ML25302A450

Enclosure 3 - Us Sfr Owner, LLC - Summary Report for the General Regulatory Audit of the Kemmerer Power Station Unit 1 Construction Permit Application

ML25302A450
Person / Time
Site:	Kemmerer File:TerraPower icon.png
Issue date:	11/21/2025
From:	Office of Nuclear Reactor Regulation
To:	TerraPower, US SFR Owner
References
EPID L-2024-CPS-0000
Download: ML25302A450 (1)

Text

OFFICIAL USE ONLY PROPRIETARY INFORMATION OFFICIAL USE ONLY - PROPRIETARY INFORMATION US SFR OWNER, LLC -

SUMMARY

REPORT FOR THE GENERAL REGULATORY AUDIT OF THE KEMMERER POWER STATION UNIT 1 CONSTRUCTION PERMIT APPLICATION PROBABILISTIC RISK ASSESSMENT As part of its review of the Kemmerer Unit 1 (KU1) Construction Permit (CP) application, the staff conducted a comprehensive audit of the applicants probabilistic risk assessment (PRA) and supporting documentation. The primary audit objective was to confirm that the PRA provides a reasonable technical foundation to support the CP application, Licensing Modernization Project (LMP), and other risk-informed approaches and programs referenced in the preliminary safety analysis report (PSAR), including Reliability and Integrity Management (RIM) and the Design Reliability Assurance Program (D-RAP).

To achieve this objective, the staff conducted a three-day in person audit with a virtual component to facilitate document access and enhanced technical staff participation. The audit provided the staff with broader access to PRA materials and facilitated direct technical exchanges to clarify key issues and verify staff understanding of PRA modeling approaches and results.

1. Applicable Regulatory Guidance The applicable regulatory guidance documents related to PRA and its use for evaluating accident probabilities, plant operational safety, and risk to public health are as follows:

RG 1.233, Guidance for a Technology-Inclusive, Risk-Informed, and Performance-Based Methodology to Inform the Licensing Basis and Content of Applications for Licenses, Certifications, and Approvals for Non-Light-Water Reactors (ML20091L620).

RG 1.247, Acceptability of Probabilistic Risk Assessment Results for Non-Light-Water Reactor Risk-Informed Activities, For Trial Use (ML21235A008).

RG 1.253, Guidance for a Technology-Inclusive Content-of-Application Methodology to Inform the Licensing Basis and Content of Applications for Licenses, Certifications, and Approvals for Non-Light-Water Reactors (ML23269A222).

2. Overview of PRA Audit Structure The staff conducted the PRA audit through a combination of virtual and in person review activities that were carried out in parallel. The virtual component provided an ongoing platform for document review and technical discussions, while the in person provided opportunities for direct engagement, in-depth discussions, and hands-on examination of the PRA models and supporting tools.

This integrated approach ensured that staff could address both high-level and detailed technical aspects of the KU1 CP PRA in a coordinated and efficient manner. It also facilitated direct interaction between the staff and the applicants PRA team as questions and clarifications arose during the KU1 CP safety review.

Virtual activities provided staff insights into the structure, content, and maturity of the KU1 PRA.

More than 35 PRA-related documents were made available for staff review through the

OFFICIAL USE ONLY PROPRIETARY INFORMATION 2

OFFICIAL USE ONLY - PROPRIETARY INFORMATION applicants electronic reading room (ERR). Based on its review of the PSAR and these supporting documents, the staff generated several technical questions related to the PRA development process, quantification methods, and the use of the PRA in support of the CP application. These questions, summarized in Enclosure 1, were discussed through multiple online meetings with the applicant. The applicants timely responses and supplemental explanations helped the staff identify areas warranting further review, especially during the in-person audit component.

The in-person audit was held over three days, from Wednesday, January 22, through Friday, January 24, 2025. Approximately ten NRC staff participated in person, with additional staff joining virtually. The third day of the audit focused on a hands-on modeling session, during which staff interacted directly with the PRA models, verified quantification logic, and assessed sensitivity analyses.

The in-person audit was specifically intended to enhance the staffs understanding of the PRA model structure, logic, quantification processes, and treatment of uncertainties. It also served to identify any information that should be established or docketed to support the staffs safety evaluation.

3. Focus Areas of the In Person PRA Audit During the in-person audit, the staff performed walkthroughs of the PRA model to confirm its understanding of the structure, logical framework, and quantification processes. The audit covered the modeling sequence, from initiating events through event sequence generation, frequency estimation, importance ranking, and treatment of uncertainties.

Particular attention was given to the models development using CAFTA software, including the logic modeling approach, model hierarchy, and parameterization. The staff assessed key assumptions, modeling parameters, and simplifications that could significantly influence the PRAs results and insights.

The staff reviewed selected sensitivity analyses to assess the robustness of the PRA and to better understand the uncertainties associated with the modeling. The applicants treatment of both epistemic and aleatory uncertainties was examined to ensure consistency with standard practices.

Gaps identified during both the applicants CP PRA self-assessment and the staffs independent review were discussed in detail. The applicant provided justifications demonstrating the impact of the identified gaps on the CP application and LMP implementation. The applicant also presented its path forward for addressing the identified issues prior to the operating license (OL) stage.

The staff audited the applicants established processes and procedures for maintaining and upgrading the PRA as the design evolves. The developed configuration control program ensures continued alignment of the PRA with the KU1 plant design and the information presented in the application.

OFFICIAL USE ONLY PROPRIETARY INFORMATION 3

OFFICIAL USE ONLY - PROPRIETARY INFORMATION The staff audited and discussed a comprehensive set of topics with the applicant to confirm the development, implementation, and technical adequacy of the KU1 PRA. The audit and discussions addressed key areas of PRA development, assessment, technical elements, specialized analyses, applications, and hands-on modeling as summarized below.

1. Plant Operating State Analysis Development of plant operating states (POS) and documentation of these calculations; and Development, documentation, and use of low power and shutdown (LPSD) PRA results for LMP implementation.

2. Initiating Event Analysis Processes for identifying plant-specific and design-specific initiating events; Database for plant-and design-specific events; Grouping methodology for initiating events, including treatment of shared system initiators (e.g., heating, ventilation, and air conditioning (HVAC), instrument air, argon); and Incorporation of interlocks and consideration of control room operator actions in event identification.

3. Event Sequence Analysis Mapping of initiating events to event sequences, barrier failures, and associated radiological consequences; Treatment of recovery actions and consistency with human reliability analysis; and Handling of containment isolation and head access area (HAA) isolation functions.

4. Success Criteria Development Functional requirements for each key safety function; Criteria for protection of confinement barriers; and Linkage between key safety functions, initiating events, and systems that fulfill these functions.

5. Systems Analysis Completeness of system evaluations, including shared systems and interfaces; Incorporation of non-reactor sources, system degradation, and hazard conditions into PRA; and Treatment of items classified as super components and resolution of not met PRA supporting requirements.

6. Human Reliability Analysis Systematic evaluation of human actions, including pre-initiator events, human-induced initiating events, and post-initiating events actions; Scope and definition of recovery actions included in the analysis; and Methodology for quantifying uncertainties and screening practices in the human reliability analysis.

OFFICIAL USE ONLY PROPRIETARY INFORMATION 4

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

7. Data Analysis Sources of reliability data, component boundaries, and distribution parameters; and Treatment of recovery, common cause failures, and equipment unavailability.

8. Hazard Screening Identification and treatment of hazards; and Rationale for screening internal and external hazards, including discrepancies in justification for similar hazards (e.g., external flooding versus rain or dam failure).

