AEP-NRC-2025-03, Unit 2 - Response to Request for Additional Information (RAI) for License Amendment Request (LAR) for Application to Adopt 10 CFR 50.69, Risk-Informed Categorization and Treatment of Structures, Systems And.

Unit 2 - Response to Request for Additional Information (RAI) for License Amendment Request (LAR) for Application to Adopt 10 CFR 50.69, Risk-Informed Categorization and Treatment of Structures, Systems And.

ML25037A168
Person / Time
Site:	Cook
Issue date:	02/06/2025
From:	Dailey S Indiana Michigan Power Co
To:	Office of Nuclear Reactor Regulation, Document Control Desk
References
AEP-NRC-2025-03
Download: ML25037A168 (1)

Text

INDIANA MICHIGAN POWER" An /UP Company BOUNDLESS ENERGY-February 6, 2025 Docket Nos.: 50-315 50-316 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D.C. 20555-0001 Donald C. Cook Nuclear Plant Unit 1 and Unit 2 Indiana Michigan Power Cook Nuclear Plant One Cook Place Bridgman, Ml 49106 indianamichiganpower.com AEP-NRC-2025-03 10 CFR 50.90 Response to Request for Additional Information (RAI) for License Amendment Request (LAR) for Application to Adopt 10 CFR 50.69, "Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors"

References:

1. Letter from K. J. Ferneau, Indiana Michigan Power Company (l&M), to U.S. Nuclear Regulatory Commission (NRC), "Donald C. Cook Nuclear Plant Units 1 and 2 Application to Adopt 1 0 CFR 50.69, "Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors"," dated March 6, 2024, Agencywide Documents Access and Management System (ADAMS) Accession No. ML24073A234.

2. Email from S. P. Wall, NRC, to M. K. Scarpello, l&M, "Final RAI - D.C. Cook 1 & 2 - License Amendment Request Regarding Adoption of 10 CFR 50.69 (EPID No. L-2024-LLA-0025),"

dated December 30, 2024, ADAMS Accession No. ML24366A003.

This letter provides Indiana Michigan Power Company's (l&M), licensee for Donald C. Cook Nuclear Plant (CNP) Unit 1 and Unit 2, response to Reference 2 by the U.S. Nuclear Regulatory Commission (NRC) regarding Reference 1.

U.S Nuclear Regulatory Commission Page2 AEP-NRC-2025-03 to this letter provides an affirmation statement. Enclosure 2 to this letter provides l&M's response to Reference 2.

The changes proposed in this letter do not impact the conclusions provided in Reference 1 that a finding of "no significant hazards consideration" is justified. There are no new regulatory commitments made in this letter. Should you have any questions, please contact Mr. Michael K. Scarpello, Regulatory Affairs Director, at (269) 466-2649.

JMT/sjh

Enclosures:

1. Affirmation

2. Response to Request for Additional Information (RAI) for License Amendment Request Regarding Application to Adopt 10 CFR 50.69, "Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors."

c:

EGLE - RMD/RPS J. B. Giessner - NRC Region Ill NRC Resident Inspector N. Quilico - MPSC R. M. Sistevaris - AEP Ft. Wayne, w/o enclosures S. P. Wall - NRC Washington, D.C.

A J. Williamson - AEP Ft. Wayne, w/o enclosures to AEP-NRC-2025-03 AFFIRMATION I, Scott A. Dailey, being duly sworn, state that I am the Site Vice President of Indiana Michigan Power Company (l&M), that I am authorized to sign and file this request with the U.S. Nuclear Regulatory Commission on behalf of l&M, and that the statements made and the matters set forth herein pertaining to l&M are true and correct to the best of my knowledge, information, and belief.

Scott A. Dailey Site Vice President SWORN TO AND SUBSCRIBED BEFORE ME THIS l,~ AY OF s;t,:uoc\\a 2025 My Commission Expires s h~31 ao3, 0 to AEP-NRC-2025-03 Response to Request for Additional Information (RAI) for License Amendment Request Regarding Application to Adopt 10 CFR 50.69, "Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors."

By letter dated March 6, 2024 (Reference 1 ), Indiana Michigan Power Company (l&M, the licensee) submitted a license amendment request (LAR) for the Donald C. Cook Nuclear Plant (CNP), Unit Nos. 1 and 2. The proposed LAR would adopt 10 CFR 50.69, "Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors."

The provisions of 10 CFR 50.69 allow licensees to use an integrated, systematic, risk-informed process for categorizing structures, systems, and components (SSCs) according to their safety significance. A licensee that has adopted 10 CFR 50.69 may specify alternative treatments for SSCs that have low safety significance.

By email dated December 30, 2024 (Reference 2), the U.S. Nuclear Regulatory Commission (NRC) staff determined that the following information is needed to complete its review.

Probabilistic Risk Assessment Licensing Branch A (APLA) Questions "Regulatory Guide (RG) 1.201, Revision 1, "Guidelines for Categorizing Structures, Systems, and Components in Nuclear Power Plants According to their Safety Significance" (ML061090627),

endorses, with regulatory positions and clarifications, the Nuclear Energy Institute (NE/) guidance document NE/ 00-04, Revision 0, "10 CFR 50.69 SSC [Structure, System, and Component]

Categorization Guideline" (ML052910035), as one acceptable method for use in complying with the requirements in 10 CFR 50. 69. Section 3. 1. 1 of the LAR dated March 6, 2024, states that J&M will implement the risk categorization process of 10 CFR 50.69 in accordance with NE/ 00-04, Revision 0, as endorsed by RG 1.201.

The following questions and topics for discussion are intended to help the NRG staff determine if the licensee has implemented the guidance appropriately in NE/ 00-04, as endorsed by RG 1.201, as a means to demonstrate compliance with all of the requirements in 10 CFR 50. 69, including acceptability of the PRA models."

APLA-RAl-1 (Audit Question 2) - Peer Review Facts and Observations (F&Os) Findings "In 10 CFR 50.69(c)(1)(i) and (ii), the regulations require that a licensee's Probabilistic Risk Assessment (PRA) be "of sufficient quality and level of detail to support the SSC categorization process" and "all aspects of the integrated, systematic process used to characterize SSC importance must reasonably reflect the current plant configuration and operating practices, and applicable plant and industry operational experience." Revision 2 of RG 1.200, An Approach for Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk-Informed Activities" (ML090410014), and Revision 3 of RG 1.200, Acceptability of Probabilistic Risk Assessment Results for Risk-Informed Activities" (ML202388871) both describe acceptable approaches for determining the acceptability of a base PRA used in regulatory decision-making for commercial light-water nuclear power plants. Both revisions of RG 1.200 endorse, with clarifications, the American Society of Mechanical Engineers (ASME) I American Nuclear Society (ANS) PRA Standard ASMEIANS RA-Sa-2009, "Standard for Level 1/Large Early Release to AEP-NRC-2025-03 Page2 Frequency Probabilistic Risk Assessment for Nuclear Power Plant Applications. " Section 3. 3 of the LAR states that the PRA models have been assessed against RG 1.200, Revision 2.

As discussed in RG 1.200, it is recognized that a PRA may not satisfy each technical requirement to the same degree (i.e., capability category as used in the ASME/ANS PRA standard); that is, the capability category achieved for the different technical requirements may vary. This variation can range from (1) the minimum needed to meet the characteristics and attributes for each technical element to (2) the minimum to meet current good practice (i.e., state-of-practice) for each technical element. Further, which capability category is needed to be met for each supporting requirement is dependent on the specific application. In general, the NRC staff considers Capability Category II (CC-II) of the ASME/ANS PRA standard to provide a level of detail that is acceptable for the majority of applications. However, for some applications, Capability Category I (CC-I) may be acceptable for some supporting requirements."

APLA-RAl-1a (Audit Question 2)- Peer Review Facts and Observations (F&Os) Findings

[Capability Category} Please confirm for all peer reviews and independent closure reviews of facts and observations (F&Os) for the Full Power Internal Events PRA and internal floods PRA, that:

(1) all supporting requirements (SRs) met at CC-I have been identified in the table in Attachment 3 of Enclosure 2 of the LAR; and (2) that no SRs were determined to be "Not met. "Alternatively, provide a list of all SRs that are CC-/ and "Not met" and a disposition for why this is acceptable for the application (i.e., risk-informed categorization of SSCs in accordance with 10 CFR 50.69).

l&M Response to APLA-RAl-1a Full Power Internal Event (FPIE) peer reviews and closure reviews were reviewed to confirm the population of SRs that are currently "Not Met" or CC-I. This population is the following:

1. SY-A4 was determined to still be Met at CC-I due to only partial closure of F&O 4-4, described in Attachment 3 of Enclosure 2 of the LAR.

2. IFSN-A17 was not specifically dispositioned as Met within the FPIE closure review, however all F&Os against the SR have been Closed, and all secondary SRs for the associated F&Os were found to be Met. For this reason, l&M judges this SR to be Met.

3. A population of LERF SRs (LE-C1, -C2, -C3, -C4, -C5, -C9, -C10, -C11, -C12, -C13, -D5, and -E2.) These are further discussed in l&M's response to APLA-RAl-1c.

APLA-RAl-1b (Audit Question 2) - Peer Review Facts and Observations (F&Os) Findings

[F&O 4-4] In order to support risk-informed categorization, PRA models need to reflect the as-built, as-operated plant. In Attachment 3 of Enclosure 2 of the LAR, the disposition for F&O 4-4 states, in part: "Some walkdowns and interviews have been performed and did not identify any necessary modeling changes, the same outcome is expected for those systems that still need walkdowns and interviews performed." Please identify which systems still need walkdowns/interviews. Provide additional justification as to why these walkdowns/interviews are not expected to result in changes to the PRA model that could impact the application (i.e., risk-informed categorization of SSCs in accordance with 10 CFR 50.69). Alternatively, propose a mechanism to ensure that the applicable walkdowns/interviews are performed prior to implementation of 10 CFR 50. 69.

to AEP-NRC-2025-03 Page 3 l&M Response to APLA-RAl-1 b The systems remaining for updated walkdowns are AC power, accumulators, Auxiliary Feedwater (AFW), Control Room Instrumentation Distribution (CRID), Containment Spray, DC power, Distributed Ignition System (DIS), Engineered Safety Feature Actuation System (ESFAS), Main Steam (MS), Pressurizer Power Operated Relief Valve (PORV), Residual Heat Removal (RHR),

and Safety Injection (SI). These notebooks still reference walkdowns that were performed for original PRA development that need to be re-performed to address F&O 4-4. However, since the original reviews, l&M has had a process for identifying system/operating changes such that changes to system design/operation would have been captured and incorporated into the PRA.

