ML23094A179
| ML23094A179 | |
| Person / Time | |
|---|---|
| Site: | Limerick  |
| Issue date: | 05/17/2023 |
| From: | Audrey Klett Plant Licensing Branch 1 |
| To: | Rhoades D Constellation Energy Generation |
| Sreenivas V, NRR/DORL/LPLI, 415-2597 | |
| References | |
| EPID L-2021-LLA-0042 | |
| Download: ML23094A179 (1) | |
Text
Enclosure 4 May 17, 2023 SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION FOR DENIAL OF PROPOSED ALTERNATIVE DEFENSE-IN-DEPTH AND PRESSURE BOUNDARY COMPONENT CATEGORIZATION PROCESSES FOR 10 CFR 50.69 CONSTELLATION ENERGY GENERATION, LLC LIMERICK GENERATING STATION, UNITS 1 AND 2 DOCKET NOS. 50-352 AND 50-353 Contents
1.0 INTRODUCTION
................................................................................................................ 1 1.1 Background.................................................................................................................... 1 1.2 Description of 10 CFR 50.69 Licensing Basis................................................................ 1 1.3 Proposed Changes......................................................................................................... 3
2.0 REGULATORY EVALUATION
........................................................................................... 3 2.1 Regulations..................................................................................................................... 4 2.2 Licensing Basis............................................................................................................... 4 2.3 Guidance........................................................................................................................ 4
3.0 TECHNICAL EVALUATION
............................................................................................... 5 3.1 Method of Review........................................................................................................... 5 3.2 Evaluation of Proposed Defense-in-Depth Categorization Approach............................. 6 3.2.1 Proposed Methodology........................................................................................... 6 3.2.2 NRC Staff Review of Alternate DID Approach......................................................... 6 3.2.3 Inconsistent and Lack of Correlation to Layers of Defense..................................... 7 3.2.4 Inappropriate Compensation for Reduced Rigor..................................................... 8 3.2.5 Insufficient Basis for Treatment of Common Cause Failures.................................. 9 3.2.6 Unjustified Mixed Use of DID Processes............................................................... 10 3.2.7 Conclusion for Proposed Defense-in-Depth Categorization Approach................. 10 3.3 Evaluation of Proposed Pressure Boundary Categorization Approach........................ 10 3.3.1 Proposed Methodology......................................................................................... 10 3.3.2 NRC Staff Review of Alternate Pressure Boundary Approach.............................. 11 3.3.3 Insufficient Technical Basis for Proposed Methodology........................................ 11 3.3.4 Applicability of RI-ISI, RI-RRA, and Previous LARs.............................................. 12 3.3.5 Evaluation of Pilot Study Results.......................................................................... 14 3.3.6 Inadequacy of Quantitative Risk-Based Criteria (Criteria 11, 12, and 13)............. 15 3.3.7 Uncertainties and Changes in Pipe Break Frequencies........................................ 17 3.3.8 Conclusion for Proposed Pressure Boundary Categorization Approach............... 18 3.4 Proposed Changes to License Conditions................................................................... 18
4.0 CONCLUSION
................................................................................................................. 18
5.0 REFERENCES
................................................................................................................. 19 6.0 ABBREVIATIONS............................................................................................................ 22 7.0 PRINCIPAL CONTRIBUTORS........................................................................................ 22
1.0 INTRODUCTION
1.1 Background
By license amendment request (LAR) dated March 11, 2021 [1], as supplemented by letters dated May 5, 2021 [2], December 15, 2021 [3], February 14, 2022 [4], and June 30, 2022 [5],
Exelon Generation Company, LLC (the licensee)1 requested amendments to Renewed Facility Operating License Nos. NPF-39 and NPF-85 for Limerick Generating Station (Limerick), Units 1 and 2, respectively. The licensee requested to revise the license condition in each license pertaining to the licensees approval to use Title 10 of the Code of Federal Regulations (10 CFR), section 50.69, Risk-informed categorization and treatment of structures, systems and components [SSCs] for nuclear power reactors, to allow the use of alternative defense-in-depth (DID), pressure boundary, and seismic processes (or approaches) for categorizing SSCs in the licensees application of 10 CFR 50.69.
This safety evaluation addresses the licensees proposal to use alternative DID and pressure boundary approaches to categorizing SSCs in the licensees application of 10 CFR 50.69.
enclosure 3 to the U.S. Nuclear Regulatory Commissions (NRCs or the Commissions) letter dated May 17, 2023 [6], addresses the NRC staffs review of the licensees proposed alternative seismic approach.
