ML25119A009
| ML25119A009 | |
| Person / Time | |
|---|---|
| Issue date: | 05/09/2025 |
| From: | Dimitrijevic V Advisory Committee on Reactor Safeguards |
| To: | Walter Kirchner Advisory Committee on Reactor Safeguards |
| References | |
| Download: ML25119A009 (1) | |
Text
UNITED STATES NUCLEAR REGULATORY COMMISSION ADVISORY COMMITTEE ON REACTOR SAFEGUARDS WASHINGTON, DC 20555 - 0001 May 09, 2025 MEMORANDUM TO:
Walt Kirchner, Chair NuScale Subcommittee Advisory Committee on Reactor Safeguards FROM:
Vesna B Dimitrijevic, Member NuScale Subcommittee Advisory Committee on Reactor Safeguards
SUBJECT:
INPUT FOR ACRS REVIEW OF THE NUSCALE STANDARD DESIGN APPROVAL APPLICATION - SAFETY EVALUATION REPORT FOR CHAPTER 19, PROBABILISTIC RISK ASSESSMENT AND SEVERE ACCIDENT EVALUATION In response to the Subcommittees request, I have reviewed the chapter of the safety evaluation report (SER) which documents the U.S. Nuclear Regulatory Commission (NRC) staffs review of Chapter 19, Probabilistic Risk Assessment (PRA) and Severe Accident Evaluation, of the NuScale Power, LLC, Standard Design Approval Application (SDAA), Part 2, Final Safety Analysis Report (FSAR), for the US460 standard plant design.
The following ise a short summary of the subcommittee discussion on the subject and my recommended course of action concerning further review of this chapter and the staffs associated safety evaluation.
SER Summary The NuScale US460 design-specific PRA addressed the full scope of internal and external initiating events for both full-power and low power and shutdown conditions. Section 19.1 presents the Level 1 and Level 2 PRA results, together with additional topics such as PRA quality, design features to minimize risk, data, uncertainties, sensitivities, and risk insights. The PRA was performed for a single module and used to develop insights for multiple modules: quantitative risk insights for multi-module internal events and qualitative risk insights for multi-module shutdown and external events. Section 19.2 provided a description and analysis of design features for the prevention and mitigation of severe accidents.
The staffs review concludes that the PRA conforms to the guidance in Standard Review Plan (SRP) Section 19.0 and DC/COL-ISG-028 (for the applicable modes and hazards). Therefore, the staff finds that the PRA is of sufficient technical adequacy to support the SDAA. Based on the integrated risk from all modes and all hazards, the staff concludes that the Commissions core damage frequency (CDF) and large release frequency (LRF) goals have been met with high margin. The staff also concluded that the applicant addressed severe accidents consistent
W. Kirchner with Commission policy and that the SDAA design for containment performance meets the containment structural integrity criteria of Regulatory Guide 1.7 and the containment leak tight criteria of SECY-93-087.
Discussion An important part of the subcommittee discussion was focused on the design changes and their impact on the changes in the risk profile between the certified design and SDAA. The most noticeable difference between these two applications was a significant reduction in the LRF.
This change is a consequence of changes made to the emergency core cooling system (ECCS): changes in actuation criteria and the removal of the inadvertent actuation block valves, as well as the addition of the ventures to the chemical and volume control system (CVCS) lines.
These design changes have allowed mitigating unisolated CVCS line breaks outside of containment without requiring an operator action to add coolant.
Such simplified mitigation is due to the ability to actuate ECCS early and depressurize the system to atmospheric pressure to stop losing coolant and keeping the core covered without need for makeup. The CDF, due to these events, has become negligible and so does the LRF.
A similar change is valid for an unisolated steam generator tube break, which presents a less challenging event with a smaller flow area. Given that the CVCS line breaks outside containment, and to a smaller degree the steam generator tube failure, constituted the main contributors to the LRF in the certified design PRA, this change is primarily responsible for the dramatic reduction in the LRF.
I did not identify any major concerns in my review. The application was well documented, and the staffs evaluation thorough. The PRA model is developed in even more detail than could be expected from a SDAA PRA. However, I have identified possible deficiencies that could impact the PRA insights used to support the other programs. These are discussed below.