9. Event Sequence Quantification Criteria for grouping event sequences into families; Calculation of radionuclide transport barrier failures, cutsets generation, key assumptions, and screened out initiating events; and Evaluation of uncertainty impacts on event sequence quantification.

10. Mechanistic Source Term Analysis Characterization of potential radiological releases; Gaps and documentation completeness related to source term uncertainty; and Rationale for relying on dose results rather than direct source term uncertainty analysis.

11. Radiological Consequence Analysis Applicability of site-specific information for consequence analysis; Comparison against LMP risk metrics; and Treatment of generic meteorological data and zero-risk early fatality results.

12. Risk Integration Criteria for establishing risk significance and integrating frequencies and consequences; Absolute risk significance for initiating events, structures, systems, and components (SSCs), functions, and event sequence families; and Integration of internal and external hazards and evaluation of uncertainties.

13. Spent Fuel PRA Construction of spent fuel PRA; Dependencies and correlations between spent fuel PRA and internal events PRA; and Shared SSCs, plant systems, and operator actions across PRAs.

14. PRA Self-Assessment Approach for validating computational models and rationale for grading PRA supporting requirements (i.e., MET, NOT MET, or MET with GAP);

Incorporation of confinement barriers, their operating conditions, and failure modes; Treatment of time-dependent event sequences and time-phased dependencies;

OFFICIAL USE ONLY PROPRIETARY INFORMATION 5

OFFICIAL USE ONLY - PROPRIETARY INFORMATION Integration of instrumentation and control systems, including identification of control functions needed for event response; and Use of expert judgment and adherence to American Society of Mechanical Engineers (ASME) / American Nuclear Society (ANS) RA-S-1.4-2021 non-light-water reactor (non-LWR) PRA standard guidance.

15. PRA Configuration Control Processes for monitoring plant design, construction, and operational changes Methods to maintain and upgrade the PRA; Implementation of self-assessment and peer review findings, including cumulative impacts on PRA applications; and Documentation and traceability of PRA.

16. Sensitivity Analysis Sensitivity analyses that were performed on key modeling assumptions; Sensitivity analyses that were conducted on data parameters; and Sensitivity analyses that were performed for staff-defined cases.

17. Model Walkthroughs and Scenario Analysis Walkthroughs and model runs for key scenarios to evaluate PRA development from initiating events to risk quantification, including:

o Loss of offsite power (LOOP);

o Total loss of primary flow (LOF);

o Transient overpower due to rod insertion (TOP-R);

o Loss of the nuclear gas system (argon, NGS-ARG);

o Fuel handling events during reactor vessel operations; and o Movement of spent fuel assemblies to the ex-vessel storage tank.

4. Audit Reference Materials USO made the following documents available in the ERR for the staffs audit.

1)

TerraPower LLC, Natrium PRA Plant Operating States Analysis, NAT-7109, Rev. A.

2)

TerraPower LLC, Natrium PRA System Analysis - Confinement, NAT-7137, Rev. A.

3)

TerraPower LLC, Natrium PRA System Analysis - Control (RPS-NIC), NAT-7138, Rev. A.

4)

TerraPower LLC, Natrium PRA System Analysis - Electrical, NAT-7139, Rev. A.

5)

TerraPower LLC, Natrium PRA System Analysis - Ex-Vessel Handling Machine (EVHM), NAT-7140, Rev. A.

6)

TerraPower LLC, Natrium PRA System Analysis - Ex-Vessel Storage Tank (EVST)

Cooling, NAT-7141, Rev. A.

7)

TerraPower LLC, Natrium PRA System Analysis - Gaseous Radwaste Processing, NAT-7142, Rev. A.

8)

TerraPower LLC, Natrium PRA System Analysis - Inherent Reactivity Feedback (IRF)

System, NAT-7143, Rev. A.

9)

TerraPower LLC, Natrium PRA System Analysis - Intermediate Air Cooling, NAT-7144, Rev. A.

OFFICIAL USE ONLY PROPRIETARY INFORMATION 6

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

10) TerraPower LLC, Natrium PRA System Analysis - Intermediate Heat Transport System, NAT-7145, Rev. A.

11) TerraPower LLC, Natrium PRA System Analysis - Miscellaneous, NAT-7146, Rev. A.

12) TerraPower LLC, Natrium PRA System Analysis - Molten Salt and Steam Generation System, NAT-7147, Rev. A.

13) TerraPower LLC, Natrium PRA System Analysis - Nuclear Island Air and Inert Gas Distribution System, NAT-7148, Rev. A.

14) TerraPower LLC, Natrium PRA System Analysis - Primary Heat Transport, NAT-7149, Rev. A.

15) TerraPower LLC, Natrium PRA System Analysis - Reactivity Control, NAT-7150, Rev. A.

16) TerraPower LLC, Natrium PRA System Analysis - Reactor Air Cooling (RAC)

System, NAT-7151, Rev. A.

17) TerraPower LLC, Natrium PRA System Analysis - Sodium Cover Gas, NAT-7152, Rev. A.

18) TerraPower LLC, Natrium PRA System Analysis - Sodium Processing System, NAT-7153, Rev. A.

19) TerraPower LLC, PRA Methodology, NAT-7444, Rev. A.

20) TerraPower LLC, Natrium PRA Plant Operating States Analysis, NAT-7109, Rev. A.

21) TerraPower LLC, Natrium Demonstration - Spent Fuel PRA, NAT-7547, Rev. B.

22) TerraPower LLC, Natrium PRA Initiating Events Analysis, NAT-7127, Rev. A.

23) TerraPower LLC, Natrium PRA Event Sequence Analysis, NAT-7154, Rev. A.

24) TerraPower LLC, Natrium PRA Success Criteria Analysis, NAT-7580, Rev. A.

25) TerraPower LLC, Natrium PRA Data Analysis, NAT-7878, Rev. A.

26) TerraPower LLC, Natrium PRA Human Reliability Analysis, NAT-7879, Rev. A.

27) TerraPower LLC, Natrium Event Sequence Quantification, NAT-7364, Rev. B.

28) TerraPower LLC, PRA Source Term and Radiological Consequences Analysis, NAT-6161, Rev. A.

29) TerraPower LLC, PRA Analysis - Risk Integration, NAT-7827, Rev. B.

30) TerraPower LLC, Natrium Demonstration - PRA Screening of External Hazards, NAT-8294, Rev. B.

31) TerraPower LLC, Natrium PRA - Risk Integration Sensitivities, NAT-8076, Rev. A.

32) TerraPower LLC, Quantitative Health Objective Updates for the IDPP, NAT-8807, Rev. A.

33) TerraPower LLC, Natrium Plant PRA Self-Assessment, NAT-8218, Rev. B.

34) TerraPower LLC, Natrium PRA Configuration Control Procedure, NAT-13718, Rev. 0.

35) TerraPower LLC, In-Person PRA Audit PPT 24-2025.

36) TerraPower LLC, In-Person PRA Sensitivities.

5. Audit Findings a) POSs Analysis PSAR section 3.1.1.6 summarizes the scope of USOs POS analysis. To further understand the description in the PSAR, the staff reviewed NAT-7109, Rev. A, Natrium PRA Plant Operating States Analysis in the audit. The review focused on the approach used to perform the POS analysis, ensuring that the analysis is generally consistent with accepted industry practices, the

OFFICIAL USE ONLY PROPRIETARY INFORMATION 7

OFFICIAL USE ONLY - PROPRIETARY INFORMATION modeling assumptions and inputs are reasonable, and the POS analysis is consistent, to the extent possible, with the preliminary plant design.

