B2 0005-04 Appendix A found several 03/15/2011 at the Palisades site to address resolved and deficiencies. These are indicated below. all additions, deletions and/or changes to the therefore is not General: The MSO panel was convened MSO assessment that have occurred due to expected to impact in 2008. Westinghouse published the post-expert panel reviews and in STI change latest MSO report in April 2009 [WCAP- consideration of the most current information evaluations NP-16933]. The current MSO reference available from the PWROG Owner's Group. performed in numbers and description in Appendix A The results of this expert panel review are accordance with the do not match the list in WCAP-NP-16933. documented in the Multiple Spurious SFCP.

There are some new issues which are not Operations Report 0247-07-0005.04:

covered by the current MSO panel report.

  • PLP-1, PLP-2, and PLP-3 were updated.

Suggest a final reconciliation of the MSO Palisades PCP seal LOCA model has panel results [either with a new panel been updated to be consistent with the meeting or a re-write of the report] with latest industry guidance (WCAP-157 49-P, WCAP-NP-16933. Revision 1, December 2008).

PLP-1, PLP-2,PLP-3: The MSO

  • PLP-10 was finalized and the PRA model descriptions in these WCAP issues are updated to include spurious valve failures intended for Westinghouse plants which to address this MSO.

have 2 diverse methods of seal cooling.

  • PLP-11 is correctly evaluated.

The MSO report states the issue is not- Simultaneous spurious closure of CV-applicable to PLP. However, it is 3031 and CV-3057 does not isolate necessary to ensure that all failure charging suction from the SIRWT.

combinations of loss of CCW seal cooling

  • PLP-12 was completed. Spurious closure are included for PLP. of MO-2087 due to fire was added to the PLP-10: Resolution not final.; PLP-11: model.

WCAP issue misunderstood bv MSO Page 29 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model F&O*I.Sta. ~U$

IAP. Plica

  • SR(s).*....*.b

. . . I.e Finding/Observation Disposition *lmport'ln~e.to 1>APp!ication panel. Issue is for closure of both RWST

  • PLP-14 was updated. The evaluation suction valves and is applicable to PLP. describes how this scenario is addressed Simultaneous spurious closure of CV3031 in the model.

and CV3057 is this issue. ;PLP-12:

  • PLP-18 was updated. The evaluation Resolution not final; PLP-14: In WCAP- describes model changes incorporated to NP-16933, issue 14 is applicable to explicitly address early drain down of the Palisades. Issue 14 is CHP runout when SIRWT and dead-heading of the ECCS RCS is depressurized. Palisades needs to pumps.

look at pump runout possibility for all

  • PLP-19 was finalized. This scenario ECCS, CCW, AFW, and SWS pumps.
  • involves early drain down of the SIRWT PLP-18: Resolution for PLP-18 states via containment spray and is addressed in RWST may drain, which is not considered the resolution to scenario PLP-18.

in PRA. If RAS occurs and CV-3029 or

  • PLP-27, PLP-34, PLP-35 were revised to CV03030 opens, RWST will not drain, address affects other than cooldown due because of check valve in sump line. to a stuck open atmospheric dump valve Other possibilities involving deadhead 1 (ADV).

NPSH of ECCS pumps are not explored.

  • PLP-39 had additional evaluation If 3029/3030 open on a spurious signal, performed. This evaluation provides the CV3031/CV3057 and CV3027/3056 will basis for excluding the blowdown valves receive a signal to close. Power is as a potential flow diversion path.

disabled to 3027/3056. Scenarios for

  • PLP-43 had additional evaluation insufficient NPSH include a) spurious SI; performed. This evaluation provides the b) opening of 3029/3030; c) closure of basis for excluding spurious opening of 3031/3057. Possibilities for ECCS the pressurizer spray valves from the deadhead include: a) spurious SI;b) model.

opening of 3029/3030; c) operator

  • PLP-45 was validated that it is correctly mistakenly restores power to 3027/3056 identified in the CAFTA model.

[based on false instruments] resulting in

  • PLP-47 was finalized with the addition of deadhead of ECCS pumps. PLP-19:

evaluations to describe the treatment of Needs final resolution; PLP-27, PLP-boron dilution events in the PRA.

34,PLP-35: MSO states SG-ADV does not need to be included because

  • PLP-57 and PLP-58 have been finalized overcooling is not an issue at Palisades. and incorporated into the PRA model.

However, need to consider other affects

  • PLP-60, PLP-80, and PLP-84 have been of SO ADV, which are: a) AFW pump finalized.

runout; b) Faulted SG may be unsuitable

  • The modeling approach for containment for decay heat removal in the long term isolation pathways was updated.

[i.e .. , not able to raise steam].; c) Heat Consequential ISLOCA and containment removal is less than effective and bypass events, potentially caused by fire, Page 30 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model I~~f)< . * *:< .* *. <>. ...........

'Status ". .. . Applicable FindinglObservation I**' Disposition ' .. Imp()~nc~to '.

SRls) ......*

. ......*.........* .. Application ....*..............

condensate inventory makeup is required. are specifically addressed in the XFR-PLP-39: Need better reason to exclude ISLOCA event tree.

blowdown valves as potential flow diversion rates appear to be significant. ;

PLP-43: Resolution for pressurizer spray valves states that SO spray valve would lead to loss of subcooling. Loss of subcooling will lead to SI signal. Spurious spray valve opening will lead to SI in [on the order of] 5 minutes. Spray valve spurious should be included in Fire PRA as leading to SI signal.

PLP-45: Basic events for pressurizer heaters could not be found in CAFTA as indicated in the resolution, PLP-47: Resolution not final. PLP-57 PLP-58: Effect of spurious operation of load sequencers no evaluated. Possible scenarios include 1) failure of cable causes spurious load shed on operating bus, 2) failure of cable causes load of DG on operating bus. PLP-60,PLP-80,PLP-84: Need final resolution.

General: No indication of search for containment isolation failure pathways which can contribute to LERF.

Correct deficiencies identified in the MSO report identified in this F&O and complete resolution of outstanding issues ES-C1-01 Open ES-C1 Since the full complement of OMAs to be Instrumentation relevant to operator actions This finding was (Resolved) included in the fire PRA has yet to be in fire scenarios were identified and validated resolved and identified instrument set is incomplete. by completion of Post-Initiator Operator therefore is not Action Questionnaires (P-IOAQ). expected to impact A copy of the Human Error Probability (HEP) STlchange Post-Initiator Calculation (PIC) and P-IOAQ evaluations were provided to current SRO licensed on- performed in shift Operations Department personnel and accordance with the Traininq Department personnel for use in SFCP.

Page 31 of 57

ATTACHMENT 2 PNP 2018*025

Open Finding F&Os Against the PNPS Fire PRA Model n accuracy.

HFEs were assigned to Operations Department Operating Crews and lor Operations training personnel for review.

Their reviews included ensuring indications, procedure selection and use, and activity performance manpower and timing is correct. Training personnel reviews included ensuring procedure selection and use were consistent with current training expectations, and the training type and frequency are accurate.

The records of the current operating crews and training personnel are provided in the final set of operator manual actions (OMAs),

and HRA Notebook NB-PSA-HR.

ES-C2-01 Open ES-C2 Instruments which provide supporting A simulator exercise was performed with This finding was (Resolved) cues for operator actions have been current Palisades' license holders in which resolved and identified and are being explicitly modeled several scenarios were evaluated to therefore is not in the fire PRA together with their determine how Operators would respond expected to impact associated power supplies given spurious or false instrument STI change Undesired operator actions potentially indications. evaluations occurring as a result of spurious plant The results of these exercises were performed in monitoring and alarm instruments do not considered in the HEP development accordance with the appear to have been addressed at the process. The process and evaluation results SFCP.

present time. Neither has a process for are documented in the HRA Notebook NB-identifying, screening and modeling such PSA-HR.

occurrences been discussed. This should be addressed.