The NRC staff performed an audit in November 2021, to support its review of the LAR. The NRC staffs audit plan is dated October 1, 2021 [7], and was supplemented by emails dated October 20, 2021 [8], January 24, 2022 [9], and February 2, 2022 [10]. The NRC staffs audit summary report is dated March 24, 2023 [11]. By emails dated April 20, 2021 [12], and May 13, 2022 [13], the NRC staff requested additional information from the licensee. The NRC staff held public meetings with the licensee on May 3, 2022 [14], June 24, 2022 [15], and February 23, 2023 [16], to discuss its review of the LAR. The licensee responded to NRC staffs requests and audit discussions by letters dated May 5, 2021 [2], December 15, 2021 [3], February 14, 2022 [4], and June 30, 2022 [5].
On August 10, 2021, the NRC staff published a proposed no significant hazards consideration (NSHC) determination in the Federal Register [17] for the proposed amendments.
Subsequently, by letter dated December 15, 2021 [3], as supplemented by letter dated February 14, 2022 [4], the licensee provided additional information that expanded the scope of the amendment request as originally noticed in the Federal Register. Accordingly, the NRC published a second proposed NSHC determination in the Federal Register on February 22, 2022 [18], as corrected by notice dated March 18, 2022 [19]. The supplement dated June 30, 2022 [5], provided additional information that clarified the LAR, did not expand the scope of the LAR, and did not change the NRC staffs revised proposed NSHC determination, as published in the Federal Register on February 22, 2022 [18], as corrected by notice dated March 18, 2022 [19].
1.2 Description of 10 CFR 50.69 Licensing Basis On July 31, 2018, the NRC issued Amendment Nos. 230 and 193 [20] for Limerick, Units 1 and 2, respectively, which added a new license condition in appendix C of the operating licenses that allowed the licensee to implement 10 CFR 50.69. The provisions of 10 CFR 50.69 1
On February 1, 2022, Exelon Generation Company, LLC was renamed Constellation Energy Generation, LLC.
2 allow adjustment of the scope of SSCs subject to special treatment requirements (e.g., quality assurance, testing, inspection, condition monitoring, assessment, and evaluation), based on a method of categorizing SSCs into one of four risk-informed safety class (RISC) categories, according to their safety significance. For SSCs categorized as low safety-significant (LSS),
alternative treatment requirements may be implemented in accordance with the regulation. The licensee determines safety significance using an integrated decision-making process that uses both risk insights and traditional engineering insights. The licensee conducts periodic assessment activities to adjust the categorization or treatment processes, as needed, so that SSCs continue to meet all applicable functional requirements.
Under 10 CFR 50.69, a licensee classifies SSCs as having either safety-significant functions (RISC-1 and RISC-2 categories) 2 or LSS functions (RISC-3 and RISC-4 categories). For high safety-significant (HSS) SSCs, 10 CFR 50.69 maintains the current regulatory requirements (i.e., it does not change requirements for these SSCs) for special treatment. For LSS SSCs, licensees can implement alternative treatment requirements in accordance with 10 CFR 50.69(b)(1) and 10 CFR 50.69(d)(2). For RISC-3 SSCs, licensees can replace special treatment with an alternative treatment. For RISC-4 SSCs, 10 CFR 50.69 does not impose new treatment requirements, and RISC-4 SSCs are removed from the scope of any applicable special treatment requirements identified in 10 CFR 50.69(b)(1).
As discussed in the NRC staffs safety evaluation for Amendment Nos. 230 and 193 [20], the licensee currently performs its DID characterization in accordance with section 6, Defense-in-Depth Assessment, of Nuclear Energy Institute (NEI) 00-04, Revision 0, 10 CFR 50.69 SSC Categorization Guideline (NEI 00-04) [21], which has guidance for addressing 10 CFR 50.69(c)(1)(iii). Section 6 of NEI 00-04 [21] states, in part:
In cases where the component is safety-related and found to be of low risk significance, it is appropriate to confirm that defense-in-depth is preserved. This discussion should include consideration of the events mitigated, the functions performed, the other systems that support those functions and the complement of other plant capabilities that can be relied upon to prevent core damage and large, early release.