For this reason, it would be unexpected to identify a deviation between the current model and the current plant. l&M's intent is to complete and document these walkdowns/interviews prior to the next FPIE model update.

APLA-RAl-1c (Audit Question 2) - Peer Review Facts and Observations (F&Os) Findings

[F&O 2-19 and FQ-D1-02] In Attachment 3 of Enclosure 2 of the LAR, the F&O Description for F&O 2-19 states that the Large Early Release Frequency (LERF) analysis uses conservative assessments, and these portions do not meet CC-I/. One example of conservatism is the assumption that a steam generator tube rupture (SGTR) is a containment bypass event without considering success of secondary side isolation. The associated disposition states that pressure and temperature induced SGTR events are modeled as progressing directly to LERF and that this ensures an over-estimation. The discussion of Fire PRA F&O FQ-D1-02 also refers to the disposition of F&O 2-19.

As discussed in NUREG-1855, Rev. 1, "Guidance on the Treatment of Uncertainties Associated with PRAs in Risk-Informed Decision Making" (ML17062A466), Section 7.2.3.3, the use of conservative assumptions in one part of the model can mask the significance of another part of the model (i.e., the importance measures of other modeled SSCs) and the impact of this conservative bias should be assessed. The dispositions provided for F&Os 2-19 and FQ-D1-02 are not specific to the impact on risk-informed categorization and do not address the possibility of masking risk insights due to conservative bias that may lead to the inappropriate categorization of SSCs. The guidance in NE/ 00-04 also recommends the use of sensitivity studies identified in the characterization of PRA adequacy, if applicable. Considering these observations, please provide:

1. Qualitative or quantitative justification (e.g., sensitivity studies) that supports the determination that conservative modeling assumptions that impact LERF results (e.g., not modeling secondary side isolation for a SGTR) will not adversely impact the categorization of SSCs per the 10 CFR 50.69 application and is not a key assumption/source of uncertainty for the application. If determined to be a key assumption/source of uncertainty, please provide updates to Attachment 6 and Section 3. 2. 7 of the LAR, as needed.

OR

2. Propose a mechanism to ensure that the issues identified in the F&Os are resolved prior to implementation of 10 CFR 50.69. For example, the model could be updated, and an independent F&O closure assessment could be performed. (Note: If any proposed PRA model changes constitute an upgrade to the PRA model as defined in the PRA Standard to AEP-NRC-2025-03 Page 4 ASMEIANS RA-Sa-2009, a focused-scope peer review would be required which could result in additional F&Os that need to be addressed prior to implementing the SSC categorization process.)

l&M Response to APLA-RAl-1c Facts and Observations (F&Os) 2-19 and FQ-D1-02 for the FPIE model and Fire PRA (FPRA) models assess a number of Large Early Release (LER) Supporting Requirements (SR) at Capability Category I (CC-I). These SRs were judged to be acceptable for general risk-informed applications because of the expected small impact on the calculated Large Early Release Frequency (LERF). This response will focus on the impact of this assumption on risk informed categorization. Specifically, that this assumption will not adversely impact the categorization of Systems, Structures, and Components (SSCs) for the 10 CFR 50.69 application and would not be a key assumption/source of uncertainty for the application.

Since these F&Os impact multiple LER SRs, the impact of each SR assessed at CC-I will be discussed individually.

1.

LE-C1 - This SR was assessed at CC-I during the 2015 full scope peer review, which remains its current status due to the subject F&Os remaining open. Subsequently, l&M performed a plant specific analysis for containment hydrogen combustion to obtain more realistic values for the likelihood of early containment failure due to hydrogen combustion.

This analysis was considered a PRA upgrade and has been subject to a peer review and closure review with all related F&Os except 2-4 closed. F&O 2-4 is discussed specifically in the response to APLA-RAl-1d. This focused scope review concluded that LE-C1 was Met at CC-II since this analysis explicitly considered containment structural capability.

This focused scope review was limited to the analysis of containment hydrogen and thus this SR is still considered Met at CC-I for other containment structural challenges unrelated to hydrogen combustion. However, these types of failures are not risk significant to the FPIE or FPRA models, as discussed below.

For FPIE, the dominant failure of containment is a bypass due to inadequate containment heat removal during medium Loss of Coolant Accident (LOCA). Because this end state includes a failure of both Containment Spray (CTS) and cooldown via the Steam Generators (SGs), this end state is not considered safe and stable as containment will continue to heat up post core damage. Reducing the risk significance of this assumption would primarily reduce the risk contribution of CTS and Auxiliary Feedwater (AFW) and would not clearly increase the risk of other SSCs. Uncertainty exists as to the time of a potential general emergency declaration, and therefore it is unclear if crediting additional time to containment ultimate failure pressure would result in these releases not being considered early.

Non-hydrogen combustion related containment failure events such as steam explosions and direct containment heating use values from NUREG-6595 which do not consider the containment ultimate failure capacity. However, these events are approximately 15% of FPIE LERF, and are phenomenological in nature such that reducing their risk would not significantly alter the risk profile. Furthermore, FPIE LERF is the lowest of the three hazard models which reduces the impact of this assumption. For these reasons, this assumption to AEP-NRC-2025-03 Page 5 is not expected to have an adverse effect on SSC categorization and would not be considered a key assumption or source of uncertainty for the FPIE Model.

For the FPRA model, the significant containment challenges are failure of containment isolation and early containment failure due to hydrogen combustion. As the FPRA model uses an identical containment model to the FPIE model (spurious valve operation is included at the fault tree level), the discussion of the CC-II hydrogen combustion model applies to the FPRA as well. Non-hydrogen related containment failure events contribute slightly less in the FPRA (approximately 11%), and as with the FPIE model, these are phenomenological events that occur independent of most system failures - the highest contributing end states are without AFW and proceed according to the Reactor Coolant System (RCS) pressure state at the time of core damage. Intentional and unintentional RCS depressurization are credited in these sequences. For these reasons, this assumption is not expected to have an adverse effect on SSC categorization and would not be considered a key assumption or source of uncertainty for the FPRA Model.

2.

LE-C2 - This SR is concerned with a realistic treatment of feasible operator actions post core damage. The LERF notebook, which as previously discussed, applies the same model to the FPIE and FPRA models, discusses the three credited operator actions in the containment event tree model. Each is discussed separately below.

a. The first action credited is an operator action to intentionally depressurize the RCS early (i.e. prior to core damage). The failure rate for this action is 0.1, which is taken directly from WCAP-16341-P, Simplified Level 2 Modeling Guidelines. This analysis was developed based on an assessment of the Westinghouse Owner's Group (WOG)

Emergency Operating Procedures (EOPs) and the Severe Accident Management Guidelines (SAMGs). This value was determined based on the short time frame between the core exit thermocouple reading of 1200°F, which would cue the action, and the time of core damage. Based on the short available timeframe, it is judged to be highly unlikely a more thorough assessment of this Human Error Probability (HEP) would result in a significant change to this value. For this reason, the value of this HEP is not expected to have an adverse effect on SSC categorization and would not be considered a key assumption or source of uncertainty for the FPIE or FPRA Model.

b. The second credited action is the operators intentionally bumping the RCPs during a non-Station Blackout (SBO) core uncovery scenario with feedwater unavailable and clearing the loop seal, raising the possibility of a Thermally Induced Steam Generator Tube Rupture (TI-SGTR). The l&M analysis uses the estimated value from WCAP-16341-P of 0.05. TI-SGTR is not a significant risk contributor in the FPIE model (<1%

of total LERF), and thus would not adversely affect the SSC categorization or be a key assumption for this model. For the FPRA, TI-SGTR is more significant, contributing approximately 10% of total LERF. This is primarily due to fire scenarios impacting Auxiliary Feedwater (AFW) availability, and not due to the operator action. While a more thorough assessment of this HEP could provide some small improvements to this value, this would not significantly alter the FPRA LERF risk profile. For this reason, the value of this HEP is not expected to have an adverse effect on SSC categorization and would not be considered a key assumption or source of uncertainty for the FPRA Model.

to AEP-NRC-2025-03 Page 6

c. The third action credited is an operator action to isolate containment is included in the fault tree and developed using the same plant-specific modeling tools used in the Level 1/CDF HRA development. This development is considered sufficiently realistic for SSC categorization as the other Level 1/CDF actions.

No additional operator actions were identified for the l&M analysis that would impact the LERF analysis. The parameters that impact containment challenges in the model are mainly RCS Pressure at the time of core damage and the availability of various systems.

RCS depressurization is discussed above, and the system models used in the containment event tree model are developed using the same modeling methods as the Level 1 /CDF model ( or are identical, such as AFW). For these reasons, the current actions are judged to be sufficient and would not have an adverse effect on SSC categorization and would not be considered a key assumption or source of uncertainty for the FPIE or FPRA Model.

3.

LE-C3 - This SR is concerned with equipment repair post-core damage. For systems such as AFW that can impact the containment event tree, repair cannot be credited as there is no acceptable industry method available to assess the likelihood of repair. Additionally, repair of equipment failures in the short time period (that are not due to start failures for which manual pump starts are credited) are highly unlikely and thus would not significantly impact the risk profile.

The SR also mentions recovery of AC power, which is credited in the Level 1 /CDF model for FPIE only. While some additional time does exist post core damage to restore systems and potentially reduce release consequences, this timeframe is short and would not significantly alter the recovery probabilities. The FPRA does not credit AC power recovery, as the standard does not allow credit for repair of fire damaged components.

For these reasons, this assumption would not have an adverse effect on SSC categorization and would not be considered a key assumption or source of uncertainty for the FPIE or FPRA Model.

4.

LE-C4 is concerned with credit for scrubbing of radionuclide releases for significant accident progression sequences. For this assessment, it is necessary to define the meaning of a significant accident progression sequence; this definition is the same as the one used in the model development and the standard:

a. A significant accident progression sequence is an accident sequence in which the summed percentage is 95% or the individual percent is 1 % of the applicable hazard group.