Recommend using screening process for alarms included in NUREG/CR 6850.

Undesired actions in the context of both the alarm response procedures and EOPs.

CS-A9-01 Open updates to the Although the data gathering is complete, it Additional update is (Unresolved) I CS-C4 original cable selection to ensure multiple was not fully implemented into the model. needed to close this hot short failures are identified. It is not Cable data for the PLP FPRA was obtained finding. For those evident that the suoolemental ::In::llvl':.il':. from two seoarate sources: the SAFE STls on which the Page 32 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model work specifically looked for proper polarity I database and NEXUS spreadsheets. The update of the cable hot shorts on ungrounded DC circuits. SAFE database was populated with the selection to reflect original Palisades Appendix R cable data, consideration of If the supplemental cable selection work the cable data collected for offsite power proper polarity hot included consideration of proper polarity components, and the initial set of shorts is determined hot shorts, update the criteria used for the components selected for cable analysis for to have a potential analysis to clearly reflect this fault mode. the fire PRA. impact, the effect is If this fault mode was not considered, expected to be additional work will be needed to meet Subsequent efforts were performed to assessed in the this supporting requirement. analyze additional components, refine change evaluations previously collected cable data and to revisit for the affected STls.

vintage data using modern criteria including ro er olarit dc hot shorts.

CS-B1-01 I Open CS-B1 I The analysis and review of electrical Palisades has documented a complete Plant (Resolved) CS-C4 overcurrent coordination and protection breaker coordination study for all are needed as a has been initiated but is not yet complete. buses considered in the fire PRA as result of this finding.

The final analysis should address described in the Safe Shutdown Associated For those STls on coordination for all Fire PRA electrical Circuits Analysis EA-APR-95-004. which the distribution buses. Refer to F&O CS-C4- Modifications will be installed in 2018 and modifications to 01 for a related discussion on 2020 to resolve the identified Electrical address electrical documentation of the coordination and Coordination Challenges to address all coordination is protection analysis. buses where electrical coordination could not determined to have be demonstrated. a potential impact, Complete the coordination study and the effect is provide supporting documentation as expected to be suggested. Ensure the analysis includes assessed in the cases where overcurrent protection for change evaluations medium voltage and low voltage for the affected STls.

switchgear might not be available due to a fire-induced loss of 125 VDC tripping power to power circuit breakers in the switch ear.

CS-C1-01 I Open I CS-C1 I The cable selection and location Although the data verification is complete, Additional update is (Unresolved) methodology is documented in Section 4 the results have not been fully implemented needed to close this the Model Development Report (0247 into the model. Section 4 of the Model finding. For those 0005.03) and associated appendices. The Development Report 0247-07-0005.03 has STls on which the methodology for completed work is been updated in a manner that ensures update of the criteria documented in a manner consistent with consistent interpretation of Fire PRA used for this supporting requirement; however, the applications. supplemental cable Page 33 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model Disposition.

.*I**=~tUSi **AP. plic:able ., . . ~1.*.n. d. . .*.*.ing/O~servation SR(s) -

--'-'--_.........."'"'-',c. .....:.mL.;:L-p~~a:;:to

~

methodology for the supplemental cable selection is selection review (Attachment 1) is not Additionally, the verification of Appendix R determined to have formally documented in a manner that Non-Safe Shutdown Cable Routing to a potential impact, ensures consistent interpretation for Fire Support the Fire PRA has been separately the effect is PRA applications and upgrades. documented in the Validation of Appendix R expected to be Additionally, the sample cable routing Non-Safe Shutdown Cable Routing to assessed in the verification check is not formally Support the Fire PRA Report PLP-RPT- change evaluations documented in the Fire PRA Report or 12-0134. for the affected STls.

any other plant document, and thus does not lend itself to consistent treatment for future Fire PRA applications and upgrades.

1. Formalize Attachment 1 and expand on the existing criteria used for the supplemental cable selection work.
2. Formalize the sample cable routing check and incorporate it in the Fire PRA Report or appropriate plant evaluation/calculation.

CS-C4-01 Open CS-B1 Unlike other elements of this Technical Palisades has documented a complete This finding was (Resolved) CS-C4 Element, the Fire PRA Report does not breaker coordination study for all resolved and address the methodology, process, or buses considered in the fire PRA as therefore is not criteria for the electrical coordination and described in the Safe Shutdown Associated expected to impact protection analysis. This information will Circuits Analysis EA-APR-95-004. STlchange need to be included in final documents to evaluations satisfy this supporting requirement. performed in accordance with the Update Section 4 of the Model SFCP.

Development Report to address electrical coordination and protection.

QLS-B2-01 Open PP-C2 See PP-C2-01 Section 2.1.2.2 of the Plant Partitioning and This finding was (Resolved) QLS-B2 Fire Ignition Frequency Development Report resolved and 0247-07-0005.02 has been updated therefore is not to satisfy the exclusion criteria of the Service expected to impact Building and Administrative Building. The STlchange buildings common boundary with the evaluations Auxiliary Building has been detailed and the performed in PAUs are retained for MCA analysis. accordance with the Page 34 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model F&O.*. . . i.~. Stat.us .*....... ***)i ~pplic~ble .. i Finding/Observation Disposition '. .... .IIllPortanceto7 . *. .*

.* > * < > i SR(s). *... ......... ...

~

...... Application* ... * *

  • c '

SFCP.

PRM-B3-01 Open PRM-B3 The fault tree model development omitted The fault tree model applied to the fire PRA This finding was (Resolved) the DC power dependency requirement was updated to include the DC power resolved and for the RCP breaker trip function. dependency for the primary coolant pump therefore is not breaker trip function. This logic was added expected to impact for each of the four primary coolant pumps. STlchange evaluations performed in accordance with the SFCP.

PRM-B3-02 Open PRM-B3 Spurious SI is not included as a potential A complete review of Safety Injection Signal This finding was (Resolved) initiating event (SIS). Containment High Pressure (CHP). resolved and Containment High Radiation (CHR). therefore is not Containment Isolation Signal (CIS) and expected to impact Recirculation Actuation Signal (RAS) logic STlchange was performed to identify potential adverse evaluations component actuations that could occur due performed in to a spurious signal from any of these accordance with the sources. SFCP.

Logic was added for 45 PRA components to consider spurious operation from any of the automatic actuation circuits. The Multiple Spurious Operation Report 0247-07-0005.04 was updated to reflect these changes.

PRM-B5-01 Open PRM-B5 The MSO expert panel issues have not All MSO expert panel issues have been This finding was (Resolved) been completely resolved and resolved and integrated into the final PRA resolved and incorporated into the PRA model. Thus. fire model as appropriate. All MSO scenario therefore is not all modeling work associated with MSO dispositions are documented in the final expected to impact incorporation has not been done at this Multiple Spurious Operation Report 0247 STI change time. 0005.04. evaluations performed in accordance with the I SFCP.

PRM-B9-01 Open PRM-B9 Failure to trip Pressurizer heaters is not A fault tree was added to the fire PRA to This finding was (Resolved) explicitly addressed model spurious operation of pressurizer resolved and heaters and failure of pressurizer spray. therefore is not Develop basis Failure of this faulttree reslJl1sjna potent~ expected to impact Page 35 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model evaluations performed in accordance with the SFCP.