As discussed in the NRC staffs safety evaluation for Amendment Nos. 230 and 193 [20], the licensee currently uses the Arkansas Nuclear One, Unit 2 (ANO-2) passive categorization method to assess passive component risk for Class 2 and Class 3 SSCs and their associated supports because passive components are not modeled in the probabilistic risk assessment (PRA) [22]. This method addresses those components that have only a pressure-retaining function and the passive function of active components. The safety evaluation [20], states:
The ANO-2 methodology is a risk-informed safety classification and treatment program for repair/replacement activities for Class 2 and Class 3 pressure-retaining items and their associated supports (exclusive of Class CC and MC items), using a modification of the ASME [American Society of Mechanical Engineers] Code Case N-660, Risk-Informed Safety Classification 2
Nuclear Energy Institute (NEI) 00-04, Revision 0, 10 CFR 50.69 SSC Categorization Guideline [21],
uses the term high-safety-significant (HSS) to refer to SSCs that perform safety-significant functions. The NRC staff understands HSS, as used in NEI 00-04, to have the same meaning as safety-significant (i.e., SSCs that are categorized as RISC-1 or RISC-2), as used in 10 CFR 50.69.
3 for Use in Risk-Informed Repair/Replacement Activities,Section XI, Division 1
. The ANO-2 methodology relies on the conditional core damage and large early release probabilities associated with pipe ruptures. Safety-significance is generally measured by the frequency and the consequence of, in this case, pipe ruptures. Treatment requirements (including repair/replacement) only affect the frequency of passive component failure. Categorizing solely based on consequences, which measures the safety-significance of the pipe given that it ruptures, is conservative compared to including the rupture frequency in the categorization. The categorization will not be affected by changes in frequency arising from changes to the treatment.
1.3 Proposed Changes In its LAR [1], as supplemented [2], the licensee indicated that several 10 CFR 50.69 categorization processes are overly conservative and resource intensive and do not provide a commensurate benefit to the health and safety of the public. To address this, the licensee proposed to revise the 10 CFR 50.69 license condition in each license to allow the use of alternative DID and pressure boundary categorization approaches,3 which are described in section 3 of this safety evaluation.
In its LAR [1], as supplemented [2] [3] [4] [5], the licensee stated that it completed the implementation items required by the license conditions prior to the implementation of the 10 CFR 50.69 categorization process, which began in October 2018. Therefore, the licensee proposed to delete the current paragraph specific to the implementation items that states, in part, Exelon will complete the implementation items... prior to implementation of the 10 CFR 50.69 categorization process, and replace it with a new insert paragraph that states:
In addition, Constellation Energy Generation, LLC (CEG) is approved to implement 10 CFR 50.69 using any of the following alternative processes for categorization of RISC-1, RISC-2, RISC-3, and RISC-4 SSCs as specified in Unit [1(2)] License Amendment No. [261(223)] dated [May 17, 2023]:
the alternative defense-in-depth approach as described in the licensees letters dated March 11, 2021, May 5, 2021, and June 30, 2022.
the alternative passive pressure boundary categorization approach as described in the licensees letters dated March 11, 2021, and June 30, 2022.
the alternative seismic approach as described in the licensees letters dated December 15, 2021, and February 14, 2022.
2.0 REGULATORY EVALUATION
The NRC staff considered the following regulations, licensing basis, and guidance during its review of the proposed changes.
3 The LAR [1] states, When implementing the 10 CFR 50.69 categorization process, pressure boundary components are those components that perform a pressure retaining function. This was previously referred to as passive categorization in section 3.5.4 of the NRC Safety Evaluation that was issued for Limerick Units 1 and 2 to implement 10 CFR 50.69.
4 2.1 Regulations Section 50.69 of 10 CFR provides an alternative approach for establishing requirements for treatment of SSCs for nuclear power reactors using a risk-informed method of categorizing SSCs according to their safety significance. Specifically, for SSCs categorized as LSS, alternative treatment requirements may be implemented in accordance with the regulation. For SSCs determined to be of HSS, requirements may not be changed. The corresponding statement of considerations (SoC) for the rulemaking are in the Federal Register notice published on November 22, 2004 [23].
Paragraph 50.69(b)(3) of 10 CFR states that the Commission will approve a licensees implementation of this section by issuance of a license amendment if the Commission determines that the categorization process satisfies the requirements of 10 CFR 50.69(c).
Paragraph 50.69(c)(1) of 10 CFR states, in part:
SSCs must be categorized as RISC-1, RISC-2, RISC-3, or RISC-4 SSCs using a categorization process that determines if an SSC performs one or more safety significant functions and identifies those functions.