Scrubbing cannot be credited for induced SGTR sequences, because the SGs must be dry for this event to occur. For SGTR induced core damage, in which a containment bypass may occur due to the ruptured SG tube, it is possible scrubbing could be credited.

For the FPIE model, the only significant accident progression sequence does not consider the availability of AFW to the ruptured SG. However, the contribution of this sequence is small (-2%) and thus potential scrubbing credit would have limited impact. The FPRA model does not include the SGTR initiator as it cannot be fire-induced.

to AEP-NRC-2025-03 Page 7 Scrubbing could potentially be credited for Interfacing Systems LOCAs (ISLOCAs). These events release radionuclides into an uncertain location within the Auxiliary Building, and thus a determination of break location would be necessary to credit scrubbing along with an assessment of potential water level. Due to the high degree of uncertainty, it is unclear how much risk reduction would be possible and any credit in these scenarios would be expected to be small. ISLOCA sequences contribute about 10% of FPIE LERF, and small reductions would not significantly alter the risk profile. Fire-induced ISLOCA is not a significant risk contributor.

Scrubbing for other non-bypass sequences due to the potential operation of containment spray could be considered. For failures caused by reactor vessel breach, the radionuclides would not be expected to remain in containment long enough for this to be effective (WCAP-16341 S 3.4). This includes hydrogen explosions for l&M, as the analysis assumes the reactor vessel breach will ignite the hydrogen which is considered realistic.

Some credit could be obtained for containment isolation failures, however the operation of containment spray during this condition is unlikely to result in significant risk improvement, as failures during the Level 1/CDF sequences often disable containment spray (e.g. support system or ECCS recirculation failures).

For these reasons, this assumption would not have an adverse effect on SSC categorization and would not be considered a key assumption or source of uncertainty for the FPIE or FPRA Model.

5.

LE-C5 is concerned with crediting realistic success criteria for significant accident progression sequences. For the FPIE model, these sequences come in the following groups:

a. Containment Bypass scenarios which are direct LERF events such as SGTR from the Level 1/CDF model. These sequences use the realistic success criteria from the Level 1 model.

b. Containment Isolation failures, which use the success criteria (>2" bypass) developed in the containment isolation notebook. This is a standard value for the size of a large release and is sufficient for this application.

c. Early containment failures dependent on RCS pressure, DIS status, and AFW status.

DIS and AFW use realistic success criteria developed in their respective system notebooks. Intentional RCS depressurization is already discussed in LE-C2 and the Pressurizer Power Operated Relief Valves (PORVs) use their Level 1 system model and success criteria which is realistic.

d. Pressure Induced SGTR (PI-SGTR) which is dependent on RCS pressure and the availability of AFW. These systems use their Level 1 system model and success criteria as discussed above.

For the FPRA, the insights are similar except containment bypass is not a significant contributor and TI-SGTR is a significant contribution. TI-SGTR depends on the status of RCS pressure and AFWwhich is realistic in the FPRA, as the FPRA uses the FPIE system models with fire induced failures injected into the fault tree.

to AEP-NRC-2025-03 Page 8 For these reasons, this assumption would not have an adverse effect on SSC categorization and would not be considered a key assumption or source of uncertainty for the FPIE or FPRA Model.

6.

LE-C9 and LE-C10 are related SRs in which LE-C9 requires justifying credit for equipment survivability and then LE-C10 requires reviewing significant accident progression sequences to determine if credit for survivability can reduce LERF.

For CNP, the in-containment systems that impact LERF are primarily the hydrogen igniters, which are qualified and credited in the adverse environment.

This is because hydrogen burns are a significant containment challenge for an ice condenser containment. Other systems are credited in the Level 1 analysis to reduce or control containment pressure, such as containment spray and/or cooldown using the SGs in large and medium LOCA sequences to prevent a bypass due to overpressure. Other in-containment systems such as the hydrogen recirculation fans, which are not needed to prevent hydrogen buildup, would not reduce LERF as they do not alter the accident progression sequence. For these reasons, this assumption would not have an adverse effect on SSC categorization and would not be considered a key assumption or source of uncertainty for the FPIE or FPRA Model.

7.

LE-C11 and LE-C12 are related to the previous two and require justification of any credit given for equipment survivability or human actions that could be impacted by containment failure (LE-C11) and then requires a review of significant accident progression sequences for potential credit (LE-C12). l&M does not currently credit any operator actions or equipment post containment failure for LERF, however, some limited modeling of late releases is done in accordance with WCAP-16341-P. This is considered a reasonable assumption because generally, existing equipment cannot arrest a large release after containment failure, and containment failure is typically assumed to occur at or shortly after the time of reactor vessel breach for early sequences. For these reasons, this assumption would not have an adverse effect on SSC categorization and would not be considered a key assumption or source of uncertainty for the FPIE or FPRA Model.

8.

LE-C13 is concerned with realistic treatment of containment bypass, including justification for any treatment of scrubbing. Scrubbing is discussed in detail in the discussion for LE-C4, which is applicable to this SR as well. There are several types of direct containment bypass events modeled in the l&M analysis, each of which is discussed individually below.

a. ISLOCA events lead to containment bypass because of the nature of the initiator.

Credit for scrubbing may lead to a more realistic treatment of this event, but credit is not taken for the reasons discussed in LE-C4 for either the FPIE or FPRA models.

b. SGTR sequences lead to a containment bypass when the flow from the faulted SG is not able to be terminated. Credit for scrubbing in SGTR is discussed in LE-C4 including the reasons why it is not credited. Secondary side isolation also could be credited, but as with scrubbing credit, the risk importance of the SGTR accident sequences for the FPIE model is small (~2% of LERF) and as with scrubbing, would not significantly alter the risk profile. The SGTR initiating event is not applicable to the FPRA.

to AEP-NRC-2025-03 Page 9

c. Medium and Large LOCA initiating events can lead to a containment bypass if containment heat removal is not maintained. This assumption is made because the ice condenser can only control pressure for a brief period of time for these events (unlike small LOCA), and because a hole in containment would eventually result in evaporation in the recirculation sump. It is possible an assessment could be made as to whether these releases would constitute an early release, but there is significant uncertainty as to the time of a general emergency declaration.

The risk significant accident progression sequences involving bypass are Medium LOCA sequences with a loss of ECCS injection or recirculation, which would result in rapid core damage. Containment spray is not credited for these sequences, however credit is not expected to have a significant impact as the Residual Heat Removal (RHR) pumps and support systems common to both systems would also fail containment spray recirculation.

Medium and Large LOCA events are not applicable to the FPRA.

For these reasons, this assumption would not have an adverse effect on SSC categorization and would not be considered a key assumption or source of uncertainty for the FPIE or FPRA Model.

9.

LE-D5 is concerned with the treatment of induced SGTR, e.g. TI-SGTR and PISGTR. The l&M containment event tree adopts the WCAP-16341-P methodology, which is considered a realistic analysis of induced tube rupture events. This analysis explicitly considers conditions such as the status of secondary side isolation, the pressure of the RCS and the SGs, and the availability of AFW. This analysis also estimates relief valve demands using a realistic assessment to determine the pressure status of each SG. While the 2015 peer review did not consider this to meet CC-II, this is the most up to date industry analysis available that is not a newly developed method for induced SGTR events and is considered acceptably realistic for SSC categorization.

For these reasons, this assumption would not have an adverse effect on SSC categorization and would not be considered a key assumption or source of uncertainty for the FPIE or FPRA Model.

10. LE-E2 is concerned with the use of realistic parameters for significant accident progression sequences. The l&M containment event tree makes use of NUREG-6595 for early containment failure events with DIS active (i.e. direct containment heating phenomena), which is considered conservative. Currently, no additional methods are available to assess these phenomena for ice condenser containments (WCAP-16341-P assess this for large dry PWR containments). As discussed in LE-C1, a more realistic value for this is unlikely to significantly change the risk profile and is judged to be acceptable for SSC categorization. WCAP 16341-P is used for TI-SGTR and PI-SGTR as discussed in LE-D5 and considered acceptably realistic for SSC categorization. For early containment failure events without DIS active, the direct containment heating phenomena are added to the containment hydrogen analysis that has been assessed as CC-II by a subsequent focused scope peer review and closure review. This analysis is discussed in detail in the discussion for LE-C1.

For these reasons, this assumption would not have an adverse effect on SSC categorization and would not be considered a key assumption or source of uncertainty for the FPIE or FPRA Model.

to AEP-NRC-2025-03 Page 10 APLA-RAl-1d (Audit Question 2) - Peer Review Facts and Observations (F&Os) Findings

[F&O 2-4] In Attachment 3 of Enclosure 2 of the LAR, the F&O Description for F&O 2-4 states that all containment failures caused by hydrogen combustion are assumed to contribute to LERF.

The associated disposition states that the current implementation is conservative and that some sequences could potentially be screened out based on containment failure location, but that improvements in modeling would not result in a significant improvement in the overall realism of the PRA results. It is unclear to the NRC staff how significant the contribution from hydrogen combustion scenarios is to LERF results.

Please provide qualitative or quantitative justification that supports the determination that conservative modeling assumptions related to containment failure due to hydrogen combustion will not adversely impact the categorization of SSCs per the 10 CFR 50. 69 application and is not a key assumption/source of uncertainty for the application. If determined to be a key assumption/source of uncertainty, please provide updates to Attachment 6 and Section 3.2. 7 of the LAR, as needed.

l&M Response to APLA-RAl-1 d F&O 2-4 identifies a conservatism in the containment hydrogen analysis where uncertainty in the exact failure location is addressed by assuming that a containment overpressure would occur in a location that would lead to a direct path to the atmosphere. l&M's containment ultimate failure analysis estimates the ultimate failure capacity for several containment locations, of which the lowest is the basemat diagonal tension failure mode. This location is underground, and thus might not be a direct release path. The next weakest locations are the equipment and personnel hatches, both of which would lead to a direct path to the atmosphere.