The KU1 POS analysis was performed drawing on insights from the ((

)) and the available design stage information. The analysis was performed following the guidance in the non-LWR PRA standard ASME/ANS RA-S-1.4-2021. The POS analysis established relevant operating conditions for the KU1 reactor to enable a robust evaluation of plant risk throughout its lifecycle.

The defined POSs were informed by expected outage schedules, operational practices, and the most current design data. The analysis included USO internal reviews to identify plant configurations or activities that could affect radionuclide release barriers, such as the metal fuel matrix, fuel cladding, sodium coolant, reactor vessel, and functional containment, or compromise key plant safety functions, including reactivity control, core flow, primary sodium heat removal, and confinement.

As stated in NAT-7109, the KU1 reactor is designed to operate at full power with over ((

))

availability. During full-power operation, the configuration and status of release barriers are expected to remain stable unless disturbed by an initiating event. However, during refueling, maintenance, or other non-power conditions, changes to one or more release barriers may be required, potentially introducing unique challenges to their integrity. Such activities may also affect SSCs that support key safety functions. Accordingly, the KU1 POS analysis emphasized identifying plant configurations and evolutions that could result in reconfiguration or impose additional demands on release barriers or safety-related (SR) SSCs.

The derivation of the KU1 POSs was conducted in two main stages: (1) identification of plant evolutions and (2) definition of the corresponding POSs. A plant evolution is defined as a sequence of related activities during which the plant transitions from one POS to another, for example, transitioning from full power to low power or shutdown. As KU1 is currently in the design phase, a representative set of plant evolutions was identified based on a review of available design and operational information, focusing on periods where significant changes in plant conditions are expected. In addition to normal refueling outages, the analysis considered at-power transitions, forced outages, and major maintenance periods.

Once plant evolutions were identified, a set of POSs was developed for each evolution. These POSs were defined based on a detailed review of KU1 specific design information, input from knowledgeable USO personnel, and consideration of ((

)). The resulting POSs collectively provide comprehensive coverage of all time intervals associated with each plant evolution.

Because plant conditions vary minimally between different evolutions during online operation, the online POSs were consolidated. As a result, eleven KU1 POSs were defined across ten distinct plant conditions. These are detailed in tables 4 through 24 of NAT-7109 and summarized below, including the estimated total duration each POS is expected to last over the plants 60-year life and the corresponding probability of being in each POS. The probability represents the fraction of time the plant is expected to be in a given POS over its 60-year (21,900-day) lifespan.

OFFICIAL USE ONLY PROPRIETARY INFORMATION 8

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

((

))

b)

Initiating Event Analysis PSAR section 3.1.1.7, specifically table 3.1-2 and table 3.1-3 summarize the hazards that were either screened out or into the PRA, respectively. To further understand the description in the PSAR regarding the PRA initiating events (IEs), the staff reviewed NAT-7127, Rev. A, Natrium PRA Initiating Events Analysis in the audit. The review focused on confirming the characteristics and attributes of the IE analysis to ensure the analysis aligns with the non-LWR PRA standard ASME/ANS RA-S-1.4-2021 and RG 1.247. The staff reviewed the approach and method applied in the IE analysis, ensuring they are reasonable and consistent with accepted industry practices and the analysis aligns with the preliminary plant design to the extent possible. The staff also assessed the assumptions to ensure reasonableness.

The KU1 IE analysis primarily focuses on events that could impact the reactor core and core components within the reactor enclosure system (RES), including in-vessel storage, core assemblies transferred to the EVST during refueling outages, and radionuclide sources associated with systems directly connected to the RES during reactor operation, such as the sodium processing system (SPS) and sodium cover gas system.

As described in NAT-7127, Rev. A, the KU1 IE analysis consists of three key steps:

1. Identification of IEs;

2. Grouping of IEs; and

3. Estimation of IE frequencies.

OFFICIAL USE ONLY PROPRIETARY INFORMATION 9

OFFICIAL USE ONLY - PROPRIETARY INFORMATION The first step emphasizes developing a comprehensive list of potential IEs through a systematic approach using multiple sources. This includes consideration of both equipment-induced and human-induced events. The identification process involved reviewing published references relevant to non-LWRs, analyzing plant-specific events based on design and safety assessments, and examining each system for special initiators arising from interactions between systems or redundant trains. Identification of human-induced events was limited by the unavailability of operating and maintenance information for the system. Some pre-IE and post-IE actions were identified as part of the human reliability analysis, but no human-induced initiating events were identified at the CP stage. This human-induced IEs evaluation to support a complete set of IEs will be reviewed by the staff at the OL stage.

The KU1 IE identification process was supported by a review of established event data from key industry and regulatory sources, including:

NUREG/CR-5750, Rates of Initiating Events at U.S. Nuclear Power Plants: 1987-1995; (ML070580080)

NUREG/CR-6928, Industry-Average Performance for Components and Initiating Events at U.S. Commercial Nuclear Power Plants; (ML070650650)

NUREG/CR-3862, Development of Transient Initiating Event Frequencies for Use in PRAs; (ML070570060) and Electric Power Research Institute (EPRI) Report 3002003129, Advanced Nuclear Technology: Advanced Light Water Reactor Utility Requirements Document, Rev. 13.

In the search for IEs, USO utilized historic data for liquid metal reactors and performed design-specific evaluations for Natrium design. The historic data consisted primarily of information from the PRISM reactor, the U.S. Advanced Liquid Metal Reactor Program, and the Experimental Breeder Reactor II (EBR-II). Specific documentation the applicant reviewed included:

Experimental Breeder Reactor II Level 1 PRA, Argonne National Laboratory, 1991; KALIMER-600 Sodium Cooled Fast Reactor, NEA/CSNI/R, 2011; Prototype Fast Breeder Reactor Level 1 PRA, Nuclear Engineering and Design, 2012; Advanced Sodium Technological Reactor for Industrial Demonstration Simplified PSA, ICAPP 2012; Japanese Liquid Metal Fast Breeder Reactor PSA Application, IAEA, 1993; PRISM All Modes Risk Integration Report, GE Hitachi Nuclear Energy, 2017; and PRA of the Advanced Liquid Metal Reactor, GE Nuclear Energy, 1994.

Concurrently, USO conducted a system-level review of the KU1 design to evaluate each systems potential to initiate an IE. Systems were screened in or out based on their potential impact on reactor or steam plant operations. Design-specific evaluations included failure mode and effects analysis (FMEA), review of the available design documentation, and discussions with the system design engineers. For systems supported by an existing FMEA, the FMEA results were utilized to identify failure modes that could result in a reactor trip or other operational disturbances classified as IEs. For systems lacking an FMEA, available design information was examined to assess whether failures of the system, train, or individual components could uniquely initiate an IE within the plant. The process used for IEs identification

OFFICIAL USE ONLY PROPRIETARY INFORMATION 10 OFFICIAL USE ONLY - PROPRIETARY INFORMATION provides reasonable assurance that the events of interest were identified, given the limitations imposed by preliminary design.

Where design information was unavailable, USOs IE identification process indicates that determinations were made leveraging PRA modeling practices, established nuclear plant design principles, and familiarity with the KU1 design. The staff observed that the process employed did not always lead to conservative assumptions, as discussed in more detail below with respect to the system (SY) analysis.