PRM-B11- Open PRM-B11 Complete work (to MODEL all operator The procedures, modification detail, Additional update is 01 (Unresolved) actions and operator influences in operations review, and detailed HRA model needed to close this accordance with the HRA element of this development are not yet complete. finding. For those Standard.) Screening values are still applied for fire STls on which the HEPs pending development of final detailed HRA is procedures, modifications, and operations determined to have reviews. a potential impact, the effect is expected to be assessed in the change evaluations for the affected STls.

FSS-A1-01 Open FSS-A1 The treatment of MCC's is not properly Section 6.1 of the Fire Scenario This finding was (Resolved) justified. FSS document 0247-07-0005.06 Development Report 0247-07-0005.06 has resolved and includes the statement "All Motor Control been revised to include a reference to the therefore is not Centers (MCC) have been treated as walkdown information and photographs expected to impact closed, sealed and robust in which which provide a basis for this statement. STI change damage beyond the ignition source will evaluations not be postulated." No documentation of performed in inspections of the MCC's, including the accordance with the top of the cabinets have been provided to SFCP.

justify not propagating fires outside the MCC.

Conduct and document a detail walkdown to gather data associated with the criteria for classification of MCC's as sealed, robust and secure panels. Use the results of the walkdowns to document the justification of the conclusion of not ro a atin fires outside the MCC's.

FSS-A3-01 Open FSS-A3 The process of mapping and accounting In the event that a cable's plant location This finding was (Resolved) for targets in the Fire PRA is not cannot be established, the process of resolved and documented. Technical discussion durin creditin b assumed routin was erformed. therefore is not Page 36 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model I;~~~ '1~~tlJSJ *'*Applicable Fh'lding/Observ~tion I DiSPositiO": '. Impo~ncet()

Applic.ation . . .

.SR(s) the review period indicate that targets with The process involved determining, with a expected to impact unknown routing are mapped to all the high degree of confidence, locations in the STlchange scenarios within a PAU unless it has been plant that do not contain the cable in evaluations verified that the target is not in a specific question. This is accomplished by performed in scenario. However, this process was not considering the likely routing of a cable and accordance with the clearly demonstrated during the review was performed by experienced plant SFCP.

and is not documented in report 0247 personnel. In many cases, this assessment 0005.06. was made by grouping components into an appropriate surrogate category. The results The process of treating targets should be of this detailed assessment are provided in clearly documented in report 0247 the Model Development Report 0247 0005.06. Specifically how unknown 0005.03.

conduit or cable tray locations have been mapped to scenarios in the PAU.

Documentation should point readers to tables where the treatment could be verified.

FSS-B1-01 Open I FSS-B1 The current Fire PRA does not consider Control room abandonment scenarios with Additional update is (Unresolved) abandonment of the main control room respect to environmental effects have been needed to close this due to lack of equipment/control due to addressed. However, the current model finding. For those fire damage. does not address abandonment due to STls on which the equipment damage. control Include in the analysis a criteria for control abandonment room abandonment due to lack of Main Control room abandonment scenarios scenarios is equipment/control failure. have been postulated based on damage to determined to have equipment and controls. Postulated fires in a potential impact, the Control Room (CR) have the potential to the effect is challenge habitability or visibility due to expected to be smoke generation or excessive heat. An assessed in the abandonment analysis documented in the change evaluations Fire Scenario Development Report 0247 for the affected STls.

0005.06 was performed to determine the response of the CR envelope given a range of possible fire events. The analysis considered three different operating states of the CR mechanical ventilation system and three different configurations of the CR Door.

FSS-B2-01 Open I FSS-B2 The CCDP quantification does not reflect Main Control room abandonment scenarios This finding was (Resolved) the human error probabilities associated have been postulated based on damage to resolved and Page 37 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Findin9 F&Os Against the PNPS Fire PRA Model

. . . . .s. ~.rv Disposition 101 portan~eit()

I.*. *. AP. pJic........................................

      • SR(s) a.*. bl.e.*

. *. . *. * **. *. . * .***I*.*.*. .F. *. . n

i. . . . din.*. . .9. ... '.*. O

. . . *. . b . . *. *. *. a. .* tiO. n. Application*

with control room abandonment and the equipment and controls. Postulated fires in therefore is not fire impacted cables may not reflect the the Control Room (CR) have the potential to expected to impact equipment/control that mayor may not be challenge habitability or visibility due to STlchange available after abandonment. smoke generation or excessive heat. An evaluations abandonment analysis documented in the performed in Properly model control abandonment Fire Scenario Development Report 0247 accordance with the using the criteria based on fire generated 0005.06 was performed to determine the SFCP.

conditions and plant operability so that the response of the CR envelope given a range CCDP accounts for the human error of possible fire events. The analysis probabilities associated with considered three different operating states of abandonment conditions and the the CR mechanical ventilation system and available equipment and controls that three different configurations of the CR Door.

may be affected by the fire.

FSS-C4-01 Open FSS-C4 The severity factor for hotwork fires of The 0.01 severity factor for hotwork is no This finding was (Resolved) 0.01 is not properly justified. The longer applied in final fire PRA model. resolved and documentation does not provide a Severity factors are now based on therefore is not description how the value was calculated NUREG/CR-6850. Section 8.3 of the Fire expected to impact and an explanation of why the value Scenario Development Report 0247 STlchange remains independent of the generic 0005.06 was updated to reflect this change. evaluations ignition frequency. performed in accordance with the Add a justification for the value of 0.0.01 SFCP.

for hotwork fires. The justification should include the process for determining the value and an explanation of why the value remains independent of the generic ignition frequency.

FSS-C5-01 Open FSS-C5 No scenario is evaluated for conditions Section 5.2 of the Fire Scenario This finding was (Resolved) where the target damage criteria is that of Development Report 0247-07-0005.06 was resolved and sensitive electronics. revised to provide further basis for excluding therefore is not scenarios with the sensitive electronics expected to impact The analysis documented in report 0247- criteria. The exclusion is based primarily on STI change 07-0005.06 suggest locations where physical cabinet distances from fire ignition evaluations sensitive electronics may be targets. The sources and that these targets are generally performed in multi compartment and single within an enclosure that provides some accordance with the compartment analysis should use the protection from the heat source. SFCP.

lower damage criteria for sensitive electronics in these locations.

Page 38 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model Open No evaluation of independence of Section 10.1 of the Fire Scenario This finding was (Resolved) suppression paths have been included in Development Report 0247-07-0005.06 was resolved and the analysis. revised to describe the treatment of therefore is not dependence between suppression paths in expected to impact Evaluate and justify that the credited the scenario suppression event tree. STI change suppression features are indeed evaluations independent in support of the current performed in analysis, or alternative, incorporate in the accordance with the rh>npnrlpn~v found. SFCP.

FSS-C8-01 Open report does not discuss the treatment Documentation was added to Section 2.2 of This finding was (Resolved) of fire barriers credited in the analysis. the Fire Scenario Development Report 0247- resolved and 07-0005.02 which discusses the treatment of therefore is not Describe and document the treatment of fire barriers credited in the analysis. expected to impact passive fire protection features credited STI change for cable protection in the Fire PRA. This evaluations include documentation of the rating of the performed in barrier and how it is incorporated in the accordance with the SFCP.

FSS-D1-01 Open FSS-D1 Although in general appropriate fire Appendix E of the Fire Scenario This tlndlng was (Resolved) models have been selected, the Development Report 0247-07-0005.06 was resolved and justification for the use of the selected updated to include further discussion on the therefore is not tools need to be improved. This finding is applicability of the MathCAD tool for expected to impact specifically applicable to the use of the calculation of the non-suppression STlchange time to damage models programmed in probability. evaluations MathCad, which are calculations that performed in have not been documented and reviewed accordance with the by the industry. SFCP.