Since the exact location of containment failure is uncertain and would depend on the compartment(s) in which hydrogen combustion occurred, the basemat tension failure was deemed appropriate for use in the hydrogen analysis. Further complicating this assessment was the fact that reactor vessel breach is considered the most likely source of hydrogen ignition, and this could result in overpressure in a number of containment compartments. If the basemat did not fail or was not subjected to the amount of overpressure the equipment and personnel hatches did, the overpressure would result in a direct atmospheric release. For these reasons, and the fact that a hydrogen combustion overpressure could go beyond both the basemat and equipment/personnel hatch failure pressures, this conservatism is judged to be acceptable.

Changes to this assumption would only reduce the risk of the DIS and slightly increase the relative contribution of the containment bypass and isolation failure modes. For these reasons, this assumption would not have an adverse effect on SSC categorization and would not be considered a key assumption or source of uncertainty for the FPIE or FPRA Model.

to AEP-NRC-2025-03 Page 11 APLA-RAl-1e (Audit Question 2)- Peer Review Facts and Observations {F&Os) Findings Section 3.1.1 of the LAR notes that previous versions of the following PRA models were submitted and accepted by the NRG for the listed applications. Further in Section 3.2.1 of the LAR the licensee states in part, "[t]he CNP categorization process for the internal events and internal flooding hazards will use a technically acceptable and independently peer reviewed plant-specific PRA model." It is unclear to the NRG staff if the FPIE 2023-R0 PRA model to be used in the categorization process is the same base model as for Revision 15MORW R1. Please confirm that FPIE 2023-R0 is the same base model that was full-scope peer reviewed in July 2015 as stated in Section 3.3 of the LAR.

l&M Response to APLA-RAl-1e The 2023 FPIE Model of Record (FPIE 2023-R0) is largely the same base model as model version 15MORW R1. The 2015 version of the FPIE model served as the "starting point" for the current version of the FPIE model, with changes being made to the model regularly to ensure it is up to date regarding data, state-of-practice methodology, plant changes, etc. The majority of changes made to the model between version 15MORW R1 and FPIE 2023-R0 were of a scope determined to be model maintenance, and therefore did not have a substantive impact on overall model technical acceptability. Significant changes (Upgrades) requiring peer review were identified and discussed in Section 3.3 "PRA Peer Review Process Results" of the LAR.

APLA-RAl-2 {Audit Question 4) - Passive Categorization In Section 3.1.2 of the LAR, for the passive categorization process, the licensee states, in part:

"Component supports, if categorized, are assigned safety based upon one of the following approaches:...

A combination of restraints or supports such that the [Low Safety Significant (LSS)] piping and associated SSCs attached to the [high safety significance (HSS)] piping are included in scope up to a boundary point that encompasses at least two supports in each of three orthogonal directions [27, 28)."

It appears the references cited are not applicable to the LAR excerpt above. Please confirm References 27 and 28 and update appropriately.

l&M Response to APLA-RAl-2 Yes. The references for the bullet beginning with "A combination of restraints... " should be to:

Standard Review Plan for Review of License Renewal Applications for Nuclear Power Plants." NUREG-1800, Revision 2. U.S. Nuclear Regulatory Commission, Washington, D.C.,

December 2010.

Standard Review Plan for Review of Subsequent License Renewal Applications for Nuclear Power Plants." NUREG-2192. U.S. Nuclear Regulatory Commission, Washington, D.C.,

July 2017.

to AEP-NRC-2025-03 Page 12 The References in this section incorrectly pointed to Reference 27 and 28, when in fact the intention was to add References 47 and 48 to the list of References and direct the reader to them.

Therefore, the bullet in Section 3.1.2 of the LAR should be changed to read:

"A combination of restraints or supports such that the {Low Safety Significant (LSS)J piping and associated SSCs attached to the [high safety significance (HSS)J piping are included in scope up to a boundary point that encompasses at least two supports in each of three orthogonal directions [47, 48)."

And the following two References should be added to Section 6.0 References, of the LAR:

"47. "Standard Review Plan for Review of License Renewal Applications for Nuclear Power Plants." NUREG-1800, Revision 2. U.S. Nuclear Regulatory Commission, Washington, D.C., December 2010.

48. "Standard Review Plan for Review of Subsequent License Renewal Applications for Nuclear Power Plants." NUREG-2192.

U.S. Nuclear Regulatory Commission, Washington, D.C., July 2017."

APLA-RAl-3 {Audit Question 6)-Total Risk Considerations Revision 3 of RG 1. 17 4, '~n Approach for Using Probabilistic Risk Assessment in Risk-Informed Decisions on Plant-Specific Changes to the Licensing Basis" (ML17317A2560), provides risk acceptance guidelines in terms of change-in-risk in combination with total core damage frequency (GDF) and LERF. The guidance in NE/ 00-04 includes an overall risk sensitivity study for all the LSS components to assure that if the unreliability of the components was increased, the increase in risk would be small (i.e., meet the acceptance guidelines of RG 1.174).

The guidance in RG 1.174 and Section 6.4 of NUREG-1855, Revision 1 states in part, for a Capability Category II risk evaluation, that the mean values of the risk metrics (total and incremental values) should be compared against the risk acceptance guidelines in RG 1.174. The mean values referred to are the means of the probability distributions that result from the propagation of the uncertainties for the PRA input parameters and model uncertainties explicitly represented in the PRA models. In general, the point estimate GDF and LERF obtained by quantification of the cutset probabilities using mean values for each basic event probability does not produce a true mean of the CDFILERF.

The LAR does not state whether the total GDF and LERF values presented in Attachment 2 of of the LAR are mean values. The NRG staff notes there is a small margin between the RG 1. 17 4 LERF threshold of 1 E-05 per year for LERF and the results for Units 1 and 2 of 9. 91 E-06 and 9. 67E-06 per year, respectively, provided in the LAR, Attachment 2 of. Under certain circumstances, a formal propagation of uncertainty may not be required if it can be demonstrated that the state of knowledge correlation (SOKC) is unimportant (e.g., the risk results are well below the acceptance guidelines). Accordingly, the risk increase due to consideration of the SOKC could impact the conclusions of the NE/ 00-04 Section 8 overall sensitivity study results by increasing the total LERF values above 1 E-05 per year for CNP Units 1 and 2.

Considering the observations above, please address the following:

to AEP-NRC-2025-03 Page 13 APLA-RAl-3a {Audit Question 6) -Total Risk Considerations

a. Provide a summary of how the SOKC investigation was performed for the base CNP PRA models used to support the 50. 69 applications. In the summary, explain which fire PRA parameters are assumed to be correlated beyond the component failure modes frequencies. Include justification that consideration of the identified parameters is sufficient to estimate the impact of the SOKC on fire risk.

l&M Response to APLA-RAl-3a The CDF and LERF values presented in Attachment 2 of Enclosure 2 are indeed point estimate values.

The PRA model database is built to support parametric uncertainty analyses using uncertainty parameters for failure rates by the use of type codes in the CAFTA software and uncertainties on individual basic events that do not use type codes. This parametric uncertainty analysis reflects the state of knowledge correlation (SOKC) considerations.

Because the 10 CFR 50.69 program application is a "delta" type application (i.e. acceptability is based on the difference in risk calculated for the base model configuration and that calculated for a configuration in which equipment is unavailable), the impact of the SOKC uncertainty on 10 CFR 50.69 estimates is considered negligible and Point Estimate values are adequate to inform the difference between plant configurations (i.e. the delta risk between different plant configurations). While mean CDF and LERF values are different than the point estimate values for the base model, the mean CDF and LERF values are also different than the point estimate values reflective of an equipment unavailable plant configuration.

The FPRA uses the FPIE uncertainty parameters for basic events that are common to both models (e.g. component failures). For parameters that are specific to the FPRA the uncertainty parameters used in the FPRA are based on the current industry guidance. The parameters include:

Fire Ignition Frequencies - NUREG-2169 or updated guidance for specific bins (e.g.,

NUREG-2230 for electrical cabinets, NUREG-2178 for main control board, etc.)

Fire HRA -Assigned by the HRA calculator Circuit Failure Mode Likelihood Analysis Probabilities-NUREG/CR-7150 The fire scenario severity factors (e.g. pump fire split fractions) are point estimates and therefore use generic error factor estimates. The fire scenario non suppression probabilities use generic NUREG-2178 error factors because the calculated non-suppression probabilities include multiple fire modeling and manual suppression parameters.

As a sensitivity, total risk is calculated using the mean values calculated as part of the parametric uncertainty analysis.

to AEP-NRC-2025-03 Page 14 APLA-RAl-3b (Audit Question 6) -Total Risk Considerations

b. Demonstrate that for the NE/ 00-04 Section 8 overall sensitivity study results, CNP will be in conformance with the RG 1. 17 4 risk acceptance guidance after the Full Power Internal Events (including flooding) and Fire PRA models are updated to include the increase associated with SOKG (if needed).

l&M Response to APLA-RAl-3b FPIE The parametric uncertainty evaluation for the FPIE PRA model is performed using the UNCERT software. A Monte Carlo simulation was performed for both CDF and LERF using 15000 samples to calculate the mean risk metrics that reflect SOKC considerations. Table APLA-06-01 below summarizes the results of the parametric uncertainty evaluation performed for the base FPIE PRA model.

Table APLA-06-01-FPIE Parametric Means U1 U2 GDF 2.44E-05 2.45E-05 LERF 1.52E-06 1.48E-06 FPRA The parametric uncertainty evaluation for the FPRA PRA model is performed using the UNCERT software. A Monte Carlo simulation was performed for both CDF and LERF using 50,000 samples to calculate the mean risk metrics that reflect SOKC considerations. Table APLA-06-02 below summarizes the results of the parametric uncertainty evaluation performed for the base FPIE PRA model.

Table APLA-06-02-FPRA Parametric Means U1 U2 GDF 3.69E-05 4.15£.:-05 LERF 3.09E-06 2.46E-06 Total Sensitivity Using the FPIE and FPRA parametric means instead of the point estimates to calculate the total risk gives the following total results.

Table APLA-06 Totals using Parametric Means U1 U2 GDF 8.23E-05 8.l0E-05 LERF 9.92E-06 9.66E-06 to AEP-NRC-2025-03 1 0CFR 50.69 Assessment Page 15 As shown in Table APLA-06-03 the FPIE and FPRA parametric mean values for CDF and LERF when substituted for the point estimate values in Attachment 2 of Enclosure 2 remain below the 10 CFR 50.69 acceptability thresholds. Given that the risk metrics haven't changed significantly, the resulting importance measures would likely remain unchanged. Therefore, the SSC categorizations would likely remain unchanged if the parametric mean values were utilized as opposed to the point estimate values.