The process must:
(ii) 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. The functions to be identified and considered include design bases functions and functions credited for mitigation and prevention of severe accidents. 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.
(iii) Maintain defense-in-depth.
(iv) Include evaluations that provide reasonable confidence that for SSCs categorized as RISC-3, sufficient safety margins are maintained and that any potential increases in core damage frequency (CDF) and large early release frequency (LERF) resulting from changes in treatment permitted by implementation of § § 50.69(b)(1) and (d)(2) are small.
(v) Be performed for entire systems and structures, not for selected components within a system or structure.
2.2 Licensing Basis The NRC issued Amendment Nos. 230 and 193 [20] for Limerick Generating Station, Units 1 and 2, regarding adoption of 10 CFR 50.69.
2.3 Guidance NRC Regulatory Guide (RG) 1.174, Revision 3, An Approach for Using Probabilistic Risk Assessment in Risk-Informed Decisions on Plant-Specific Changes to the Licensing Basis (RG 1.174) [24], provides guidance on the use of PRA findings and risk insights in
5 support of changes to a plants licensing basis. This RG provides risk acceptance guidelines that are acceptable for evaluating the results of such evaluations. The SoC for the 10 CFR 50.69 rulemaking [23] discusses how RG 1.174 influenced the development of the rule. For example, section III.7.0 of the SoC for the 10 CFR 50.69 rulemaking [23]
states:
The Commission concludes that § 50.69 provides reasonable assurance of adequate protection of public health and safety because the principles listed below were used in the development of § 50.69 and because these principles will continue to be employed in the NRC's continuing regulatory oversight of § 50.69 implementation. Those principles are:
(a) Reasonable confidence that the net increase in plant risk is small; (b) Defense-in-depth is maintained; (c) Reasonable confidence that safety margins are maintained; and (d) Monitoring and performance assessment strategies are used.
These principles were established in RG 1.174, which provided guidance on an acceptable approach to risk-informed decision-making consistent with the 1995 Commission policy on the use of PRA. Section 50.69 was developed to incorporate these principles, both to ensure consistency with Commission policy, and to ensure that the rule maintains adequate protection of public health and safety.
NEI 00-04, Revision 0 (NEI 00-04) [21] describes a process for determining the safety significance of SSCs and categorizing them into the four RISC categories defined in 10 CFR 50.69.
3.0 TECHNICAL EVALUATION
3.1 Method of Review In determining whether an amendment to a license will be issued, the NRC is guided by the considerations that govern the issuance of initial licenses to the extent applicable and appropriate. The NRC staff evaluated the licensees LAR to determine whether the proposed changes are consistent with the regulations, licensing basis information, guidance, and precedents, as applicable, which are discussed in section 2 of this safety evaluation.
Paragraph 50.69(b)(3) of 10 CFR states that the Commission will approve a licensees implementation of 10 CFR 50.69 by issuing a license amendment if it determines that the licensees process for categorizing SSCs satisfies the requirements to 10 CFR 50.69(c). The NRC staff reviewed the proposed alternative DID and pressure boundary categorization approaches against the requirements in 10 CFR 50.69(c) to determine whether the proposed approaches support the proposed license condition changes and provides reasonable assurance of continued compliance with 10 CFR 50.69.
6 3.2 Evaluation of Proposed Defense-in-Depth Categorization Approach 3.2.1 Proposed Methodology The licensee proposed the DID approach described in the Pressurized Water Reactor Owners Group (PWROG) document PWROG-20015-NP [25], as an alternative method to the current NRC-approved DID categorization process described in section 6 of NEI 00-04 [21]. As described in section 2.2.2 of PWROG-20015-NP, Revision 3 (as revised in the supplement dated June 30, 2022 [5]), the licensee proposed to use an alternate DID approach for core damage DID in lieu of that described in section 6.1 of NEI 00-04 [21], while continuing to apply the guidance in section 6.2 of NEI 00-04 [21] for containment DID. PWROG-20015-NP, Revision 3 [5] indicates that the alternate DID process is used to identify whether SSCs that are identified as candidate LSS from other steps in the 10 CFR 50.69 categorization process remain candidate LSS after the DID assessment. It further indicates that the remainder of the system categorization steps need to be completed and that the integrated decision-making panel (IDP) has to review and approve the system categorization, prior to an SSC being assigned as LSS and assigned a RISC. Finally, the alternate core damage DID process initial screening would be completed in a single plant level analysis for all SSCs, with results incorporated into the system categorization.