In the second step of the IE analysis, to optimize the PRA and reduce the number of IEs requiring detailed analysis, the identified IEs were grouped based on key characteristics.

Specifically, IEs exhibiting similar plant responses, success criteria, timing, and effects on the operability and performance of operators and mitigating systems were consolidated. Events that could be bounded by the worst-case impact within the group were also combined. The frequency of these grouped events was calculated as the sum of the individual event frequencies.

In the third step of the IE analysis, frequency estimates were developed using generic data sources, augmented by available design-specific information and fault tree modeling where applicable. Section 4.3 of NAT-7127 provides detailed descriptions of the calculation methodologies and underlying bases used to determine frequencies for each IE.

The staff also observed areas where the failure frequencies selected for SSCs from reference data had limited bases. For example, data supporting values associated with SPS cell and lifting devices were not necessarily applicable to KU1 design. In other cases, values used were the minimum value for a large range of applicable data, without fully accounting for the level of uncertainty.

The KU1 IE analysis also incorporated parametric and modeling uncertainties. Parametric uncertainty was characterized through probability distributions representing uncertainty in input parameters, with distribution types and associated uncertainty parameters sourced from reference literature when available. In the absence of such information, assumptions were made regarding distribution types and uncertainty parameters. Modeling uncertainties were addressed through documented assumptions necessitated by the lack of detailed operational data at the CP stage.

USO initially identified ((

)) potential IEs as shown in table 4.1-1 of NAT-7127. Following a systematic grouping and screening process, this set was consolidated into approximately

((

)) combined IEs, summarized in table 5-1 of NAT-7127, including their frequency estimates, distribution types, and associated variance or error factors. The staff also audited the assumptions documented in table 3.3-1 of NAT-7127 and the sources of modeling and completeness uncertainty associated with the assumptions in the IE analysis, which are detailed in table 4.4-1 of NAT-7127. The staffs audit confirmed that USOs approach to IE analysis and the process used to identify IEs align with the guidance of RG 1.247, Appendix B.

OFFICIAL USE ONLY PROPRIETARY INFORMATION 11 OFFICIAL USE ONLY - PROPRIETARY INFORMATION c) Event Sequence Analysis The KU1 event sequence (ES) analysis, which encompasses both at-power and LPSD conditions, is described in PSAR sections 3.1.1.8 and 3.1.1.14. The ES analysis integrates POSs, IEs, safety functions, and the success or failure of SSCs, culminating in end states that may involve potential release of radioactive material. USO performed the ES modeling and quantification using the EPRI Phoenix Architect 2.0, which includes CAFTA, PRAQuant, UNCERT, and FRANX, and FTREX Version 1.8.

To further understand the description in the PSAR, the staff reviewed NAT-7154, Rev. A, Natrium PRA Event Sequence Analysis in the audit. The review focused on confirming the characteristics and attributes of the ES analysis to ensure the analysis aligns with the non-LWR PRA standard ASME/ANS RA-S-1.4-2021 and RG 1.247. The audit examined the approaches and methodologies used in the ES analysis, ensuring they are reasonable, technically defensible, and consistent with established industry practices. The staff also examined associated assumptions to ensure the analysis reflects a level of detail suitable for the preliminary plant design.

The KU1 ES analysis, as documented in NAT-7154, Revision A, integrates POSs, IEs, safety functions, and the success or failure of SSCs, culminating in end states that may involve potential release of radioactive material. The primary output of the ES analysis is a set of event trees (ETs), which delineate possible event progressions. Each ET represents a time-independent, system-level response to a specific IE. The results provide essential inputs for plant risk quantification.

For the at-power ES analysis, each release category is defined by a quantified level of fuel barrier damage and the corresponding configuration of intact or failed containment release barriers. The ETs for fuel release events in the reactor vessel are constructed based on five critical barriers designed to mitigate radionuclide release from the reactor core:

1. Metal Fuel - Retains many radionuclides unless the fuel undergoes melting.

2. Fuel Cladding - Serves as a barrier to gaseous fission products and radionuclides.

3. Sodium Coolant - Acts as a third barrier by trapping radionuclides through plate-out or adsorption mechanisms; however, noble gases and other non-retained radionuclides may be released into the cover gas region above the sodium hot pool.

4. Reactor Vessel - Prevents fission gases that have reached the cover gas space from escaping the primary functional containment as a low leakage barrier. Settling of radionuclides is generally accounted for during holdup in the reactor vessel.

5. HAA - Functions as an additional and generally final barrier to radionuclide release.

The end states defined in the KU1 ES analysis encompass a range of potential plant conditions, accounting for the range of IEs. Radionuclide release scenarios, of varying magnitude, were identified and classified as either a safe and stable condition, referred to as the OK state, or a release to the environment. These end states primarily fall into two categories: short-term reactivity sequences and long-term decay heat removal sequences.

OFFICIAL USE ONLY PROPRIETARY INFORMATION 12 OFFICIAL USE ONLY - PROPRIETARY INFORMATION Short-term sequences begin with an IE followed by a failure to automatically scram the reactor.

The resulting end states from these sequences are detailed in table 3.1-2 of NAT-7154.

Conversely, long-term sequences involve a successful reactor scram, after which decay heat must be removed to prevent damage to the radionuclide release barriers. For the KU1 design, long-term decay heat removal is accomplished passively via the RAC system, which operates through natural convection. The end states associated with long-term sequences are summarized in table 3.1-3 of NAT-7154.

The KU1 design incorporates three fundamental safety functions (FSFs) essential for preventing damage to one or more radionuclide release barriers and mitigating radionuclide releases:

controlling heat generation; controlling heat removal; and retaining radionuclides.

All mitigating systems and features modeled in the PRA support one or more of these FSFs.

The KU1 ES analysis specifically evaluated failures of active reactivity control systems (e.g.,

control rods) and active heat removal systems (e.g., primary sodium, intermediate sodium, and forced air circulation via the intermediate air condenser (IAC)). When these systems perform as intended, the reactor promptly achieves a subcritical state, and decay heat is removed effectively, preventing significant increases in core or primary sodium temperature and thereby maintaining stable plant conditions.

A mission time of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> is assigned to event sequences where the reactor successfully scrams and primary sodium heat removal is accomplished through forced circulation utilizing the IAC. For sequences relying on passive decay heat removal systems, such as the RAC system or the passive IAC, the plant design accommodates these modes of operation. However, stable plant conditions may not be reached within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> due to the additional time required for decay heat to decrease below the removal capacity of these passive systems. To address this, a 72-hour extended mission time is assigned to such sequences, ensuring that the plant attains a safe and stable condition.

The plant design features credited in the success criteria were explicitly evaluated as part of the system analysis. Two distinct sets of ETs were developed for the KU1 PRA:

1. Fuel Damage ETs - These serve as the foundation for the ES analysis and were constructed based on the IE groups identified in the IE analysis.

2. Confinement ETs - These assess the integrity of radionuclide confinement systems throughout the progression of each ES.

The fuel damage ETs developed for the KU1 PRA encompass a range of scenarios, including but not limited to:

((

OFFICIAL USE ONLY PROPRIETARY INFORMATION 13 OFFICIAL USE ONLY - PROPRIETARY INFORMATION

)).

Confinement ETs were developed to evaluate release categories associated with these fuel damage sequences. A total of ((

)) confinement ETs were constructed as part of the KU1 PRA. Detailed descriptions of each ET, including their underlying logic, are provided in section 3.1.4.3 and Appendix A of NAT-7154. Each ET node represents a top event corresponding to a specific mitigating function. In total, ((

)) mitigating functions were defined and integrated within the confinement ET structure. These include:

((

))

The top events listed above are described in detail in section 3.2 of NAT-7154.