Probabilistic Risk Assessment Licensing Branch C (APLC) Questions APLC-RAl-1 (Audit Question 1)- Integrated PRA Hazards Model 10 CFR 50.69(b)(2)(i) requires a licensee voluntarily choosing to implement 50.69 to apply for a license amendment with "a description of the process for categorization of R/SC-1, RISC-2, RISC-3, and RISC-4 SSCs." Paragraph (c)(1)(ii) of 10 CFR 50.69 requires that the SSC functional importance be determined using an integrated, systematic process. NE/ 00-04, Section 5. 6, "Integral Assessment," discusses the need for an integrated computation using available importance measures. It further states that the "integrated importance measure essentially weighs the importance from each risk contributor (e.g., internal events, fire, seismic PRAs) by the fraction of the total core damage frequency (or large early release frequency) contributed by that contributor." The guidance provides formulas to compute the integrated Fussell-Vesely importance (FV) and integrated Risk Achievement Worth (RAW).

Based on the information provided in the LAR, it is not clear to the NRG staff how the licensee proposes to address the integration of importance measures across all hazards (i.e., internal events, fire, and seismic).

Please address the following:

APLC-RAl-1a (Audit Question 1)- Integrated PRA Hazards Model

a. Explain how the integration of importance measures across hazards for the 10 CFR 50. 69 categorization process will be performed.

l&M Response to APLC-RAl-1 a For components categorized as LSS from the internal events PRA and categorized as HSS from one of the other hazard PRAs, an integrated assessment is performed to determine an overall categorization for all hazards. The integrated assessment, across hazards, will be performed using the process described in NEI 00-04, Section 5.6. In order to provide an overall assessment of the risk significance of SSCs, an integrated computation is performed using the available importance measures from each hazard. This integrated importance measure essentially weights the SSC importance from each hazard (e.g., internal events, fire, seismic PRAs) by the fraction of the total core damage frequency contributed by that hazard. The following formulas from NEI 00-04 define how the Fussell-Vesely importance and Risk Achievement Worth are computed for CDF. The same process is used for LERF.

to AEP-NRC-2025-03 Integrated Fussell-Vesely Importance L *(FV:- *

• CDF-)

IFV:. =

J l,]

J 1

LjCD0

where, IFVi = Integrated F-V Importance of Component i over all CDF Contributors FVi,j = F-V Importance of Component i for CDF Contributor j CDFj = CDF of Contributor j Integrated Risk Achievement Worth Importance L *(RAW'.* * - 1)

• CDF,-

IRAW'.* = 1 +

1 l,J 1

1 LjCD0

where, IRAWi = Integrated Risk Achievement Worth of Component i over all CDF Contributors RAWi,j = Risk Achievement Worth of Component i for CDF Contributor j CDFj = CDF of Contributor j Page 16 Once calculated, an assessment is made of these integrated values against the screening criteria of F-V >0.005, RAW> 2.0 for independent basic events, and RAW> 20 for common cause basic events. In no case should the integrated importance become higher than the maximum of the individual measures. However, it is possible that the integral value could be significantly less than the highest contributor, if that contributor is small relative to the total CDF/LERF.

Per the guidance of NEI 00-04, if a component is categorized as LSS from the internal events PRA but categorized as HSS for one of the other hazard PRAs, and the integrated assessment also leads to a categorization of LSS, the component may be presented to the IDP as a potential categorization of LSS.

to AEP-NRC-2025-03 Page 17 APLC-RAl-1b (Audit Question 1) - Integrated PRA Hazards Model

b. Discuss how the importance measures for the PRA models (e.g., FVand RAW) are derived and justify why the importance measures generated do not deviate from the NE/ guidance or Table 3-1 of the LAR. If the practice or method used to generate the integrated importance measures is determined to deviate from the NE/ guidance, then provide justification to support why the integrated importance measures computed are appropriate for use in the categorization process.

l&M Response to APLC-RAl-1 b The component importance measures (F-V and RAW) are derived from basic event importance reports generated from the CDF or LERF cutsets. Events in the PRA that represent each SSC are selected consistent with NEI 00-04, Section 5.

PRA model basic events are mapped to system components. Each mapped PRA basic event is identified as either common cause or independent. Basic events related to human errors, testing activities, and planned maintenance activities to components are not mapped since those basic events are either not applicable to hardware (human error) or represent a voluntary removal of a component from service (testing and planned maintenance) and therefore do not represent the performance of an SSC. The exception is when a human error basic event implicitly models a component that is not explicitly modeled. This is consistent with NEI 00-04, Section 5.1, which states:

"The assessment of importance for an SSC involves the identification of PRA basic events that represent the SSC. This can include events that explicitly model the performance of an SSC (e.g., pump X fails to start), events that implicitly model an SSC (e.g., some human actions, initiating events, etc.) or a combination of both types of events."

For the purposes of performing the active component risk categorizations, only metrics associated with the performance of the mitigation capability of components are considered. Component failures due to the occurrence of the hazard are not included in the mapping of basic events to components. For example, a fire induced failure of the component is not included in the component-specific failure mapping. The basis for this is that the source of the component failure is the hazard itself and not representative of the performance of an SSC. This is consistent with NEI 00-04, Section 5.2, which states:

"The risk importance process is slightly modified to consider the fact that most fire PRAs do not have the ability to aggregate the mitigation importance of a component with the fire initiation contribution. For that reason, components are evaluated using standard importance measures for their mitigation capability only."

The same logic is also applied to generally exclude from consideration seismic failures with respect to fragilities where the seismic event induces the component failure. Similar to fire, these seismic-induced failures are not viewed as representative of SSC performance. This is consistent with NEI 00-04, Section 5.3, which states:

"The generalized safety significance process for plants with a seismic PRA is the same as the process for a fire PRA. The risk importance process is slightly modified to consider to AEP-NRC-2025-03 Page 18 that plant components cannot initiate seismic events. Aside from that small change, the process is the same as the internal events PRA process."

Components that have multiple applicable events (for example, a pump that has both a Fail to Start and a Fail to Run event) have these multiple events considered consistent with NEI 00-04, Section 5.1, specifically:

Sum of F-V for all basic events modeling the SSC of interest, including common cause events> 0.005 Maximum of component basic event RAW values> 2 Maximum of applicable common cause basic events RAW values> 20."

This approach for applying importance measures to the categorization process is consistent with NEI 00-04, and therefore is appropriate for use in l&M's 10 CFR 50.69 categorization process.

APLC-RAl-1c (Audit Question 1)- Integrated PRA Hazards Model

c. Describe how the importance measures for the seismic PRA (e.g., FV and RAW,, are derived considering that the seismic hazard is discretized into bins. The discussion should include how the same basic events, which were discretized by binning during the development of the seismic PRA, are then combined (i.e., combined across bins as well as across failure modes such as seismic and random failure modes) to develop representative importance measures.

Further, discuss how they are compared to the importance measure thresholds in NE/ 00-04.

Provide justification to support the determined impact on the categorization results and describe how the approach is consistent with the guidance in NE/ 00-04.

l&M Response to APLC-RAl-1 c Cutsets are quantified separately for CDF and LERF for each of the seismic initiator bins. This process produces a unique cutset file for each bin with the Minimum Cutset Upper Bound (MCUB) estimate. The next step in the process is to append each individual bin cutset file together, which produces a MCUB estimate for total seismic CDF or LERF. Because each bin uses a unique initiating event, it produces unique cutsets relative to the other bins. For this reason, the appended cutset file contains all cutsets from all bins and does not alter the cutset results for each bin during combination. This appended file is then post-processed with ACUBE software to correct for the overestimation of risk due to the rare event approximation in accordance with Seismic Code Case Supporting Requirement SPR-E2.

Basic event importances (F-V and RAW) are reported by ACUBE software for the resulting CDF and LERF. For random failures which are used in risk categorization, identical basic events are used in all seismic bins and thus the calculated importances are relative to-the total ACUBE software calculated Seismic CDF and LERF. This is consistent with the methodology used for internal events and Fire, and consistent with standard practice for risk applications. These basic event importances (F-V and RAW) are compared to the criteria from NEI 00-04 to determine whether a specific component is ranked as LSS or HSS. The same is done for LERF. Therefore, derivation of importance measures is consistent with NEI 00-04 guidance.

to AEP-NRC-2025-03 Page 19 APLC-RAl-1a (Audit Question 1) - Integrated PRA Hazards Model

d. In the context of the integral assessment described in NE/ 00-04, Section 5. 6, it is understood that importance evaluations performed in accordance with the process in NE/ 00-04 are determined on a component basis. However, the LAR and NE/ 00-04 guidance does not make clear how the integrated importance measures are calculated for certain components.

Specifically, in the seismic PRA, basic events that represent different failure modes for a component may not align with basic events in other PRA models. Examples of such basic events include those that are specific to the seismic PRA (including implicitly modeled components) or basic events that represent a subcomponent modeled within the boundary of a component in the internal events PRA.

l&M Response to APLC-RAl-1 d All basic events associated with a particular system will be mapped to a system component, such that any basic events that appear only in a fire PRA or only in a seismic PRA will be mapped to one of the system components and any basic event importances will be included in the categorization of the component in both the initial categorization and in any integrated assessment.

APLC-RAl-2 (Audit Question 2) - 2013 PRA Standard 10 CFR 50.69(c)(1)(i) requires the PRA to be "subjected to a peer review process assessed against a standard or set of acceptance criteria that is endorsed by the NRG."

In Section 3.3 of the LAR, an excerpt from PWROG-18062-P refers to an "effective" endorsement of Part 5 of Addendum B of the ASMEIANS RA-SB-2013 PRA standard for seismic modeling.