Develop and justify the application of the damage time data available in NUREG/CR-6850 in the approach implemented in the Fire PRA. Suggest a sensitivity analysis for key scenarios is documented describing the range of applicability and discussing the impact when compared with the use of Appendix H tables only and a deterministic heat transfer calc (e.o. THI Page 39 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model Open I FSS-D2 I No fire detection analysis has been Section 10.1 of the Fire Scenario This finding was (Resolved) conducted in support of the activation of Development Report 0247-07-0005.06 was resolved and fixed suppression systems or the time to revised to describe the treatment of therefore is not smoke detection. automatic suppression system activation expected to impact times on the suppression probability. STlchange I

Include in the analysis time to detection evaluations calculations. performed in accordance with the SFCP.

FSS-D4-01 Open FSS-D4 This finding is associated with treatment Section 7.0 of the Fire Scenario This finding was (Resolved) of transient fires. 1) Fire elevation for Development Report 0247-07-0005.06 was resolved and transient fires has been assumed to be on revised to describe the treatment of fire therefore is not the floor. 2) the heat release rate for elevation and heat release rate for transient expected to impact transient fires have been assumed to be fires. The transient heat release rate was STI change characterized by electric motor fires. increased to 317 kW; 98th percentile heat evaluations These are important input values for release rate for transient combustibles, in performed in determining zone of influence. lieu of the value for electric motor fires. accordance with the SFCP.

Consider a higher fire elevation to account for transient fires elevated from the floor. Consider using the heat release rate probability distribution for transient fires instead of the one for electric motors.

FSS-D7-01 I Open , FSS-D7 ,Items a, b, and c in the Cat II requirement Section 10.0 of the Fire Scenario This finding was (Resolved) are not explicitly addressed in the Development Report 0247-07-0005.06 was resolved and analysis. revised to describe the basis for availability therefore is not of automatic suppression systems and the expected to impact Conduct a qualitative or quantitative study impact on suppression probability. STlchange addressing items a, b, and c in the Cat II evaluations requirement in support of the use of performed in generic values. accordance with the SFCP.

FSS-D8-01 I Open , FSS-D8 , The Fire PRA currently does not include Section 10.1 of the Fire Scenario This finding was (Resolved) an assessment of the effectiveness of the Development Report 0247-07-0005.06 was resolved and fire suppression and detection systems revised to provide an assessment of the therefore is not credited in the analysis. effectiveness of automatic suppression expected to impact systems and the impact on suppression STI change Evaluate time to detection and I probability. evaluations performed in suppression. Include walkdown notes Page 40 of 57

ATTACHMENT 2 PNP 2018*025 . Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model

[~~i,'"i2.*. ***.*.\I.*~~t.U$' AP. P.liC Sl~.ls)

.* ableil .**.*.**.F

. . . iQd. . . ing'.O

. *. *. . bserVatio. n I. DiSpoS. . ition Il11p°rlcinc(;ltO*.

Application documenting the inspection and accordance with the evaluation of the effectiveness of the SFCP.

system to control fires in the postulated scenarios.

FSS-E3-01 Open I FSS-E3 A qualitative characterization of the A characterization of the parameters used in Additional update is (Unresolved) parameters used in the fire modeling in the fire modeling in significant fire scenarios needed to close this significant fire scenarios have not been has not been completed. However, this does finding.

completed as the Fire PRA still needs not impact the point estimate numerical Characterization of detailed analysis to reduced the plant results of the fire PRA. fire modeling CDF. The qualitative discussion required parameters is to meet category 1 should be completed expected to have no once key scenarios are identified. impact on the fire PRA results or the Add a qualitative discussion of the STlchange uncertainty in fire modeling parameters evaluations for the significant scenarios once those performed in are identified. accordance with the SFCP.

FSS-F1-01 Open I FSS-F1 The report 0247-07-0005.08, which The definition of a significant fire hazard was This finding was (Resolved) documents structural steel analysis, does added to Section 2.0 of the Exposed resolved and not describe what is a "high hazard fire". Structural Steel Analysis Report 0247 therefore is not Consequently, it is not clear what specific 0005.08: A significant fire hazard was expected to impact fires where considered as high hazard defined as having at least the same or STI change during the walkdowns and analysis to greater combustible loading equivalent to 50 evaluations conclude that a scenario should be gallons of fuel oil, which is in excess of a performed in quantified in the analysis. heat value of 7E+6 BTU." accordance with the SFCP.

Add a definition of "high hazard fire" and apply such definition in the analysis.

FSS-F2-01 Open I FSS-F2 The criteria for identifying and analyzing The appropriate criteria for fire damage to This finding was (Resolved) fire scenarios associated with damage to structural steel were added to Section 3.0 of resolved and structural steel is not clearly documented. the, Exposed Structural Steel Analysis therefore is not The criteria utilized has been inferred Report 0247-07-0005.08 to clearly document expected to impact from the analysis and is considered the criteria used for identifying and STlchange appropriate. The criteria includes 1) analyzing fire scenarios associated with evaluations possibility of a high hazard fire, 2) structural steel damage: 1) presence of performed in exposed structural steel, and 3) a steel significant fire hazard, 2) presence of accordance with the temperature of 1000 F. exposed structural steel, 3) steel surface SFCP.

temperature in excess of 1OOO°F for fire Page 41 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model Clearly document the criteria for identifying scenarios associated with dama e to structural steel elements.

Open FSS-F3 The four scenarios selected for evaluation Section 3.2.2 of the Exposed Structural Steel This finding was (Resolved) have been screened and therefore not Analysis Report 0247-07-0005.08 was resolved and included in the CDF calculation for the revised using the frequencies found in therefore is not plant. The screening process for one of NUREG/CR-6850 and EPRI TR 1016735 to expected to impact the scenarios is based on the frequency calculate a new turbine-generator STI change of such an event (PAU-23, turbine catastrophic fire frequency. Quantitative evaluations generator fire). The calculated frequency calculations and factors applied are also performed in is not based on fire ignition frequencies documented. Site Specific frequencies accordance with the documented in current Fire PRA EPRI documented in the Fire Ignition Frequency SFCP.

guidance. and Plant Partitioning Report 0247 0005.02 were implemented in the Revise the frequency analysis for the I quantitative assessment of the FPRA.

turbine generator scenario and re-evaluated the screenin decision.

FSS-G2-01 Open FSS-G2 Elements of the qualitative criteria require The assumptions in Section 1.1 and the This finding was (Resolved) further evaluation. Specifically, "exposing screening criteria in Table 3-1 of the Multi- resolved and PAU is outdoors; no HGL postulated" and Compartment Analysis 0247-07-0005.07 therefore is not "exposed PAU has a sufficient volume were revised to add discussion of outdoor expected to impact that any hot gases that may enter PAU transformers near turbine building walls and STlchange would dissipate before significant damage hot gas layer. evaluations would occur." In the former, the qualitative performed in assessment should include a discussion accordance with the of yard transformer fires near turbine SFCP.

building walls. In the later, assessment of hot gas layer conditions should be quantitatively address.

Some combinations of PAU's may require a description/justification of the applicability of the qualitative screening criteria invoked.

FSS-G2-02 Open FSS-G2 The quantitative screening criteria do not Section 3.5 of the Multi-Compartment This finding was (Resolved) include consideration for the cumulative Analysis 0247-07-0005.07 was revised to resolved and risk screened out due to multi describe the cumulative impact of CDF therefore is not compartment combinations. Currently, screening at 1E-7. expected to impact multi com artments are screened at a STI Page 42 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model evaluations performed in screened. accordance with the SFCP.