The proposed alternate DID process would use the minimal cutset4 (accident sequence scenarios) results of the full power internal events PRA model at the plant level. Specifically, the process examines the minimal cutsets that result in core damage. The process would use a qualitative screening to filter for retention as HSS for those minimal cutset configurations consisting of an initiating event with either a single basic event cutset element for initiating event frequencies between 10-3 per year and 10-4 per year, or two basic event cutset elements for initiating event frequencies greater than 10-3 per year. The basic event cutset elements can represent an independent random failure of an SSC, common cause failure of multiple SSCs, or human failure events related to one or more SSCs. PWROG-20015-NP, Revision 3 [5] also describes a set of rules for cutset elements that are not counted toward this configuration, including items present in the PRA model for modeling convenience, such as flags and phenomenological events. Cutsets with initiating event frequencies that are less than 10-4 per year or those that do not pass the screening criteria of one or two basic events, are screened out and any associated SSCs are preliminary categorized as candidate LSS for core damage DID.
In its LAR [1], the licensee indicated that it piloted the alternate DID categorization process for ten Limerick systems that were previously categorized by current methods and compared the results to the existing process. The licensee also indicated that the systems selected represent both front line and support systems. The licensee proposed that both processes (the currently approved process [20] and the proposed process) be available for DID categorization.
3.2.2 NRC Staff Review of Alternate DID Approach The NRC staff reviewed the LAR and the subsequent revisions made to PWROG-20015-NP by the licensee during the review. The NRC staff finds that the licensee did not demonstrate that the proposed alternate DID process provides for an integrated and systematic categorization process, as required by 10 CFR 50.69(c)(1)(ii), and provides reasonable assurance that DID is 4
NUREG-2122, Glossary of Risk-Related Terms in Support of Risk-Informed Decisionmaking [34],
defines the term minimal cutset.
7 maintained, as required by 10 CFR 50.69(c)(1)(iii). The NRC staffs evaluation is discussed below.
3.2.3 Inconsistent and Lack of Correlation to Layers of Defense In its LAR, the licensee draws parallels between figure 6-1, Defense-in-Depth Matrix, in section 6 of NEI 00-04 [21] and the licensees proposed alternate DID process. In its supplement (section 2.2.6, Level of Defense-in-Depth Basis, of PWROG-20015-NP) [5], the licensee indicated that the proposed process is similar to the approved figure 6-1 in section 6 of NEI 00-04. However, the NRC staff notes that there are fundamental differences between the two methodologies and that the proposed methodology is less conservative.
The approved DID guidance in NEI 00-04 [21] specifies the number of trains or systems necessary for mitigation (by grouping based on initiating event frequency) and assesses the remaining mitigation capability without crediting any function or SSC that has been proposed as LSS or for any identical redundant SSCs within the system that are also classified as LSS. The licensee-proposed alternate process examines the results of the internal events PRA by reviewing the minimal cutsets (accident sequence scenarios) that result in core damage and allows credit for functions or SSCs that have been proposed as LSS or for identical redundant SSCs within the system that are classified as LSS. Furthermore, under the proposed process, common cause failure events would be considered using a limited common cause component grouping of four, which is discussed in section 3.2.5 of this safety evaluation. In contrast with NEI 00-04 [21], the proposed alternate process has all SSCs default to LSS for DID unless shown to be HSS and does not assess for remaining capability. In addition, the use of the PRA model overlooks other issues not analyzed or accounted for in the PRA, such as various modeling assumptions and those items associated with completeness uncertainties, as described in Revision 1 of NUREG-1855, Guidance on the Treatment of Uncertainties Associated with PRAs in Risk-Informed Decisionmaking [26], which can also result in miscategorizing the SSCs. The NRC staff finds this approach is a potential reduction in DID because (1) LSS SSCs, subject to reduced regulatory treatment, are credited for DID and (2) the proposed approach does not account for issues not analyzed in the PRA. The licensee did not provide sufficient justification for this potential reduction in DID.
Additionally, as described above, the licensee-proposed alternate process considers for HSS categorization those core damage cutsets having two basic events (two failures) for initiating event frequencies greater than 10-3 per year and one basic event (one failure) for initiating event frequencies greater than 10-4 per year, but less than 10-3 per year. Section 2.2.6 of PWROG-20015-NP, Revision 3 [5] indicates that each basic event cutset element represents an independent source of DID protection and that it maps directly to the NEI 00-04 [21] DID matrix.