The NRG staff accepted ASMEIANS RA-S Case 1, "Case for ASMEIANS RA-Sb-2013 Standard for Level 1/Large Early Release Frequency Probabilistic Risk Assessment of Nuclear Power Plant Applications," dated November 22, 2017, for use in licensing actions on an interim basis in Jetter dated March 12, 2018 (ML18017A964). Further, in Revision 3 of RG 1.200, Appendix B endorses (with clarifications and qualifications) ASMEIANS RA-S Case 1, "Case for ASMEIANS RA-Sb-2013 Standard for Level 1/Large Early Release Frequency Probabilistic Risk Assessment of Nuclear Power Plant Applications," dated June 2020.

APLC-RAl-2a (Audit Question 2) - 2013 PRA Standard

a. Given available regulatory basis for applicability of Code Case 1 for licensing actions, explain the purpose of CNP's statements about "effective" endorsement from a different context.

l&M Response to APLC-RAl-2a This statement was taken from the Seismic PRA (SPRA) peer review, which at the time was intended to contextualize the status of the SPRA standard being used for the review. Explicit endorsement of the code case is provided in RG 1.200, Revision 3. This provides the current basis for acceptability of the SPRA peer review and closure.

to AEP-NRC-2025-03 Page 20 APLC-RAl-2b (Audit Question 2) - 2013 PRA Standard

b. Confirm that the seismic PRA supporting the LAR addresses the staff's clarifications and qualifications on Code Case 1 as provided in the enclosure to the letter dated March 12, 2018 (ML18017A966) and included in Revision 3 of RG 1.200.

l&M Response to APLC-RAl-2b The SPRA peer review (PWROG-18062-P) references the clarifications and qualifications provided in ML18017A966. This is explicitly stated as included in the review section 1.1, "Purpose":

"This peer review was conducted against the requirements of the Code Case for ASM E/ ANS RA-Sb-2013 (Reference 1 ), as amended by the Nuclear Regulatory Commission (NRC) on March 12, 2018 (Reference 2)."

1. ASME/ANS RA-S CASE 1, Case for ASME/ANS RA-Sb-2013, "Standard for Level 1 /Large Early Release Frequency Probabilistic Risk Assessment for Nuclear Power Plant Applications", ASME and ANS, November 2017. NEI 12-13, "External Hazards PRA Peer Review Process Guidelines," August 2012.

2. NRC Letter, U.S. Nuclear Regulatory Commission Acceptance of ASME/ANS RA-S Case 1, March 12, 2018 (ADAMS access ML18017A964 and ML18017A966).

Any deviations between the PRA modeling methods and the clarifications and qualifications would therefore have been in the scope of the peer review, and documented as an F&O.

APLC-RAl-3 (Audit Question 3) - Open or Partially Resolved F&O's (Seismic PRA) 10 CFR 50.69(c)(1)(i) requires the PRA to "... be of sufficient quality and level of detail to support the categorization process and must be subjected to a peer review process assessed against a standard or set of acceptance criteria that is endorsed by the NRG."

In Attachment 3 of Enclosure 2 of the LAR, there are numerous open and partially resolved findings from prior peer reviews of the seismic PRA. The following questions are related to those findings and the possible impact to SSC categorization under 10 CFR 50. 69.

APLC-RAl-3a (Audit Question 3) - Open or Partially Resolved F&O's (Seismic PRA)

a. Seismic F&O 20-7 (later incorporated into F&O 1-1) states that more than one method should be considered to evaluate soil liquification to address epistemic uncertainty.

Provide the evaluation of the liquification hazard using an alternate method. If such an evaluation has not been performed, demonstrate that the open finding level F&O does not impact SSC categorization under 10 CFR 50. 69.

to AEP-NRC-2025-03 l&M Response to APLC-RAl-3a Page 21 An alternate method has not been used to evaluate the liquefaction hazard. However, the plant damage from liquefaction events would be expected to lead to a global failure mode, resulting in the simultaneous failure of multiple Structures, Systems, and Components (SSCs) and directly impacting core damage. In the context of 10 CFR 50.69, the global failure mode is directly modeled to core damage in PRA analyses and therefore dominates the risk assessment. Given the global failure mode and direct core damage, the liquefaction hazard has no impact on SSC categorization as greater direct core damage risk will only lower the risk significance of random component failures which are used in SSC categorization. Therefore, further assessment of the liquefaction hazard would have no impact on SSC categorization for 50.69 APLC-RAl-3b (Audit Question 3) - Open or Partially Resolved F&O's (Seismic PRA)

b. Seismic F&O 20-7 (later incorporated into F&O 1-1) requested the evaluation of the lateral spreading hazard at the site. The original F&O states that Figures 6-7 and 6-9 of DC-COOK-PR-09 indicates there is a continuous layer of potentially liquefiable soils (in direction towards the lake) on borings 8120, 8124, 8133, 8142, and 8141 between elevations of about 560 and 555 ft. l&M in contrast has stated that they consider Figure 6-7 to show no lateral continuity of the liquefiable boreholes.

Explain in detail why l&M has reached a different conclusion from the Seismic Peer Review Team for lateral spreading. Additionally, demonstrate that the open finding level F&O does not impact SSC categorization under 10 CFR 50.69.

l&M Response to APLC-RAl-3b Report DC-COOK-PR-09 concludes that the potential for site-wide liquefaction at the power block is low on the basis that not all borings in the power block are susceptible to liquefaction. For the boreholes that contain liquifiable soils, including those identified in the F&O, the report evaluates the effect of localized liquefaction on individual structures. This evaluation of localized liquefaction is judged to provide a sufficient level of detail to ensure that risk insights from lateral spreading are represented in the current seismic PRA. For this reason, an additional evaluation to further disposition sitewide liquefaction has not been performed, however the plant damage from liquefaction events would be expected to lead to a global failure mode. This would result in the simultaneous failure of multiple Structures, Systems, and Components (SSCs) and directly impacting core damage. In the context of 10 CFR 50.69, the global failure mode is directly modeled to core damage in PRA analyses and therefore dominates the risk assessment. Given the global failure mode and direct core damage, the liquefaction hazard has no impact on SSC categorization as greater direct core damage risk will only lower the risk significance of random component failures which are used in SSC categorization. Therefore, further assessment of the liquefaction hazard would have no impact on SSC categorization for 50.69.

to AEP-NRC-2025-03 Page 22 APLC-RAl-3c (Audit Question 3) - Open or Partially Resolved F&O's (Seismic PRA)

c. Seismic F&O 2-1 remains open because the potential for cracking in the auxiliary building has not been fully dispositioned. The conservative or non-conservative impacts on the seismic PRA results has the potential to impact overall risk insights as they apply to the application of 50.69.

Demonstrate that the open finding level F&O does not impact SSC categorization under 10 CFR 50. 69.

l&M Response to APLC-RAl-3c l&M has performed a sensitivity evaluation to examine the potential for Auxiliary Building cracking.

This sensitivity concluded that there would be no cracking at the.5*Response Level Earthquake (RLE) level, and that there would be some localized cracking at the RLE level. This cracking was present in the spent fuel pool area. This area does not contain any risk-PRA-credited components, and therefore the earthquake level would need to be increased considerably beyond the RLE to have a significant impact on plant response. At these higher earthquake levels, the damage to PRA-credited SSCs would be sufficiently widespread that the auxiliary building cracking would not become the dominating failure mode, and the risk insights would already be captured under the generic auxiliary building fragility. Based on this assessment, explicit consideration of cracking in the auxiliary building would not impact SSC categorization under 10 CFR 50.69.

APLC-RAl-3d (Audit Question 3) - Open or Partially Resolved F&O's (Seismic PRA)

d. Seismic F&O 28-2 identified gross simplifications including modeling of the nuclear steam supply system (NSSS) equipment as Jumped masses only (without inclusion of stiffness).

Some recommendations were made to close the F&O by either performing a sensitivity study to evaluate the impact of NSSS stiffness on the overall structure or correctly implementing ASCE 4-16, Section 3. 7 criteria.

Provide a sensitivity study to evaluate the impact of NSSS stiffness on the overall structure and seismic PRA. Alternatively, demonstrate that the open finding level F&O does not impact SSC categorization under 10 CFR 50. 69.

l&M Response to APLC-RAl-3d Structural support failure of Nuclear Steam Supply System (NSSS) components, which includes the reactor vessel, the reactor coolant pumps, steam generators, the pressurizer, and other components, will result in Loss of Coolant Accident (LOCA) greater than the capability of the Emergency Core Cooling System (ECCS) can mitigate. This failure is mapped to the LOCA beyond ECCS capability initiator in the Seismic PRA, which is a direct core damage contributor.

In the context of 10 CFR 50.69, the global failure mode is directly modeled to core damage in PRA analyses and therefore dominates the risk assessment. Given the global failure mode and direct core damage, the simplifications have no impact on SSC categorization as greater direct core damage risk will only lower the risk significance of random component failures which are used in SSC categorization. Therefore, the simplifications in NSSS system modeling, while significant to plant seismic risk determination, has no impact on SSC categorization for 50.69.

to AEP-NRC-2025-03 Page 23 APLC-RAl-3e {Audit Question 3) - Open or Partially Resolved F&O's {Seismic PRA)

e. Seismic F&O 28-11 identified that structural variability is ignored when determining the seismic response of SSCs. In /&M's disposition of the F&O, a statement is made that licensee staff "will review the small number of risk-significant components on a case-by-case basis, adjusting the frequency range of interest (FRO/) by an additional +/- 15 percent to ensure structural variability is captured in the fragility calculations." NE/ 00-04 Rev. 0 requires sensitivity studies for seismic PRAs (Table 5-4), including "any applicable sensitivity studies identified in the characterization of PRA adequacy."

Provide documentation of the referenced FRO/ adjustments. If the FRO/ adjustments have not been performed, will the adjustments be made as part of the 50.69 SSC categorization process?

l&M Response to APLC-RAl-3e A documented evaluation of the FROI adjustments has not been performed. A qualitative review of high-risk fragilities was performed to determine the potential impact of broadening the FROI on the fragilities used in the SPRA model. This review concluded that for the reviewed fragilities, further assessment of structural variability would not impact the model results. For this reason, it is not expected to impact SSC categorization and will not be made as part of the categorization process. An itemized table of the risk-significant fragilities and their dispositions is provided in the table labelled "APLC-RAl-3C FROI Dispositions" to AEP-NRC-2025-03 Page 24 Table: APLC-RAl-3C FROI Dispositions Fragility ID Fragility Description Disposition of Structural Variability The subject fragilities are Separation of Variable (SoV) calculations developed for collapse of the control room ceiling. As a SoV, different sources of variability have been explicitly considered. In addition, the SF-CR-CEIL-1, -

Control Room Ceiling Section 1-5 governing failure mode was identified as resulting from X direction 2, -3, -4, -5 horizontal demand, which shows limited variability outside of the considered FROI, and Z direction (vertical) demand which is taken as rigid. Therefore, no changes in fragility would result from consideration of increased structural variability.