Include in the screening criteria a verification of the cumulative risk screened out so that the screening nrnr.Al':l': consistent with QNS-A 1.

FSS-G4-01 Open FSS-G4 The SR requires confirmation of al _ was (Resolved) credit, assessment of effectiveness and Analysis 0247-07-0005.07 was revised to resolved and reliability, and evaluation of random describe the applicability and basis for the therefore is not failures of passive barriers. No analysis random failure probability of passive fire expected to impact has been presented or documented barriers from NUREG/CR-6850 used in the STlchange addressing these requirements. mUlti-compartment analysis. evaluations performed in Provide an assessment of the rating and accordance with the integrity of the barriers that would support SFCP.

the failure probabilities of the barriers incorporated in the analysis. This may consist of a walkdown to inspecUconfirm boundaries and reference to inspection res and results.

FSS-G5-01 Open FSS-G5 The SR requires quantification of Appendix A of the Multi-Compartment This finding was (Resolved) effectiveness, reliability and availability of Analysis 0247-07-0005.07 was revised to resolved and the active fire barriers. No analysis has describe the applicability and basis for the therefore is not been presented or documented random failure probability of active fire expected to impact addressing these requirements in addition barriers from NUREG/CR-6850 used in the STlchange of using the generic values in mUlti-compartment analysis (MCA). Table evaluations NUREG/CR-6850. 3-4 of this analysis reflects the quantification performed in of MCA interaction failures. accordance with the Refer to inspection records of the barriers SFCP.

to identify if the generic values for barrier failure probabilities in NUREG/CR 6850 are applicable and document this assessment.

The treatment of hydrogen fires is The documentation for treatment of This finding was incorrectly documented in report 0247 hydrogen fires in the Fire Scenario resolved and 0005.06. Development Report 0247-07-0005.06 was therefore is not to be consistent with Page 43 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model associated with change treatment of hydrogen fires to reflect evaluations current practice discuss during the peer performed in review. accordance with the SFCP.

Open FSS-H5 The Fire PRA is in process. Fire modeling The final fire modeling output results for This finding was (Resolved) results are not complete. Documentation each analyzed fire scenario were resolved and of output results should be consistent with documented in Section 6.0 of the Fire Risk therefore is not current approach for scenarios analyzed Quantification and Summary Report 0247- expected to impact while the fire PRA is completed. 07-0005.01. STlchange evaluations Document fire modeling outputs performed in consistent with the requirements of Cat II. accordance with the SFCP.

FSS-H9-01 Open FSS-H9 Sources of uncertainty in the fire modeling Sources of uncertainty in the fire scenario This finding was (Resolved) analysis are not documented in 0247 selection process were documented in the resolved and 0005.06. Fire Risk Quantification and Summary therefore is not Report 0247-07-0005.01. expected to impact Document sources of uncertainty in the STI change fire modeling analysis. evaluations performed in accordance with the SFCP.

IGN-A7-01 Open IGN-A7 Page 3-1 of report 0247-07-0005.02 The Plant Partitioning and Fire Ignition This finding was (Resolved) appears to suggest that no frequency for Frequency Development Report 0247.07 resolved and miscellaneous hydrogen fires has been 0005.02 and Fire Ignition Frequency therefore is not assigned to applicable physical analysis calculation database have been updated to expected to impact units. This may affect the PAU level assign miscellaneous hydrogen fires to STI change quantification by reducing the fire ignition all applicable Physical Analysis Units evaluations frequency assigned to the applicable plant (PAUs). The frequency associated with Bin performed in locations. 19, Miscellaneous Hydrogen Fires, has been accordance with the allocated based on linear feet, valve location SFCP.

Add the contribution from miscellaneous and tank location in PAUs where hydrogen hydrogen fires to applicable PAU'S. equipment exists. Applying these criteria has apportioned miscellaneous hydrogen frequency to the following PAUs: 04 (1C Switchgear Room), 13 (Reactor Building),

and 23 (Turbine Bui Page 44 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model Open The characterization of uncertainties in The characterization of uncertainties in the This finding was (Resolved) the fire ignition frequencies has not been fire ignition frequencies has been addressed resolved and addressed in the report qualitatively or in the Fire Risk Quantification and Summary therefore is not quantitatively. Report 0247.07.005.01. expected to impact STI change The uncertainties in the fire ignition The report describes the sensitivities run by evaluations frequencies should be discussed changing the bin ignition frequencies to the performed in qualitatively or quantitatively in the report 5th and 95th percentile values of the original accordance with the so that a category can be assigned. frequencies for both EPRI and NUREG/CR- SFCP.

6850 values. This sensitivity provides an adequate upper and lower bound of the final CDF which used the mean HRA-A2-0 HRA-A2 The identification of fire response actions The final identification of fire response This finding was (Resolved) is not yet complete. Additional fire safe actions was completed and documented in resolved and shutdown actions are still being identified the Human Reliability Analysis Notebook therefore is not as the Fire PRA analysis continues to be NB-PSA-HR-1. These actions were expected to impact refined. incorporated into the final fire PRA model STlchange where appropriate. evaluations Complete the identification of fire performed in response actions necessary to make the accordance with the response to all risk significant fire SFCP.

scenarios realistic.

HRA-A3-01 Open HRA-A3 Section 6.3 of the HRA Notebook A simulator exercise was performed with This finding was (Resolved) discusses the review that was performed current Palisades' license holders in which resolved and with the licensed operators for the several scenarios were evaluated to therefore is not identification of the new, undesired determine how Operators would respond expected to impact operator actions in response to spurious given spurious or false instrument STlchange indications. However, the detailed indications. The results of these exercises evaluations documentation for the evaluation process were considered in the HEP development performed in and the justifications for the conclusion process. The process and evaluation results accordance with the that no undesired operator actions will be are documented in the Human Reliability SFCP.

taken in these instrumentation failure Analysis Notebook NB-PSA-HR.

conditions was not yet completed for the reviewers to confirm the conclusion that no undesired operator actions need to be considered.

Complete the documentation for the irl">ntifir<>tinn and eval Page 45 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model for the justifications of the evaluation conclusion. Also, complete the review of the ARPs. It is expected that Category II can be met when the documentation is comolete and conclusion verified.

HRA-A4-01 Open I H RA-M As the fire scenario refinement continues, The procedures, modification detail, Additional update is (Unresolved) additional fire response actions will be operations review, and detailed HRA model needed to close this identified and evaluated, which will development are not yet complete. finding. For those require the performance of additional A copy of the Human Failure Event (HFE) STls on which operator interviews. As such, this task is Post-Initiator Calculation (P-IC) and detailed HRA is not fully completed yet. Also, operator associated Post-Initiator Operator Action determined to have interviews for those fire response actions Questionnaire (P-IOAQ) were provided to a potential impact, that are still using screening values (e.g., current SRO licensed on-shift Operations the effect is ACP-DGOT-B5B-DG, ACP-PMOE-383- Department personnel and Training expected to be 11A, ACP-PMOE-383-12A, AFW-PMOA- Department personnel for use in validating assessed in the P8B-CRAB, etc.) may not have been HEP information accuracy. change evaluations completed. for the affected STls.

HFEs were assigned to Operations Complete the fire scenario refinement, Department Operating Crews and lor HRA for fire response actions, and Operations training personnel for review.

operator interviews for the new fire Their reviews included ensuring indications, response actions. procedure selection and use, and activity performance manpower and timing is correct. Operator comments were reviewed and discussed with PRA personnel and resolutions forwarded to the comment initiator for acceptance.

Significant HFEs were evaluated and developed in further detail. Screening values are still applied for fire HEPs pending development of final procedures, modifications, and operations reviews.