In addition to the containment, ultimate heat sink and module protection system, the only structures, systems, and components identified as risk significant are the ECCS components.
That questions a depth in defense in depth. The other important structures, systems, and components could be discovered if the relative risk importance measures (as compared to the CDF/LRF values) are also used, or the other importance related questions are considered. That is relevant when providing the PRA inputs to the operational programs (e.g., Maintenance Rule) or the operational requirements that support the design, inspection, construction, and operation of the plant (e.g., inspections, tests, analyses, and acceptance criteria; the reliability assurance program; and technical specifications, etc.).
To evaluate realistic uncertainty in the results, and the underlying mean value for the risk measures, the ECCS components with high risk importances should receive a detailed evaluation of uncertainties in the applied data, common cause assumptions, the passive heat transfer failure likelihood. It should be ensured that the failure distributions for the ECCS valves are not treated as independent, and, that the state of knowledge relationship is considered. It is puzzling that the uncertainty evaluation resulted in equal point estimate and mean values.
Similarly, the related sensitivity results, should be done in more detail and could be reflected in the uncertainty evaluations - like high sensitivity to the common cause failures done as a big lump.
Sensitivities are mostly calculated as single runs; the selected combination of sensitivities are not considered. The overall results could be very sensitive to underestimating multiple factors.
W. Kirchner For example, it would be interesting to combine sensitivity to the steam generator tube failure initiating frequency, with sensitivity to assumptions of single tube failure on a single steam generator.
Recommendation As lead reviewer for NuScale Chapter 19, I recommend that the Committee not perform any additional review of this chapter.
As the plant-specific PRA model develops further, improvements in the evaluation of the risk importance, uncertainty and sensitivity should be considered to provide realistic PRA inputs to the operational programs and requirements.
References
- 1. U. S. Nuclear Regulatory Commission, Safety Evaluation of NuScale Standard Design Approval Chapter 19, Probabilistic Risk Assessment and Severe Accident Evaluation, February 12, 2025 (ADAMS Accession Nos. ML25037A169 (Public) ML24352A361 (Non-Public)).
- 2. NuScale Power, LLC, Standard Design Approval Application, Part 2, Chapter 19, Probabilistic Risk Assessment and Severe Accident Evaluation, October 31, 2023 (ADAMS Accession Nos. ML23304A385 (Public) and ML23304A386 (Non-Public)).
- 3. U. S. Nuclear Regulatory Commission, DC/COL-ISG-028, Assessing the Technical Adequacy of the Advanced Light-Water Reactor Probabilistic Risk Assessment for the Design Certification Application and Combined License Application, November 22, 2016 (ADAMS Accession No. ML16130A468)
- 4. U. S. Nuclear Regulatory Commission, Regulatory Guide 1.7, Control of Combustible Gas Concentrations in Containment, Revision 3, March 23, 2007 (ADAMS Accession No. ML070290080).
- 5. U. S. Nuclear Regulatory Commission, SECY-93-0087, Policy, Technical, and Licensing Issues Pertaining to Evolutionary and Advanced Light-Water Reactor Designs, April 21, 1993 (ADAMS Accession No. ML003708021).
W. Kirchner May 09, 2025
SUBJECT:
INPUT FOR ACRS REVIEW OF THE NUSCALE STANDARD DESIGN APPROVAL APPLICATION - SAFETY EVALUATION REPORT FOR CHAPTER 19, PROBABILISTIC RISK ASSESSMENT AND SEVERE ACCIDENT EVALUATION Package Accession No: ML25091A091 Accession No: ML25119A009 Publicly Available (Y/N): Y Sensitive (Y/N): N If Sensitive, which category?
Viewing Rights:
NRC Users or ACRS only or See restricted distribution OFFICE ACRS SUNSI Review ACRS ACRS NAME MSnodderly MSnodderly LBurkhart V Dimitrijevic DATE 4/28/25 4/28/25 4/28/25 5/09/25 OFFICIAL RECORD COPY