The subject fragilities are structural failure or collapse fragilities for the major structures. The F&O is concerning structural stiffness variability SF-SCIB-AB, in seismic response models and the impact on supported SSCs.

SF-NSCIB-TB1, Auxiliary Building, Turbine Building, and Structural failure fragilities are determined based on internal forces and SF-SCIB-CONT Containment Building fragilities moments which are governed by the fundamental soil and structural frequencies. These are at low frequencies where soil variability has been shown to envelope structural variability. Therefore, no changes in fragility are necessary to account for structural variability.

The subject fragilities are electrical component fragilities that consider

> 8Hz FROI for one or more horizontal directions and rigid for the SF-250DCPNL 1, vertical and other horizontal direction, if applicable. For each horizontal SF-BATCD 250 VDC Distribution Panel, CD Battery direction where> 8Hz frequency is considered, a higher frequency peak is included in the seismic demand and would govern even if the FROI was widened by+/- 15%. Therefore, no substantive changes in fragility would result from consideration of increased structural variability.

SF-LSP, SF-The subject fragilities are generic fragilities developed based on MLOCA, SF-Offsite Power, Medium LOCA, Very Small LOCA industry values and engineering judgement. These do not include a VSLOCA fragility analysis using a FROI so no changes are applicable.

SF-SDG Supplemental Diesel Generator System These components are founded on grade, rather than mounted in a Components structure, and therefore are not impacted by structural variability.

to AEP-NRC-2025-03 Page 25 Fragility ID Fragility Description Disposition of Structural Variability The subject fragilities are developed for large groups of equipment SF-SCIC-AB1, SC-I Components in Auxiliary Building - EL. 573-based on bounding spectrum capacities and enveloping seismic SF-INV 591, 120VAC Inverter Room demands over building areas. Peak spectral demands are used in all cases, so adjustments to FROI will have no effect on fragilities.

The subject fragility is for CST piping and is found to be governed by an expansion joint with fragility determined for a separate piping system.

The expansion joint fragility is developed based on scaling from design Seismic-Induced Flood from Condensate Storage documents based on weighted average amplification of displacement SF-CST-PIPE Tank (CST) Piping considering response frequencies and modal participation factors from the original design documents. Based on the process used and the fact that maximum displacements will occur at lower frequencies where soil variabilities envelope structural variabilities, no changes in fragility are justified to account for structural variability.

The subject fragilities are SoV fragilities for relays that consider a broad RELAY_D_1, FROI, 2 to 40Hz, with In-Structure Response Spectra (ISRS) and RELAY D 2 Turbine Building and Screenhouse Relay Groups cabinet amplification factors. Based on the process used and the already broad FROI, no changes in fragility are justified to account for structural variability.

The subject fragilities are CDFM (Conservative, Deterministic Failure Margin) fragilities that consider cabinet-specific demand amplification RELAY _B_2_U2, using GENRS software. The CDFM considers a broad FROI of 2 to 40 RELAY_B_ 4_U1, Hz, however the GENRS amplification could be affected by structural RELAY _B_2_U1, frequencies and the cabinet frequencies are at higher frequencies 4KV Switchgear Complex Relay Groups where structural variability could be non-negligible relative to soil RELAY B 5 U1 variability. With this in mind, the GENRS output was reviewed, and it was generally found that In Cabinet Response Spectra (ICRS) aligned RELAY _B_ 4_U2 with structural peaks, and in no cases were adjacent peaks missed by the GENRS process. Based on this review, no changes in fragility are justified to account for structural variability.

to AEP-NRC-2025-03 Page 26 APLC-RAl-3f (Audit Question 3) - Open or Partially Resolved F&O's (Seismic PRA}

f.

Seismic F&O 28-13 states that the gap in power spectral density (PSD) as described in the F&O should be addressed per latest fragility guidance document. In /&M's disposition of the F&O, l&M indicates that there are not significant gaps in energy near frequencies that are important to risk-significant fragilities based on a review of the non-interpolated PSDs and PSDs developed using a linear frequency interpolation.

Provide a graphical representation of the non-interpolated PSDs and PSDs developed using a linear frequency interpolation that supports the /&M's position that no significant PSD gaps exist near the frequencies important to risk-significant fragilities. Confirm that the ground motion time histories used in fragility calculations are adequate for SSC categorization under 10 CFR 50. 69.

l&M Response to APLC-RAl-3f PSD plots are provided in the figures below for the foundation-level time histories developed for seismic response analysis of the Containment Structure, Turbine Building, and Auxiliary Building. PSDs are presented in their raw format and using a +- 20% frequency averaging window consistent with guidance from SRP 3. 7.1 Appendix A and B.

Based on a review of the PSDs, potential gaps in energy are only observed at low frequencies, less than ~0.6 Hz. The most significant gap in energy is at 0.4 Hz in the vertical direction for the Auxiliary Building. These observed frequencies are well below any building response frequencies considering Soil Structure Interaction (SSI) effects and will have no impact on the fragility results. This confirms that the ground motion time histories used in fragility calculations are adequate for SSC categorization under 10 CFR 50.69.

to AEP-NRC-2025-03 0.1 0.01 0.001 0.0001 3

0 lE-05

~

lE-06 lE-07 lE-08 lE-09 0.01 f

I

/

0.1 Freqoency [Hz]

Figure 1: Containment Structure Raw PSDs Page 27 TH_UHRSF _X_PSD_R""'

TH_UHRSF _Y_PSO_Rzrw TH_UHRSF _Z_PSD_RilW 10 100 to AEP-NRC-2025-03 Page 28 TH UHRSF X PSD RA TH=UHRSF=Y=PSD=RA TH_UHRSF _Z_PSD_RA 0.01 -+------------+-------------+-----------+------------.-<

0.001

i;

'.9 0

0.0001

n.

l E-05 lE-06-+------------+-------------+-----------+------------U*--i l E-07-+------~~~~...+------~~~r-r-r+------~.....-~r-r...+---r---.....-.....-~r-r'T""i 0.01 0.1 10 100 F l"'luency [Hz]

Figure 2: Containment Structure +-20% Rolling Average PSDs to AEP-NRC-2025-03 Page 29 TSC_TH_UHRSF _X_PSD_Raw TSC TH UHRSF Y PSO Raw 0.1==1------------+------------+----------- -

rsc:rH:UHRSF:z:Pso:Raw 0.01 =<-----------------------------------+------------<

l E-09 -t------,----,--r---,-...--,-,-,-+-----,----,--r---,-...--,-,-,-+-----,-~--r---r-T""T-,-r+------.-~--,--,-T""T-,-rl 0.01 0.1 Frequency (Hz)

Figure 3: Auxiliary Building Raw PSDs 10 100 to AEP-NRC-2025-03 Page 30 TSC TH UHRSF X PSD RA TSC=TH=UHRs(::Y=PSD=RA TSC_TH_UHRSF _Z_PSD_RA 0.1 =<----------,----"--+-------------t-----------+------------<

0.01-------------.-+-----------+-----------+------------<

0.001

'Jq

~

0 "'

0.0001 c..

l E-05 l E-06 l E-07

~

0.01 0.1 10 100 Frequency (Hz)

Figure 3: Auxiliary Building +-20% Rolling Average PSDs to AEP-NRC-2025-03 Page 31 0.1 0.01 0.001 0.0001

-i:r

'.9 0

lE-05 0..

1E-06 1E-07 1E*OB 1E*09 TH_UHRSF_X_PSO_Raw TH_UHRSF_Y_PSD_Raw

=1--------------1------------1------------ -

TH_UHRSF_Z_PSD_Raw 0.01 0.1 Frequency [Hz]

Figure 4: Turbine Building Raw PSDs 10 100 to AEP-NRC-2025-03 Page 32 TH UHRSF X PSO RA TH=UHRSF=Y=Pso:RA TH_UHRSF _Z_PSD_RA 0.01=1--------------+------,--------,--------t-----------,,--+--------,------j 0.001

~

.!!I C

a.

0.0001 l E-06 -+-----,----,-~-.-T"'"'"T-.-r+------,----,-~-.-T"'"'"T-.-rt------,--,-----,--.--r-T"""i--rr+-----,--,-----,--r---r-T"""i--rT""i 0.01 0.1 10 100 Frequency (Hz]

Figure 5: Turbine Building +-20% Rolling Average PSOs to AEP-NRC-2025-03 Page 33 APLC-RAl-4 (Audit Question 4) - External Hazard Screening 10 CFR 50.69(c)(1)(ii) requires an applicant while performing an SSC Categorization Process to "determine SSC functional importance using an integrated, systematic process for addressing initiating events (internal and external), SSCs, and plant operating modes, including those not modeled in the plant-specific PRA... All aspects of the integrated, systematic process used to characterize SSC importance must reasonably reflect the current plant configuration and operating practices, and applicable plant and industry operational experience."

The following questions are related to the screening of external hazards as documented in the LAR Attachment 4 of Enclosure 2.

APLC-RAl-4a (Audit Question 4) - External Hazard Screening

a. As part of the LAR, a variety of external hazards were screened out of consideration for the 10 CFR 50. 69 categorization process in Attachment 4 of Enclosure 2. For the external hazard of ice cover - portions of the circulating water deicing system were credited as HSS to justify the exclusion of this hazard from the categorization process. Screening Criterion C1 (event damage potential is less than events for which the plant is designed), C4 (event is included in the definition of another event), and C5 (event develops slowly, allowing adequate time to eliminate or mitigate the threat) are all referenced as partial bases for screening the ice cover hazard.