OFFICIAL USE ONLY PROPRIETARY INFORMATION 14 OFFICIAL USE ONLY - PROPRIETARY INFORMATION Section 3.3 of NAT-7154 presents the sequences generated from each ET. For each ET, the sequences are thoroughly described, including the success or failure of top events, to illustrate the progression of events. Section 3.4 of NAT-7154 documents the outcomes of the confinement ETs, specifically those sequences that culminate in a defined release category.

Section 3.5 of NAT-7154 provides the ET models applicable to LPSD conditions. Generally, the success criteria for the LPSD PRA align with those established for the at-power PRA. During LPSD operations, spent fuel may reside within the reactor vessel, the EVHM, or the EVST.

In the LPSD PRA, unprotected end states are not considered because the reactor is either shut down with all control rods inserted or operating at sufficiently low power levels where core damage cannot occur before IRF mechanisms effectively limit power. However, short-term damage scenarios are evaluated, for example, the accidental drop of a fuel assembly during core alterations could compromise multiple radionuclide barriers, potentially leading to an immediate release of radioactive material.

Table 3.5-4 of NAT-7154 provides an overview of the systems and features supporting FSFs during LPSD conditions. Consistent with the at-power ET analysis, two sets of ETs were developed for the LPSD PRA: fuel damage ETs and confinement ETs. The developed LPSD fuel damage ETs include the following:

((

)).

For certain events, such as loss of offsite power, multiple plant areas (i.e., the reactor vessel, EVHM, and EVST) may be affected in distinct ways. For these cases, separate ETs were developed for each impacted area to reflect their unique responses. These include:

((

)).

OFFICIAL USE ONLY PROPRIETARY INFORMATION 15 OFFICIAL USE ONLY - PROPRIETARY INFORMATION A single LPSD confinement ET, ((

)), was developed to evaluate release categories for LPSD fuel damage sequences. The thermal-mechanical loads specific to each sequence were characterized as either short-term challenges, primarily arising from load-drop events that cause immediate fuel cladding damage, or long-term heat challenges caused by decay heat removal mismatches.

The top events in this ET are summarized in table 3.6-1 of NAT-7154, with the corresponding functional logic detailed in table 3.6-2. Further descriptions of the sequences generated by each LPSD ET are provided in section 3.7 of NAT-7154 and the confinement ET logic is documented in Appendix A of NAT-7154.

The staff audited the assumptions supporting the KU1 ES analysis, as documented in table 2.2-1 of NAT-7154. The staff also reviewed the sources of modeling and completeness uncertainty related to the ES analysis, as identified in table 4.0-1 of NAT-7154. The staff notes that while uncertainty judgments appear reasonable, given the preliminary nature of the KU1 design, additional refinements are anticipated by the OL stage.

d) Success Criteria Analysis PSAR section 3.1.1.9 describes the CP applications success criteria (SC) analysis. To further understand the description in the PSAR, the staff reviewed NAT-7580, Rev. A, Natrium PRA Success Criteria Analysis in the audit. The review focused on confirming the characteristics and attributes of the SC analysis to ensure the analysis aligns with the non-LWR PRA standard ASME/ANS RA-S-1.4-2021 and RG 1.247. The staff reviewed the SC analysis approach, confirming general reasonableness and that the assumptions are appropriate, and the analysis aligns with the preliminary plant design.

The KU1 SC analysis covers both active and passive systems modeled in the PRA that contribute to maintaining the integrity of radionuclide release barriers. These barriers include the metal fuel matrix, fuel cladding, sodium coolant, reactor vessel, and confinement for the at-power analysis, as well as the EVHM and EVST pressure boundaries for the LPSD analysis.

For active systems, SC were established using conventional reliability analysis techniques, with system performance expectations and failure modes derived from design specifications documented in the system analysis. The SC evaluation for the KU1 design utilized the SAS4A/SASSYS-1 system transient analysis code.

SC were based on the design features of the passive components. The analysis considered the IEs that passive systems are intended to mitigate, assessing the likelihood of successful performance during relevant event sequences.

The staff reviewed the assumptions underlying the KU1 PRA SC analysis, as detailed in table 1 of NAT-7580. The staff expects USO to reassess these assumptions at the OL stage, using finalized design information, and either confirm their continued validity or update the SC analysis accordingly. Similarly, the staff audited the sources of modeling and completeness uncertainty associated with the SC assumptions, identified in table 8 of NAT-7580, and expects these to be confirmed or updated at the OL stage as well.

OFFICIAL USE ONLY PROPRIETARY INFORMATION 16 OFFICIAL USE ONLY - PROPRIETARY INFORMATION e) Systems Analysis PSAR section 3.1.1.10 describes the CP applications systems analysis (SY) analysis. To further understand the description in the PSAR, the staff reviewed the following PRA system notebooks during the audit:

NAT-7137, Revision A, Natrium PRA System Analysis - Containment; NAT-7138, Revision A, Natrium PRA System Analysis - Control (RPS/NIC);

NAT-7139, Revision A, Natrium PRA System Analysis - Electrical; NAT-7140, Revision A, Natrium PRA System Analysis - Ex-Vessel Fuel Handling Machine (EVHM);

NAT-7141, Revision A, Natrium PRA System Analysis - Ex-Vessel Storage Tank (EVST) Cooling; NAT-7142, Revision A, Natrium PRA System Analysis - Gaseous Radwaste Processing (RWG);

NAT-7143, Revision A, Natrium PRA System Analysis - Inherent Reactivity Feedback Feature (IRF);

NAT-7144, Revision A, Natrium PRA System Analysis - Intermediate Air Cooling (IAC);

NAT-7145, Revision A, Natrium PRA System Analysis - Intermediate Heat Transport System (IHT);

NAT-7146, Revision A, Natrium PRA System Analysis - Miscellaneous; NAT-7147, Revision A, Natrium PRA System Analysis - Molten Salt and Steam Generation System (MSS/SGS);

NAT-7148, Revision A, Natrium PRA System Analysis - Nuclear Island (NI) Air and Inert Gas Distribution System (NGS);

NAT-7149, Revision A, Natrium PRA System Analysis - Primary Heat Transport (PHT);

NAT-7150, Revision A, Natrium PRA System Analysis - Reactor Control (CRD);

NAT-7151, Revision A, Natrium PRA System Analysis - Reactor Air Cooling (RAC);

NAT-7152, Revision A, Natrium PRA System Analysis - Sodium Cover Gas (SCG); and NAT-7153, Revision A, Natrium PRA System Analysis - Sodium Processing System (SPS).

The staff found the system notebooks were developed in a manner generally consistent with the current design information; however, they do not include the fault tree logics or the results from system quantifications, such as cutsets and importance rankings.

The staff reviewed the assumptions underpinning the KU1 PRA systems analysis, as documented within the respective system notebooks. Each system notebook contains a dedicated section that discusses sources of modeling and completeness uncertainty related to the assumptions for that system, accompanied by tables documenting these uncertainties.

The staff observed that where design information was unavailable, the associated assumptions made did not always lead to conservative results. For example, in some instances where additional SSCs or functions were likely needed in the final system design but design information was unavailable at the CP stage (e.g., HVAC connections on confinement cells),

these SSCs or functions were generally not included in the CP PRA. Because the additional SSCs provide additional ways in which the system can fail to perform its function and sometime

OFFICIAL USE ONLY PROPRIETARY INFORMATION 17 OFFICIAL USE ONLY - PROPRIETARY INFORMATION additional paths for radionuclide release, this may lead to non-conservative frequencies or consequences. The staff expect that updates to this analysis will be made when final design details are available enabling further staff review at the OL stage.