The records of the current operating crews and training personnel are provided in the Human Reliability Analysis notebook NB-R.

HRA-B2-01 HRA-B2 new, fire response actions fire response This finding was of the identified fire nd documented in resolved and Page 46 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model the Human Reliability Analysis notebook NB- therefore is not PSA-HR. These actions were incorporated expected to impact into the final fire PRA model where STI change Complete the incorporation of the appropriate. evaluations identified fire response actions and the performed in identification of new, fire response HFEs accordance with the as the refinement of fire scenario analysis SFCP.

HRA-B3-01 Open HRA-B3 The impact of loss of all The procedures, modification detail, Additional update is (Unresolved) redundant/diverse instrumentation on operations review, and detailed HRA model needed to close this HEPs has been modeled by OR-ing the development are not yet complete. finding. For those instrumentation logic with its associated The simulator exercise performed with STls on which HEP. Thus, in cases where total current Palisades' license holders detailed HRA is instrument failure (by hardware fault or evaluated operator response to several determined to have fire) occurs (including the failure of the scenarios with false, partial or total a potential impact, only instrument available), the HEP is loss of instrument indications. The results of the effect is appropriately failed. However, the failure these exercises were considered in the HFE expected to be impact of partial instrumentation on an development process for purposes of assessed in the HEP has not yet been implemented. developing timing of cues and time windows. change evaluations There are cases in the model where for the affected STls.

multiple instruments provide cues to the The final developed fire HFEs incorporate operators to perform actions. Operator task complexity and procedural guidance as actions based on false indication have not documented in the Post-Initiator Operator been considered. In addition, HFEs Action Questionnaire (P-IOAQ) provided to modeled using screening values (for current SRO licensed on-shift Operations some of the fire response actions Department personnel and Training identified; e.g., ACP-DGOT-B5B-DG, Department personnel for use in validating FPS-PMOE-START-L, ACP-PMOE-383- HFE information accuracy.

12A, ACP-PMOE-383-11A, etc.) and those fire response actions that will be Significant HFEs were evaluated and identified as the fire scenario refinement developed in further detail. Screening values continues have not yet accounted for the are still applied for fire HEPs pending scenario context including timing, development of final procedures, procedural guidance, instrumentation, modifications, and operations reviews.

task complexity, etc. Also, HRA Calculator The final list of fire HFEs and their evaluation sheets cannot be located for associated documentation are provided PCP-PMOF-P-50X-LOC and EDG- in the Human Reliability Analysis notebook PMOE-PORT-PUMP, and AFW-AVOA- NB-PSA-HR.

CV-2010-D. SWS-AVOA-CV-0823-26 Page 47 of 57

ATTACHMENT 2 PNP 2018*025* Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model F&O Disposition ., Jm.*porta". .c.eJo

  • IS~~~s. ~~r~)cable .. 'Find~:g'Obse~ation .._____

Application and SWS-AV08-CV-082447M still need to be modified for fire related conditions.

This can be addressed by developing an associated HEP considering partial instrumentation and modifying the logic between the instrumentation and HEPs to properly reflect the dependence on instrumentation. Alternatively, this impact could be addressed in post-processing.

Complete the definition and evaluation of the fire re~onse actions.

HRA-C1-01 Open HRA-C1 Fire response HFEs modeled with The procedures, modification detail, Additional update is (Unresolved) screening values have not yet been operations review, and detailed HRA model needed to close this evaluated in a manner accounting for development are not yet complete. finding. For those relevant PSFs (e.g., ACP-DGOT-8S8- Significant HFEs were evaluated and STls on which DG, FPS-PMOE-START-L, ACP-PMOE- developed in further detail as documented in detailed HRA is 383-11A, ACP-PMOE-383-12A, etc.). the Human Reliability Analysis notebook N8- determined to have Also, HRA Calculator evaluation sheet PSA-HR. Screening values are still a potential impact, cannot be located for PCP-PMOF-P-50X- applied for fire HEPs pending development the effect is LOC and EDG-PMOE-PORT-PUMP, and of final procedures, modifications, and expected to be AFW-AVOA-CV-2010-D, SWS-AVOA-CV- operations reviews. assessed in the 0823-26, and SWS-AV08-CV-082447M change evaluations still need to be modified for fire related for the affected STls.

conditions. This task is not completed.

Complete detailed assessment of HFEs (for fire response actions) associated with risk significant fire scenarios.

HRA-D1-01 Open HRA-D1 Identification and evaluation of recovery The final developed fire HEPs incorporate This finding was (Resolved) actions for risk significant scenarios are task complexity and procedural guidance as resolved and expected to continue as the refinement of documented in the Post-Initiator Operator therefore is not fire scenario analysis continues. Action Questionnaire (P-IOAQ) provided to expected to impact, Currently, some of the top core damage current SRO licensed on-shift Operations STI change fire scenarios still do not account for Department personnel and Training evaluations realistic recovery actions. This task is not Department personnel for use in validating performed in completed yet. HEP information accuracy. accordance with the SFCP.

Com~ete identification and evaluation of Significant HEPs were evaluated and Page 48 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model Status . Applicable Finding/()bservation Disposition .** Irnf.)O~riceto . ****.****

I>~~? . * .*' .* *.'**i*"*..*.<

SR(sr> '>. ....... .. . .

recovery actions for risk significant developed in further detail. Screening values

. ......... Application ."" . . . . .

scenarios as the refinement of fire are still applied for some HEPs, however, the scenario analysis continues. values selected for screening are considered conservative. The final list of fire HEPs and their associated documentation are provided in the Human Reliability Analysis notebook NB-PSA-HR.

HRA-D2-01 Open HRA-D2 Many of the operator recovery actions The procedures, modification detail, Additional update is (Unresolved) associated with fire response are still operations review, and detailed HRA model needed to close this modeled with screening values; i.e., not development are not yet complete. finding. For those accounting for all of the relevant PSFs. Screening values are still applied for fire STls on which Dependency analysis has been HEPs. detailed HRA is performed for the current set of fire determined to have scenarios and operator actions in the 'T' A dependency analysis was completed to a potential impact, model. The results generated from the "Q" identify combinations of human failure the effect is model did not incorporate the dependency events (HFEs) in which dependencies expected to be analysis. The dependency analysis needs between actions may contribute to an assessed in the to be re-analyzed before finalization of the increase in CDF when compared to the CDF change evaluations Fire PRA model. This task is not complete calculated when nominal screening values for the affected STls.

yet. Also, HRA Calculator evaluation for human error probabilities (HEPs) are sheets cannot be located for PCP-PMOF- used. The fire PRA HRA dependency P-50X-LOC and EDG-PMOE-PORT- analysis is documented in the Human PUMP, and AFW-AVOA-CV-2010-D, and Reliability Analysis notebook NB-PSA-HR.

SWS-AVOA-CV-0823-26, and SWS-AVOB-CV-082447M still need to be Dependencies between actions were modified for fire related conditions. assigned based on sequence-specific evaluations of cues, timing, location, and Complete the detailed assessments for available resources, and the HEPs adjusted recovery actions identified in risk if necessary to represent the level of significant scenario. dependence; the CDF was then recalculated using the modified HEPs. HFEs not explicitly evaluated for dependence were assigned HEPs of 1 (i.e., the represented operator actions are assumed to fail with a probability of unity), and thus the resulting CDF represents an upper bound for the potential impact of dependencies upon the results. Shared cues conservatively assumed 100% dependence.

Page 49 of 57

ATTACHMENT 2 PNP 2018*025

Open Finding F&Os Against the PNPS Fire PRA Model This approach has identified important HFEs for which the completion of detailed human reliability analyses may be beneficial; those anal ses have not been completed.