However, the NRC staff notes that basic events elements within a cutset can share a common layer of DID (e.g., for high pressure coolant injection and reactor core isolation cooling, two steam-driven trains have identical pathways for reactor pressure vessel inventory control),
which results in dependent systems or components to provide protection to layers of DID. The NRC staff further notes that basic events in PRA models generally reflect equipment failures, common cause failures, failures of operator actions, and other modeling assumptions and simplifications that can be dependent on the modeling style of the PRA modeling engineer. The licensee did not demonstrate how counting two basic events provides sufficient confidence in the robustness of the proposed process to maintain DID because there is no direct correlation between cutset elements and layers of DID mitigation.
8 In the LAR, as supplemented [5], the licensee indicated that the proposed method meets the guidance for DID in RG 1.174 [24]; however, the NRC staff found instances where the proposed method is not consistent with RG 1.174 [24]. While RG 1.174 [24] credits insights of PRA results, it also credits deterministic engineering approaches and specifically cautions against using PRA as a sole source of DID in risk-informed decision-making. Section 3.1.2, Alternate Defense-in-Depth Categorization Process, of enclosure 1 to the LAR [1] states, [u]se of the PRA model logic structure presents a more effective and consistent method of assessing defense-in-depth for 10 CFR 50.69 categorization. However, variations in PRA modeling approaches could result in inconsistent categorization results. Therefore, the NRC staff finds that using a PRA approach as a sole substitute for deterministic DID can overlook the impact of SSCs that are not explicitly modeled in the PRA but that can have a contribution to higher risk scenarios, thus leading to potential miscategorization of an HSS SSC to LSS. Such miscategorization can result in less regulatory treatment being applied to safety-related SSCs important to safety, which has the potential to increase the likelihood of failures being undetected for extended periods of time for HSS SSCs. Therefore, the licensee did not demonstrate that the proposed process can provide reasonable assurance that DID is maintained, as required by 10 CFR 50.69(c)(1)(iii), and support an integrated and systematic 10 CFR 50.69 categorization process, as required by 10 CFR 50.69(c)(1)(ii).
3.2.4 Inappropriate Compensation for Reduced Rigor In its supplement [5], the licensee indicated that the level of technical adequacy in developing its RG 1.200 [27]-compliant PRA model is sufficient to evaluate DID, in lieu of the approved process in section 6 of NEI 00-04 [21]. Although the NRC staff is not questioning the veracity of the risk models and resolution of findings and observations, using the limited scope of a PRA model may involve SSCs not modeled in the PRA which the NRC staff finds is not sufficient to evaluate DID. Specifically, 10 CFR 50.69(c)(1)(ii) states:
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.
Section 2.1.1.3, Evaluating the Impact of the Proposed Licensing Basis Change on Defense in Depth, of RG 1.174 [24] for DID consideration 1 (preserve a reasonable balance among the layers of defense), provides additional guidance that states, in part:
...however, to address the unknown and unforeseen failure mechanisms or phenomena, the licensees evaluation of this defense-in-depth consideration should also address insights based on traditional engineering approaches.
Results and insights of the risk assessment might be used to support the conclusion; however, the results and insights of the risk assessment should not be the only basis for justifying that this defense-in-depth consideration is met.
The licensee should consider the impact of the proposed licensing basis change on each of the layers of defense.
As described in PWROG-20015-NP, Revision 3, section 2.2.2, item 8.d, the proposed DID categorization process assumes that SSCs are LSS, unless selected to be HSS by the proposed screening process. The licensees proposed categorization process has the potential to result in a significantly increased number of SSCs categorized as LSS that are passed to the
9 IDP for review and approval.5 These potential LSS SSCs could both be modeled in the PRA and not modeled in the PRA by default. Consequently, in the RAI (RAIs-10 and 12) [13], the NRC staff questioned how the licensee filters similar initiating events, filters cutsets, accounts for multiple human failure events, and categorizes potentially HSS unmodeled SSCs. In response [5] to the NRC staffs questions, the licensee discussed the role of the IDP in identifying possibly miscategorized SSCs and indicated that the IDP would require reviews from its engineering organization. The licensee further indicated in its RAI response [5] that, in addition to the IDP, the engineering team would be required to perform the evaluations that the IDP is expected to perform, as outlined in section 9.2.2 of NEI 00-04 [21]. These evaluations include evaluating the seven considerations for reviewing risk information and five considerations for DID, as outlined in section 9.2.2 of NEI 00-04 [21].