Human Reliability Analysis PSAR section 3.1.1.11 describes the CP applications human reliability analysis (HRA). To further understand the description in the PSAR, the staff reviewed NAT-7879, Rev. A, Natrium PRA Human Reliability (HR) Analysis in the audit. The review focused on confirming the characteristics and attributes of the HRA analysis to ensure the analysis aligns with the non-LWR PRA standard ASME/ANS RA-S-1.4-2021 and RG 1.247.

The KU1 HRA methodology is based on the ((

)).

The analysis was performed using the ((

)), which facilitates the identification of operator actions, evaluation of dependencies via cutset quantification, and detailed documentation in alignment with the non-LWR PRA standard ASME/ANSI RA-S-1.4-2021.

The KU1 HRA addressed three primary categories of human actions:

Type A: Pre-IE human actions; Type B: IEs caused by human actions; and Type C: Post-IE human actions.

The human failure events, along with their estimated probabilities, are documented in table 4.1.1, table 4.2.1, and table 4.3.1 of NAT-7879. These tables identify Type-A and Type-C operator actions, but no Type-B operator actions were found. By committing to conform to the non-LWR PRA standard RA-S-1.4-2021, the applicant is expected to reassess Type-A and Type-C operator actions and to fully identify and evaluate Type-B operator actions at the OL stage.

The staff reviewed the assumptions supporting the KU1 HRA, as documented in table 1 of NAT-7879. The staff also reviewed the identified sources of modeling and completeness uncertainty related to the HRA assumptions, as presented in table 3.5.1 of NAT-7879. The staff expect that updates to HRA analysis, including assumptions and uncertainties, will be made as the design is finalized enabling further staff review at the OL stage.

f) Data Analysis PSAR section 3.1.1.12 describes the CP applications data analysis (DA). To further understand the description in the PSAR, the staff reviewed NAT-7878, Rev. A, Natrium PRA Data Analysis. The staff review focused on confirming the characteristics and attributes of the DA analysis to ensure the analysis aligns with the non-LWR PRA standard ASME/ANS RA-S-1.4-2021 and RG 1.247.

The DA and parameter estimation for the KU1 PRA are documented in NAT-7878, Revision A.

The KU1 DA included both demand-based and time-based failure rates, depending on the

OFFICIAL USE ONLY PROPRIETARY INFORMATION 18 OFFICIAL USE ONLY - PROPRIETARY INFORMATION component type and failure mode. For the auxiliary system components, the underlying database was developed using data and guidance from NUREG/CR-6928. When data were unavailable from this source for specific failure modes, additional references were consulted, including:

((

))

For sodium and molten-salt system components, reliability data was selected using the following data sources:

((

)).

The staffs audit found that some failure probabilities used did not have sufficient basis or were non-conservative. It is the staffs expectation that additional justification for the selection of failure probabilities will be provided at the OL, particularly in areas where the frequency impacts the licensing basis event or SSC classifications.

The KU1 database was developed in the CAFTA database format, incorporating code definitions and standardized basic event nomenclature. This structure enables automatic linkage of basic events to type codes within CAFTA, facilitating fault tree (FT) quantification.

Parametric uncertainty was addressed using reference sources such as NUREG/CR-6928.

Beta, gamma, and lognormal probability distributions were applied to model uncertainty, i.e.,

beta distributions for demand-type events (e.g., failure to start or open) and gamma distributions for time-dependent events (e.g., fail-to-run scenarios).

The KU1 PRA considered four types of dependencies: plant-level functional dependencies, intersystem dependencies, human action dependencies, and intra-system dependencies.

Common cause failure (CCF) analysis was based primarily on the methodology outlined in NUREG/CR-6268. The ((

)). At this stage, CCF modeling is limited to individual systems; however, USO indicated that intersystem dependencies will be incorporated as the PRA progresses, considering factors such as similarities in design, operation, maintenance, and environment.

OFFICIAL USE ONLY PROPRIETARY INFORMATION 19 OFFICIAL USE ONLY - PROPRIETARY INFORMATION The staff audited the assumptions underlying the KU1 PRA data analysis, as presented in table 1 of NAT-7878. The staff reviewed the sources of modeling and completeness uncertainty associated with these assumptions, as detailed in table 10 of NAT-7878, and expects them to be confirmed or updated at the OL stage.

g) Hazard Screening PSAR section 3.1.1.5 describes the CP applications hazard screening (HS) analysis. To further understand the description in the PSAR, the staff reviewed NAT-8294, Natrium PRA Screening of External Hazards. The audit review focused on confirming the characteristics and attributes of the HS analysis to ensure the analysis aligns with the non-LWR PRA standard ASME/ANS RAS1.42021. USO identified potential hazards and evaluated each for potential inclusion in the PRA. The hazards were assessed using the qualitative or quantitative screening criteria outlined in the non-LWR PRA standard ASME/ANS RA-S-1.4-2021.

In NAT-8294, USO identified 125 potential hazards and evaluated each for potential inclusion in the PRA. The hazards were assessed using qualitative or quantitative screening criteria, consistent with the guidance in the non-LWR PRA standard ASME/ANS RA-S-1.4-2021.

Hazards were qualitatively screened out if conservatively bounding evaluations demonstrated they would not impact the KU1 plant. Hazards not eliminated through qualitative screening were subject to quantitative evaluation. Quantitative screening excluded hazards if their mean frequency of occurrence was conservatively estimated to be less than 1x10-7 per plant-year.

Table 3.1-2 of the KU1 PSAR identifies internal and external hazards and indicates whether they were screened using qualitative or quantitative criteria. Table 3.1-3 of the KU1 PSAR lists the hazards retained within the PRA scope. Table 6-1 of NAT-8294 provides detailed information for each hazard, including a description, screening method, and the technical basis for the screening decision. Hazards that could not be screened out were retained for further PRA development.

The staff audited the assumptions supporting the KU1 HS analysis, as documented in chapter 4 of NAT-8294. The staff assessed the sources of modeling and completeness uncertainty discussed in section 6.5 of NAT-8294. The staff expect that updates to HS analysis, including assumptions and uncertainties, will be made as the design is finalized enabling further staff review at the OL stage.

h) Event Sequence Quantification Analysis PSAR sections 3.1.1.13 and 3.1.1.14 describe the CP applications event sequence quantification (ESQ) analysis. To further understand the description in the PSAR, the staff reviewed NAT-7364, Rev. B, Event Sequence Quantification. The audit review focused on confirming the characteristics and attributes of the ESQ analysis to ensure the analysis aligns with the non-LWR PRA standard ASME/ANS RA-S-1.4-2021.

NAT-7364, Revision B, documents the ESQ analysis, which integrates event sequences, system models, HRA, and data to quantify the frequencies of each modeled event sequence and event sequence family. These quantified results serve as inputs to risk integration tasks, supporting evaluations against quantitative health objective (QHO) risk metrics. PRA

OFFICIAL USE ONLY PROPRIETARY INFORMATION 20 OFFICIAL USE ONLY - PROPRIETARY INFORMATION quantification was conducted using ((

)). Individual sequence results were aggregated for reporting and served as input for the LMP risk evaluation. Quantification results are expressed in terms of minimal cutsets representing the smallest combinations of failures or conditions that lead to radionuclide release.