HRA-E1-01 Open HRA-E1 Documentation for HFEs associated with The procedures, modification detail, Additional update is (Unresolved) selected fire response HFEs (e.g., FPS- operations review, and detailed HRA model needed to close this PMOE-START-L, ACP-PMOE-383-11A, development are not yet complete. finding. For those ACP-PMOE-383-12A, etc.) in the risk Screening values are still applied for fire STls on which significant fire scenarios need to be HEPs pending development of detailed HRA is provided. Also, HRA Calculator evaluation final procedures, modifications, and determined to have sheets cannot be located for PCP-PMOF- operations reviews. a potential impact, P-50X-LOC, EDG-PMOE-PORT-PUMP, the effect is and PULLFUSE; AFW-PMOT-P-8B-LOC expected to be seems to have been changed to AFW- assessed in the PMOT-P-8B-SBO in HRA notebook (but change evaluations not changed in Fire PRA model); and for the affected STls.

AFW-AVOA-CV-2010-D, SWS-AVOA-CV-0823-26, and SWS-AVOB-CV-082447M still need to be modified for fire related conditions. This task is not complete.

Complete the detailed assessments and documentation of HFEs associated with fire response HFEs in the risk significant fire scenarios.

SF-A1-01 Open SF-A1 The current seismic fire interactions The Seismic-Fire Interaction Report 0247- Additional update is (Unresolved) analysis relies on the IPEEE study. The 07-0005.05 evaluates Palisades needed to close this report needs to demonstrate that the with respect to NUREG/CR-6850 Task 13, finding. Qualitative scope of that work meets the objectives of Seismic-Fire Interactions Assessment. analysis of seismic-the Standard and that plant changes fire interactions is since the work was performed do not The seismic fire interactions analysis has not expected to have no compromise the conclusions. been updated. However, since the Standard impact on the fire only requires a qualitative analysis, there is PRA results or the The report should clarify the scope of the no impact on the quantified results in fire STI change IPEEE review and demonstrate it is PRA model. evaluations adequate to fully address the issues performed in identified in the SRs and lor supplement accordance with the the work as necessary. Furthermore the SFCP.

Page 50 of 57

ATTACHMENT 2 PNP 2018*025 . Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model

'~~~~;~~~';~~,~H0~!;~~J~,~~;~~~~0~i~i~l""

report needs to provide assurance that the conclusions of that work have not been compromised by plant hardware/procedural changes since the IPEEE work was performed.

FQ-A4-01 Open FQ-A4 Many of the accident sequences involve a Recovery actions and proposed This finding was (Resolved) Fire initiator which goes straight to core modifications have been incorporated into resolved and damage [i.e .. , there is no success path]. the final version of the fire PRA model and therefore is not This implies a single fire event can fail documentation. The final model has no expected to impact both trains of safe shutdown capability. sequences with a conditional core damage STlchange The CDF is too high to accept so many probability of 1. Results are described in evaluations individual sequences with no success Appendix B of the Fire Risk Quantification performed in path. There has not been sufficient and Summary Report 0247-07-0005.01. accordance with the investigation done to indicate whether SFCP.

recovery actions are truly not possible, or simply not modeled yet.

Comolete accident seauence modelin FQ-B1-01 Open FQ-B1 QU-B3 requires demonstration of The convergence process to determine This finding was (Resolved) acceptable truncation value by an iterative acceptable truncation limits for the final fire resolved and convergence process. The PLP fire PRA PRA model was documented in Section 6.0 therefore is not does not have this process. Although of the Fire Risk Quantification and Summary expected to impact there is no indication the current Report 0247-07-0005.01. STI change truncation value is not acceptable, the evaluations convergence process exercise was not performed in done. accordance with the SFCP.

Provide a demonstration of iterative truncation FQ-C1-01 Open FQ-C1 PRA document NB-PSA-HR-1, Rev 3 The procedures, modification detail, Additional update is (Unresolved) provides an HEP dependency analysis operations review, and detailed HRA model needed to close this and develops adjustment factors to apply development are not yet complete. finding. For those to the cutsets. Multiple HFE's are STls on which evaluated for dependencies using the A dependency analysis was completed to detailed HRA is EPRI HRA calculator. Dependency identify combinations of human failure determined to have adjustment factors are developed and events (HFEs) in which dependencies a potential impact, applied in the cutsets. However, the "Q" between actions may contribute to an the effect is model [which was reviewed] does not increase in CDF when compared to the CDF expected to be incoroorate this work. Therefore the F&O calculated when nominal screenina values assessed in the Page 51 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model and the not met assessment. for human error probabilities (HEPs) are change evaluations used. The fire PRA HRA dependency for the affected STls.

Complete the T model quantification analysis is documented in the Human Reliability Analysis notebook NB-PSA-HR.

Dependencies between actions were assigned based on sequence-specific evaluations of cues, timing, location, and available resources, and the HEPs adjusted if necessary to represent the level of dependence; the CDF was then recalculated using the modified HEPs. HFEs not explicitly evaluated for dependence were assigned HEPs of 1 (i.e., the represented operator actions are assumed to fail with a probability of unity), and thus the resulting CDF represents an upper bound for the potential impact of dependencies upon the results. Shared cues conservatively assumed 100% dependence.

This approach has identified important HFEs for which the completion of detailed human reliability analyses may be beneficial; those analvses have not been comoleted.

FQ-E1-01 Open FQ-E1 The discussion of dominant results is not Section 6.0 of the Fire Risk Quantification This finding was (Resolved) presented in the 0247-07-0005.01. The and Summary Report 0247-07-0005.01 resolved and results are categorized and sorted in was revised to include a discussion of the therefore is not terms of the dominant contributors [as per dominant results. expected to impact FQ-E1], but there is no discussion as STlchange required by this SR. evaluations performed in Revise the model to yield representative accordance with the results and then develop a discussion of SFCP.

dominant seauences.

UNC-A1-01 Open UNC-A2 Only a limited number of parameter and The Fire Risk Quantification and Summary (Resolved) QU-E1 modeling uncertainties and associated Report 0247-07-0005.01, was revised to QU-E2 assumptions have been identified. The list include additional discussion and evaluation QU-E4 is incomolete and not defined in sufficient of the state-of-knowledae correlation and the Page 52 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model iiFinding/pbserVation, , ,

"Li:i/r;.::,;,,": ;>~,;:,:::,:;/,::c');':' _":~ :,;

detail to support a reasonable impact of uncertainty associated with STlchange characterization or evaluation. severity factors and non-suppression evaluations Uncertainties have been propagated probability. The resulting distributions were performed in through a Monte Carlo approach. found to have greater range factors and thus accordance with the However, correlation of state of represent a more realistic analysis of the fire SFCP.

knowledge uncertainties has not been PRA uncertainty.

addressed, i.e. all initiators have been treated as independent variables, Severity Factor (SF) and Non Suppression Probabilities (NSP) and spurious actuation probabilities are not correlated.

(Uncertainties carried over from the internal events analysis are correlated).

This approach has led to unrealistically narrow predictions of CDF and LERF distributions (error factor of 2) and the potential underestimation of the mean values for scenarios which are quantified based on the product of like distributions (e.g. multiple spurious actuation probabilities).

Compile a comprehensive list of sources of model uncertainty, including related assumptions, and their potential impact on the Fire PRA model. Use NUREG/CR 6850 to identify generic issues and individual task analysis documents to identify plant specific issues. Develop and implement strategies for addressing each issue identified.