With respect to the screening of the ice cover external hazard - please provide one of the following:

1. Justify why categorizing the proposed portions of the circulating water deicing system as HSS is adequate for the screening of this hazard consistent with NE/ 00-04, Figure 5-6, "Other External Hazards," and the discussion that if a component is credited as part of the safe shutdown paths evaluated, it is considered safety significant. "

2. Provide a different technical basis and/or criterion for the screening this hazard.

l&M Response to APLC-RAl-4a The primary effect of ice cover events at CNP would be a loss of offsite power due to ice buildup on offsite power lines, resulting in their eventual failure. The impact of this is captured under the modeling and frequency for the weather-related loss of offsite power initiating event in the internal events PRA. Because the frequency and impact are covered under another event, this portion of the hazard screens under Criterion C4.

A secondary effect of ice cover events would be the buildup of ice on Lake Michigan, which serves as the site's heat sink. However, plant processes/design preclude a significant impact on plant operation due to this ice buildup. The plant design includes a de-icing operation mode of the intake system, as described in the UFSAR, Section 10.6.2:

"De-icing capability to the intake cribs is provided by shutting off flow in the middle intake pipe to the screen house, by closing its motor operated sluice gate and sending a portion of "warm" discharge water from either the Unit No. 1 or Unit No. 2 discharge tunnel back through the middle to AEP-NRC-2025-03 Page 34 pipe to the lake. The three intake cribs are arranged in a triangular pattern with the middle pipe at the apex (farthest from shore) so that the heated water will recirculate to the two other intake pipes thus keeping the intakes free of ice."

l&M procedures prompt consideration of entering de-icing mode at a circulating water temperature of 35°F, which is expected to be reached long before significant ice buildup. This de-icing strategy is to maintain plant operation/generation and is not required to safely shut down the plant in an emergency. If another event were to require a plant trip during an ice cover event, the circulating water pumps would trip off. This would allow the systems used for plant shutdown (Essential Service Water, Non-Essential Service Water) to operate using the forebay without concern of the circulating water system drawing in lake ice (their combined flow rate is a small fraction of the circulating water flow rate). Because ice cover is slow-developing, and there are adequate plant processes/procedures to mitigate the impact on plant intake, the secondary effect of ice cover (ice buildup on Lake Michigan) is screened out under Criterion C5.

While successful entry into de-ice, mode is not necessary to mitigate any potential screened initiating events at CNP, and therefore no de-icing components are required to be categorized HSS due to external event screening or safe shutdown criteria, a conservative decision was made to keep the de-icing valves (1-WMO-16, 2-WMO-26) HSS in categorization due to their significance in maintaining plant operation during cold weather.

While preparing the updated response for APLC-RAl-4a, the team concluded that Criterion 4 was the more appropriate criteria for the basis provided in the original Table 4.4. The ASME standard for Criterion 1 requires an evaluation of plant design bases to estimate the resistance of plant structures and systems, which was not performed as part of the basis originally linked to Criterion

1. Rather, the original part of the screening linked to Criterion 1 better matches Criterion 4, which is that the event is included in the definition of another event (LOOP). Therefore, Table 4.4 of the original LAR is being revised as shown, below.

Ice Cover G1-,C4, The principal effects of such events would be to cause a loss of C5 offsite power and are addressed in the weather-related Loss of Offsite Power initiating event in the internal events PRA model for CNP (C-4-4).

+l=le J:)eteRtial l=la~aFEI ef ise sle§§iR§ tl=le iRtake is J:)FeveRteEI by tl=le Qe isiR§ System. +l=lis system 91:)eRs YRit 1 aREI 2 FReteF 91:)eFateEI val1.ies 1 WMQ 1 e aREI 2 WMQ 2e, Fesl:)estively, i.vl=lisl=I aFe iR tl=le 591' elevatieR ef tl=le SsFeeR l=ie1::1se aREI J:)FeviEles siFs1::1latiR§ wateF Elissl=laF§e frnm tl=le seREleRseFs te tl=le Ele isiR§ t1::1RRel te J:)FeveRt ise b1::1ilEl1::1J:) iR tl=le feFebay (e.18) {Ga). +l=leFefeFe, tl=lese semJ:)eReRts { 1 WMQ 1 e aREI 2 VVMG 2e) sFeEliteEI feF miti§atiR§ ise sle§§iR§ tl=le iRtake aREI wl=lisl=I aFe J:)eFtiReRt te tl=lis l=la~aFEI SGFeeRiR§ will be tFeateEI as l=iSS if sate§eFi~eEI, iR 8669F98R6e witl=I tl=le §1::liElaRGe J:)F8¥iEleEI iR Nel QQ Q4 i;i§l::IFe a e (e.Ql +l=le IQP will be iRfeFmeEI ef tl=le basis feF ise sle§§iR§ tl=le iRtake {lse GeveF) ssFeeRiR§ El1::1FiR§ tl=leiF Fe1*1iews ef sate§eFi~atieR Fes1::1lts.

to AEP-NRC-2025-03 Page 35 A secondary effect of ice cover events would be the buildup of ice on Lake Michigan, which serves as the site's heat sink. However, plant processes/design preclude a significant impact on plant operation due to this ice buildup. The plant design includes a de-icing operation mode of the intake system, as described in the UFSAR, Section 10.6.2:

"De-icing capability to the intake cribs is provided by shutting off flow in the middle intake pipe to the screen house, by closing its motor operated sluice gate and sending a portion of "warm" discharge water from either the Unit No. 1 or Unit No. 2 discharge tunnel back through the middle pipe to the lake. The three intake cribs are arranged in a triangular pattern with the middle pipe at the apex (farthest from shore) so that the heated water will recirculate (o the two other intake pipes thus keeping the intakes free of ice."

Procedures 1(2)-OHP-4021-057-002, "Placing In/Removing From Service Circulating Water Deicing System" prompt consideration of entering de-icing mode at a circ water temperature of 35°F, which is expected to be reached long before significant ice buildup.

This de-icing strategy is to maintain plant operation/generation and is not required to safely shut down the plant in an emergency. If another event were to require a plant trip during an ice cover event, the circulating water pumps would trip off. This would allow the systems used for plant shutdown (Essential Service Water, Non-Essential Service Water) to operate using the forebay without concern of the circulating water system drawing in lake ice (their combined flow rate is a small fraction of the circulating water flow rate).

Because ice cover is slow-developing, and there are adequate plant processes/procedures to mitigate the impact on plant intake, the secondary effect of ice cover (ice buildup on Lake Michigan) is screened out under Criterion C5.

While successful entry into de-ice mode is not necessary to mitigate any potential screened initiating events at CNP, and therefore no de-icing components are required to be categorized HSS due to external event screening or safe shutdown criteria, a conservative decision was made to keep the de-icing valves ( 1-WMO-16, 2-WMO-26) HSS in categorization due to their significance in maintaining plant operation during cold weather.

See also "External Flooding" (C4).

Based on this review, the Ice Cover impact hazard can be considered to be small.

to AEP-NRC-2025-03 Page 36 APLC-RAl-4b (Audit Question 4) - External Hazard Screening

b. [Release of Chemicals from On-site Storage] In order to screen out the external hazard of a release of chemicals from on-site storage, CA-03-01 documents an analysis showing a hydrazine spill at a storage location would not impact control room habitability.

Has a hydrazine spill been evaluated for locations during transit or use within the power block

- not just storage? If not, justify why performance of such an evaluation would not impact SSC categorization under 10 CFR 50. 69.

l&M Response to APLC-RAl-4b A calculation has been completed to evaluate Control Room habitability following postulated release of hydrazine and states:

"Damage to the hydrazine container by external forces can be attributed to such accidents as a forklift mishap. However, such damage would be less severe, and an instantaneous release of the entire contents would not be probable. Consequently, any accidental spills occurring during transportation of containers would be less severe than the postulated scenario. Such a transportation accident would not lead to a severe consequence since the spill would be reported in a short time."

CA-03-01 makes the case that the on-site storage release bounds any potential impact from an in-transit hydrazine spill. Based on the qualitative disposition provided above, the characteristics that make an in-transit spill less severe (lower volume, immediately identified) do not rely on specific plant components. For this reason, it is not expected that an in-transit hydrazine spill analysis would identify any additional components as HSS.

APLC-RAl-4c (Audit Question 4) - External Hazard Screening

c. [Turbine-Generated Missiles] In order to screen out the external hazard of turbine-generated missiles, a turbine missile probability analysis was performed for the "Unit 1 and Unit 2 Alstom low-pressure turbines."

Describe whether a similar analysis was done for any high-pressure turbines on-site at CNP.

If not, justify why the performance of such an analysis would not impact SSC classification under 10 CFR 50. 69.

to AEP-NRC-2025-03 Page 37 l&M Response to APLC-RAl-4c The CNP UFSAR, in section 14.1.13.2 discusses failure of turbine-generator rotating elements.

This section provides qualitative assessment of high-pressure turbine missiles, stating:

And "The high-pressure turbine rotors of both units can withstand, without failure, the maximum theoretical runaway speed of 200% of rated speed, i.e., 3600 rpm. If there is premature failure, thrown parts should be retained by the cast, heavy wall section of the bolted, high-pressure shells."

"The high pressure turbine rotor is surrounded by heavy cast steel shells. Based on the several overspeed accidents, which have taken place at other power plants, it is believed that even with severe troubles, few fragments can be released, and that they are of lower speed and mass than the pieces already discussed in the low pressure turbine."

Due to these qualitative assessments, no detailed analysis of High Pressure (HP) turbine missiles has been performed. However, from these qualitative assessments, it is reasonable to conclude that additional assessment of HP turbine missiles would not produce insights beyond those produced by the Low Pressure (LP) turbine missile analysis, and therefore would not impact SSC categorization.

References:

1. Letter from K. J. Ferneau, Indiana Michigan Power Company (l&M), to U.S. Nuclear Regulatory Commission (NRC), "Donald C. Cook Nuclear Plant Units 1 and 2 Application to Adopt 10 CFR 50.69, "Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors"," dated March 6, 2024, Agencywide Documents Access and Management System (ADAMS) Accession No. ML24073A234.

2. Email from S. P. Wall, NRC, to M. K. Scarpello, l&M, "Final RAI - D.C. Cook 1 & 2-License Amendment Request Regarding Adoption of 10 CFR 50.69 (EPID No. L-2024-LLA-0025),"

dated December 30, 2024, ADAMS Accession No. ML24366A003.