The KU1 ESQ analysis utilized the EPRI CAFTA Architect 2.0 software for ET and FT modeling, the PRAQuant tool for model integration and quantification, and FTREX as the quantification engine for generating minimal cutsets.

USO conducted an internal review of the KU1 PRA event sequences to verify model accuracy.

Cutset reviews, documented in Appendix D of NAT-7364, focused on the following:

Verification of cutset correctness, ensuring proper integration of ES, system models, ETs, and fault trees; Evaluation of OK branch sequences to ensure consistency with the ES analysis; Review of cutsets to identify any potential understatements of risk; and Examination of bottom cutsets to uncover modeling issues not apparent among top contributors.

Truncation limits were applied on ((

)). To ensure convergence and minimize the risk of excluding significant contributors, truncation thresholds as low as ((

)) per year were used. Truncation results are summarized in table 4 of NAT-7364.

NAT-7364 also documents the quantification results, equipment and operator importance analyses, parametric uncertainty analyses, and associated design insights. Appendices A and B of NAT-7364 provide detailed results for release category frequencies and importance measures, respectively.

The staff audited the quantification-specific assumptions documented in chapter 4 of NAT-7364.

Since ESQ analysis brings the other components of event frequency together, its completeness and quality is limited by the completeness and quality of the pieces of the other components.

The preliminary nature of the design introduced limitations in some other PRA elements, and those limitations will propagate into the ESQ.

i) Mechanistic Source Term Analysis PSAR sections 3.1.1.13 and 3.2 describe the CP applications mechanistic source term (MST) analysis. To further understand the description in the PSAR, the staff reviewed NAT-6161, PRA Source Term and Radiological Consequences in the audit. The KU1 MST analysis evaluates radionuclide release pathways from in-vessel reactor core events and from fuel assemblies during transport and storage in the spent fuel pool. It also addresses radionuclide transport within the plant and potential environmental release pathways. Additional evaluations that consider potential releases from supporting systems, including the SPS, SCG system, and gaseous radwaste system.

OFFICIAL USE ONLY PROPRIETARY INFORMATION 21 OFFICIAL USE ONLY - PROPRIETARY INFORMATION Multiple computational tools were employed to develop the source term estimates, including

((

)), and RADTRAD. Bounding source terms were calculated to account for uncertainties inherent in the source term estimation process.

The staff audited the assumptions supporting the KU1 MS analysis, as documented in table 5-2 of NAT-6161. The staff also assessed the identified sources of modeling and completeness uncertainty associated with the MS analysis assumptions, as documented in table 8-1 of NAT-6161. The staff expect that updates to MS analysis, including assumptions and uncertainties, will be made as the design is finalized enabling further staff review at the OL stage.

j) Radiological Consequence Analysis To further understand the radiological consequence (RC) analysis described in PSAR sections 3.1.1.13, 3.2, and 3.3, the staff further reviewed NAT-6161, PRA Source Term and Radiological Consequences. The review focused on confirming the characteristics and attributes of the RC analysis to ensure the analysis aligns with the non-LWR PRA standard ASME/ANS RA-S-1.4-2021 and RG 1.247.

The radiological consequences evaluated for each source term release category include:

Total effective dose equivalent (TEDE) to a receptor located at the exclusion area boundary (EAB)

Frequency of exceeding 100 mrem TEDE at the site boundary Average individual early fatality risk within one mile of the EAB Average individual latent cancer fatality risk within ten miles of the EAB.

The MACCS code was used to compute uncertainties associated with TEDE, early fatality risk, and latent cancer fatality risk. Both mean values and 95th percentile estimates were developed for the plant cumulative risk. For the TEDE at the EAB, mean 5th, and 95th percentile estimates were developed, based on analyses of nominal and bounding source terms to ensure a robust evaluation.

k) Risk Integration PSAR sections 3.1.1.13 and 3.1.1.14 describe the CP applications risk integration (RI) analysis and PSAR section 4.1 provides the overall integrated risk performance of the plant. To further understand the description in the PSAR, the staff reviewed NAT-7827, Risk Integration, and NAT-8807, Quantitative Health Objective Updates for the IDPP. The review focused on confirming the characteristics and attributes of the RI analysis to ensure the analysis aligns with the non-LWR PRA standard ASME/ANS RA-S-1.4-2021 and RG 1.247.

The KU1 integrated risk analysis combined ESQ, the spent fuel pool model, and the RC analysis into a unified, fully integrated model with a common top event. In accordance with Appendix A of RG 1.253, this revision of the KU1 RI analysis does not include external hazards such as internal fires, internal flooding, seismic events, or high winds.

OFFICIAL USE ONLY PROPRIETARY INFORMATION 22 OFFICIAL USE ONLY - PROPRIETARY INFORMATION The integrated model was developed using the CAFTA Architect 2.0 code and evaluated against three plant-level cumulative risk metrics that characterize public radiological risk. These metrics include:

Site boundary dose risk: The total mean frequency of exceeding a dose of 100 mrem at the site boundary from all LBEs must not exceed one per plant-year, consistent with the public dose limits in 10 CFR Part 20, Standards for Protection Against Radiation.

Early fatality risk at the EAB: The mean individual risk of early fatality within one mile of the EAB, based on the frequencies and consequences of all LBEs, must not exceed 5.0x10-7 per plant-year, consistent with the QHO for early fatality.

Latent cancer fatality risk: The mean individual risk of latent cancer fatality within ten miles of the EAB must not exceed 2.0x10-6 per plant-year, consistent with the QHO for latent cancer risk.

To ensure adequate resolution of low-frequency contributors, a truncation threshold of

((

)) per year was applied to each release top event. This level is at least five orders of magnitude below the lowest event frequency considered. For non-release top events, a truncation threshold of ((

)) per year was applied, consistent with the LMP methodology allowing the applicant to define an appropriate truncation threshold.

l) KU1 Spent Fuel PRA PSAR sections 3.1.1.3 and 3.1.1.6 describe the spent fuel (SF) PRA. To further understand the description in the PSAR, the staff reviewed NAT-7547, Rev. B, Scoping PRA Analysis for Spent Fuels from EVST to Spent Fuel Pool (SFP) and in the SFP, which documents the development and results of the SF PRA.

The analysis included IEs identification, event tree and system fault tree development, quantification, and uncertainty evaluation. Release category frequencies were estimated for a spectrum of potential event outcomes, ranging from cladding damage of a single fuel pin to large-scale fuel assembly damage caused by uncovering and overheating in the SFP.

The KU1 SF PRA considered potential IEs from fuel handling events outside of the reactor vessel and EVST, including events involving the bottom loading transfer cask, pin removal cell, pool immersion cell, cask loading pit, pool handling machine, and the SFP cooling (FPC) system. The SF PRA identified and evaluated plant systems and operator actions needed to mitigate damage to radionuclide barriers, should such barriers be challenged. Specifically, the analysis addressed risks to fuel integrity from removal from the EVST, through transfer, and subsequent storage in the SFP. The current scope of the KU1 SF PRA is limited to internal IEs, where the spent fuel handling activities from the reactor vessel to the EVST are addressed in the LPSD PRA.

The staff reviewed the assumptions used in the development of the KU1 SF PRA, as documented in table 4.1 of NAT-7547. The staff also reviewed the treatment of modeling and completeness uncertainties, as discussed in section 6.8.12 of NAT-7547.