UNC-A2-01 Open UNC-A2 The uncertainty intervals assigned to Fire The approach for performing the parametric Additional update is (Unresolved) IGN-A1O IEs, Severity Factors and Non uncertainty evaluation has not yet been needed to close this IGN-B5 Suppression Probabilities are not based updated. finding. A parametric FSS-E3 on acceptable systematic methods. uncertainty analysis CF-A2 1) Uncertainty distributions for fire IEs The parametric uncertainty analysis is is expected to have DA-D3 have been assigned the same error factor presented in Section 7.1 of the Fire Risk no impact on the fire IE-C15 of 10 rather than using posterior Quantification and Summary Report 0247- PRA point estimate QU-A3 distributions from Bavesian update 07-0005.01. The issues identified have not results or the STI Page 53 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model

2) SF distributions have been assigned been fully addressed, but this primarily change evaluations without an underlying basis. impacts the potential range of the uncertainty performed in
3) NSP uncertainty distribution has been distribution and does not have a significant accordance with the derived on the basis of NUREG/CR 1278. impact on the mean value; and has no SFCP.

This provides guidance on HEP impact on the point estimate mean values uncertainty assessment. However, NSP used in the analysis.

terms are an output of a combination of fire growth and suppression modeling and guidance in NUREG/CR 1278 has therefore little relevance. A valid approach would be to address the uncertainties in damage times in combination with uncertainties in suppression probabilities based on specific contributing factors.

4) Uncertainties associated with spurious actuation probabilities have been characterized according to a set of rules defined for severity factors. In this case spurious actuation probabilities with a failure probability of> 0.25 are assigned an error factor of 1.0. In contrast NUREG/CR 6850 recommend use of a uniform distribution with the following limits:

Cables with 15 or less conductors: +20%

Cables with more than 15 conductors:

+50%

Alternatively the values included in tables 10-1 to 10-5 NUREG/CR 6850 could be used where limits appear to be wider. The Palisades analysis has not accounted for larger uncertainties associated with cables with> 15 conductors.

Identify sources of parametric uncertainty and estimate uncertainty intervals for significant ignition frequencies and fire growth modeling parameters using an method such as Page 54 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 4.

Open Finding F&Os Against the PNPS Fire PRA Model updating, frequentist method or expert

  • udqment.

MU-A1-01 Open MU-A1 The Palisades PRA Model Update Section 3.3 of the PSA Model Configuration This finding was (Resolved) procedure includes maintenance and Control Notebook NB-PSA-CC has resolved and upgrades to the PRA to be consistent with been revised to include a requirement for a therefore is not the as-built, as-operated plant. Resolution peer review against the ASME standard for expected to impact of the Full Power Internal Events (FPIE) PSA model upgrades. STlchange Peer Review F&Os and incorporation of evaluations design and operational information performed in relevant to a Fire PRA should result in accordance with the meetinq the Standard. SFCP.

Page 55 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 5.

List of SRs Assessed as ((-lor Not Reviewed for the PNPS Fire PRA Model S~s . ,{i l'0pic ...* **.* ***.** . i ** *.*****..*..*** *i *.******* *......... ............ ....

.Status

... ... .... . .* Ifl'Ip0r'tanceto,6.pplip~ti()I1.*. ****** i* *.*. * *.* *.**.**..*

PRM-A3 Construct the fire PRA plant response model so it SR was not reviewed. The sequence infrastructure to The structure of the plant response is capable of determining the significant provide this capability is not available at this time, model is not expected to impact STI contributors to fire-induced risk when quantified. although the risk importance in a single scenario can be change evaluations performed in (CC-I/IIIIII) calculated. accordance with the SFCP evaluations.

PRM-B2 Verify the peer review exceptions and deficiencies SR was not reviewed. The internal events F&O closure The impact of internal events PRA for the Internal Events PRA are dispositioned, and review documents the disposition of the internal events findings on STI change evaluations the disposition does not adversely affect the peer review exceptions and deficiencies. performed in accordance with the development of the fire PRA plant response SFCP evaluations is discussed in model. (CC-IIII/III) Table 2.

PRM-B4 Model any new initiating events identified in PRM- SR was not reviewed. This SR was not required as no For those STls on which new B2 in accordance with the SRs for HLR-IE-A, -B, new initiating events were identified. However, PRM-B3 initiating events are determined to and -C. Address the SRs in the context of a fire indicates spurious SI should be considered, so that the have a potential impact, the effect is inducing the initiating events excluding the SR must be completed. expected to be assessed in the initiating events that cannot be induced by a fire. change evaluations for the affected Develop a basis to support the non-applicability of STls.

any of the HLR-IE requirements. (CC-IIII/III)

PRM-B14 Identify any new accident progressions beyond SR was not reviewed. The LERF analysis did not For those STls on which new the onset of core damage that would be assess whether there were any LERF phenomena accident progressions are applicable to the fire PRA that were not addressed applicable to the fire PRA which were not included in determined to have a potential for LERF in the Internal Events PRA. (CC-IIII/III) the internal events PRA. The fire PRA peer reviewers impact, the effect is expected to be were not aware of any Fire PRA which looked for assessed in the change evaluations "beyond internal events" LERF J)henomena. for the affected STls.

HRA-A3 Finding HRA-A3-01, also see Table 4. SR was assessed as CC-1. This finding was resolved and The documentation for the identification and A simulator exercise was performed with current therefore is not expected to impact evaluation process as well as the detailed Palisades' license holders in which several scenarios STI change evaluations performed in justifications for the conclusion of no undesired were evaluated to determine how Operators would accordance with the SFCP.

operator actions need to be completed and respond given spurious or false instrument indications.

reviewed to confirm the evaluation conclusion. The results of these exercises were considered in the HEP development process. The process and evaluation results are documented in the Human Reliability Analysis Notebook NB-PSA-HR.

HRA-B4 Include HFEs for cases where fire-induced SR was assessed as CC-1. For those STls on which undesired instrumentation failure of any single instrument Finding ES-C2-01 also pertains to undesired operator operator actions due to instrument could cause an undesired operator action actions, see Table 4. failure is determined to have a consistent with HLR-ES-C and in accordance with Instrument failures are built into the fire PRA and potential impact, the effect is the SRs for HLR-HR-F. Develop a basis to directly impact the HRA. A simulator exercise was expected to be assessed in the support the non-applicability of the HLR-HR-F performed with current Palisades' license holders in change evaluations for the affected requirements. which several scenarios were evaluated to determine STls.

Page 56 of 57

ATTACHMENT 2 PNP 2018-025 - Probabilistic Risk Assessment Technical Adequacy Table 5.

List of SRs Assessed as ((-lor Not Reviewed for the PNPS Fire PRA Model

~Status
..;;;;; ;;;;,;il~~~~;~~~~~~~~8*!t~

how Operators would respond given spurious or false instrument indications.

The results of these exercises were considered in the HEP development process. The process and evaluation results are documented in the HRA Notebook NB-PSA-HR.

HRA-C1 Finding HRA-C1-01, also see Table 4. SR was assessed as CC-I. Additional update is needed to close HFEs in risk significant scenarios need to receive The procedures, modification detail, operations review, this finding. For those STls on which detailed assessment accounting for relevant and detailed HRA model development are not yet detailed HRA is determined to have PSFs. complete. a potential impact, the effect is Significant HFEs were evaluated and developed in expected to be assessed in the further detail as documented in the Human Reliability change evaluations for the affected Analysis notebook NB-PSA-HR. Screening values are STls.

still applied for fire HEPs pending development of final rocedures, modifications, and operations reviews.

Page 57 of 57

Retrieved from "https://kanterella.com/w/index.php?title=ML18270A323&oldid=2399165"