OFFICIAL USE ONLY PROPRIETARY INFORMATION 23 OFFICIAL USE ONLY - PROPRIETARY INFORMATION As noted in NAT-7547, the SF PRA will be refined as additional design and operational details become available, allowing for a more accurate assessment of risks associated with spent fuel operations.

m) Risk Integration Sensitivities To further confirm that the KU1 CP stage PRA has been performed in accordance with RG 1.247 and RG 1.253 guidance, the staff reviewed NAT-8076, Rev. A, Risk Integration Sensitivities. This document presents a sensitivity analysis that examines the impact of key assumptions and model elements used in the RIs portion of the KU1 PRA.

Table 2 of NAT-8076 summarizes the principal assumptions made across the PRA elements currently included in the KU1 model, along with the rationale for screening out certain assumptions based on their minimal impact on overall PRA results. Table 3 of NAT-8076 provides detailed assessments of assumptions retained for evaluation in the sensitivity analysis.

Table 4 of NAT-8076 identifies assumptions or events that were not assessed further due to the need for model revisions or re-quantification. Some of these assumptions were identified as potentially significant; the staff expects these to be evaluated and available for review at the OL stage when final design details are available.

In response to staff questions regarding the sensitivity analyses performed, USO also conducted the sensitivity cases listed below and presented the results to the staff during the audit. The applicant clarified that the outcomes of these sensitivity runs did not affect any of the conclusions, results, or insights documented in support of the PSAR submittal.

The ((

)).

The staff found the performed sensitivity analyses useful in gaining insights into the influence of key assumptions and understanding the sensitivity of the KU1 PRA model.

n) PRA Self-assessment The staff also reviewed NAT-8218, Rev. B, Natrium PRA Self-Assessment Against ASME/ANS RA-S-1.4-2021 Standard, to confirm that the self-assessment was performed in accordance with RG 1.247 and RG 1.253, as noted in section 3.1.1.2 of the PSAR. The USO self-assessment encompassed the following internal events PRA elements:

POS Analysis IE Analysis ES Analysis Success Criteria Development (SC)

Systems Analysis (SY)

HRA DA HS Analysis

OFFICIAL USE ONLY PROPRIETARY INFORMATION 24 OFFICIAL USE ONLY - PROPRIETARY INFORMATION ESQ Analysis MST Analysis RC Analysis RI.

The following PRA elements were not developed at the CP stage and were therefore excluded from the self-assessment:

Internal Flood PRA (FL)

Internal Fire PRA (F)

Seismic PRA (S)

High Wind PRA (W)

External Flooding (XF)

Other External Hazards PRA (O).

Table 6-1 of NAT-8218 presents the overall results of the assessment. Based on its review of the information in NAT-8218 the staff identified the following with respect to the 1,233 PRA supporting requirements (PRA SRs) in the non-LWR PRA standard ASME/ANS RA-S-1.4-2021:

((

)) PRA supporting requirements were assessed as MET with Gap, indicating general compliance with the standard but with self-identified gaps that could potentially result in findings. These gaps are detailed in table 6-2 of NAT-8218.

((

)) PRA supporting requirements were assessed as Not Applicable, as they apply to operating plants and are not relevant at this design stage.

((

)) PRA supporting requirements were assessed as Not Met, primarily due to the absence of certain elements such as operator and maintenance procedures.

((

)) PRA supporting requirements were found to be not applicable to the KU1 design.

((

)) PRA supporting requirements were evaluated and found to be Met.

Comprehensive assessments for each supporting requirement are provided in appendices A through L of NAT-8218. Results from the KU1 self-assessment are summarized in the table below.

OFFICIAL USE ONLY PROPRIETARY INFORMATION 25 OFFICIAL USE ONLY - PROPRIETARY INFORMATION

((

))

The audit encompassed a thorough examination of KU1s PRA self-assessment results, including the grading of PRA supporting requirements, evaluation of gaps and their potential impact on the CP application and LMP application, and the justifications provided for acceptability.

During the audit, the staff engaged in detailed discussions on over 50 technical topics with USO personnel. These discussions addressed gaps identified through both KU1s self-assessment and the staffs safety review, focusing on the potential impact of each gap on risk-informed applications and USOs proposed actions to address gaps by the OL stage. Furthermore, USO provided archival documentation demonstrating the implementation of the KU1 PRA configuration control program.

The staffs review of the KU1 PRA self-assessment identified opportunities to enhance future assessment processes. The current self-assessment appeared to emphasize individual evaluations of PRA standard supporting requirements versus team-based analysis and coordinated resolution planning. A more collaborative approach could support a more comprehensive understanding of the PRA models and their technical foundations. In addition, portions of the assessment may have benefited from broader engagement of personnel with specialized PRA expertise to help ensure consistent interpretation and application of the supporting requirements.

Overall, the NRC staff observed that the KU1 PRA self-assessment was reasonably conducted and that the results were sufficiently documented. Consistent with guidance from RG 1.247 and RG 1.253, the staff observed that the self-assessment identified assumptions made in lieu of as-

OFFICIAL USE ONLY PROPRIETARY INFORMATION 26 OFFICIAL USE ONLY - PROPRIETARY INFORMATION built and as-operated design information, provided justification for Not Met PRA supporting requirements, and documented associated gaps to support future resolution efforts. USO stated that a PRA peer review will be conducted to support the OL stage PRA, which will replace the self-assessment and provide an additional level of confidence in the model and results.

o) Kemmerer Unit 1 PRA Configuration Control Program PSAR section 3.1.1.2 describes the KU1 PRA configuration control (CC) process. To further understand the program, the staff reviewed NAT-13718, Rev. 0, PRA Configuration Control Procedure, which defines the process for maintaining and managing the KU1 PRA model and its documentation This CC procedure addressed key elements consistent with the PRA standard high-level requirements of the non-LWR PRA standard ASME/ANS RA-S-1.4-2021, including:

Monitoring changes to plant design, operations, PRA technology, and relevant industry experience, as well as collecting updated performance data that may affect PRA inputs.

Maintaining and upgrading the PRA model to ensure alignment with the as-designed, as-to-be-built, or as-operated plant, depending on the plant development phase and intended use.

Evaluating the cumulative impact of pending changes on the performance of risk-informed applications.

Maintaining CC over computer codes and associated files used in PRA quantification.

Documenting the CC programs implementation to ensure traceability of all updates and modifications to the PRA model.

USO confirmed that the PRA is developed and maintained under CC in accordance with the USO Quality Assurance Program Description, as detailed in topical report TP-QA-PD-0001, and the Demonstration Project Quality Execution Plan, as detailed in topical report NATD-QA-PLAN-0001. During the design phase, USOs design change control procedure requires all deliverable modifications, that impact the design baseline, to be consolidated into a design change package, which undergoes a comprehensive review process to evaluate scope and impact thoroughly, which serves as input to the PRA update.

USO also employed its corrective action program to document and track improvements, corrective actions, and compensatory measures related to the PRA, including discrepancies between the design and the PRA model. This program also facilitated incorporation of relevant industry and operating experience into the PRA.

6. PRA Audit Conclusion Through the KU1 CP PRA audit, the staff gained a comprehensive understanding of the KU1 PRA, including its structure, modeling assumptions, quantification methods, and treatment of uncertainties. This audit supported the staffs safety evaluation of the KU1 CP application, confirming summary information provided in the PSAR regarding the KU1 CP stage PRA quality and its use to support risk-informed decision-making, including the implementation of the LMP, RIM, and D-RAP.