However, since the proposed process only focuses on the PRA minimal cutsets, categorizing the impact of SSCs that are not explicitly modeled in the PRA for DID would fall out of the process and rely on the IDP and engineering team, without any detailed guidance on how to evaluate issues that are not analyzed or not accounted for in the PRA model. Therefore, the NRC staff finds that the licensee did not demonstrate that the proposed process considers issues not analyzed or accounted for in the PRA model and did not provide additional guidance on how the IDP and engineering team would address such issues and, therefore, does not support an integrated and systematic 10 CFR 50.69 categorization process, as required by 10 CFR 50.69(c)(1)(ii).
3.2.5 Insufficient Basis for Treatment of Common Cause Failures The licensees approved current methodology of section 6 of NEI 00-04 [21] excludes like components for remaining capability in DID categorization (i.e., it does not allow credit for any identical, redundant SSCs within the system that are classified as LSS). However, in the proposed alternate process in PWROG-20015-NP, Revision 3 [5], the licensee proposes to exclude from HSS categorization, cutsets that contain common cause failure events comprising of four or more SSCs within the common cause component grouping.
The licensee indicated in section 2.2.3.4 of PWROG-20015-NP, Revision 3 [5], that the failure of four or more components within a common cause component grouping reflect significant redundancy, and that this consideration parallels the core damage DID matrix in NEI 00-04 [21].
However, the NRC staff finds that the licensee did not adequately justify these assertions because each SSC failure event within a common cause component grouping may not correspond to a standalone independent layer of DID mitigation and all SSC members of the common cause failure component grouping could share a common layer of DID. The NRC staff notes that an increased number of like components in a system does not always reflect significant redundancy, as success of more than one component may be necessary for system success. The NRC staff finds that the proposed approach has the potential to overlook the impact of common cause failure interactions of four or more SSCs, which is inconsistent with section 2.1.1.2, consideration 4, Preserve adequate defense against potential CCFs [Common Cause Failures], of RG 1.174 [24]. The licensee did not provide a justification for why limiting the grouping size to four or more can be categorized as LSS for DID for each SSC. Therefore, the NRC staff finds that the proposed approach has the potential to miscategorize specific common cause failures as LSS (because it can overlook the impact of common cause failure interactions of four or more SSCs) and, thus, the licensee did not demonstrate how the 5
The NRC staffs findings documented in this section are related to the proposed categorization process and not on the establishment of the IDP itself.
10 proposed process provides reasonable assurance that DID is maintained, as required by 10 CFR 50.69(c)(1)(iii).
3.2.6 Unjustified Mixed Use of DID Processes The licensee requested to use both the currently approved [20] NEI 00-04 [21] DID methodology and the proposed alternate DID process to categorize SSCs without justification for why using these unrelated processes results in an integrated and systematic categorization process, as required by 10 CFR 50.69(c)(1)(ii). The currently approved method assumes that safety-related SSCs are HSS, unless proven otherwise in an individual matrix evaluation using deterministic factors; whereas, the proposed method assumes that SSCs are LSS unless shown to be proven otherwise in a one-time PRA based evaluation. The currently approved [20] and proposed categorization processes would be performed at different points in the categorization procedure, which has the potential to introduce inconsistency if performed together. The NRC staff finds that this mixed use of processes has the potential to introduce miscategorization of SSCs because of the opposite approaches to DID. Therefore, the NRC staff finds that the licensee did not demonstrate how using these unrelated processes would result in an integrated and systematic categorization process, as required by 10 CFR 50.69(c)(1)(ii).
3.2.7 Conclusion for Proposed Defense-in-Depth Categorization Approach Based on its review, the NRC staff finds that the licensee did not demonstrate that the proposed DID methodology meets 10 CFR 50.69(c)(1)(ii) and (iii) for the reasons discussed in sections 3.2.1 through 3.2.6 of this safety evaluation. Therefore, the NRC staff concludes that the licensees proposed alternate DID approach described in the LAR, as supplemented, is not acceptable for the licensees categorization process under 10 CFR 50.69.
3.3 Evaluation of Proposed Pressure Boundary Categorization Approach 3.3.1 Proposed Methodology The licensees proposed pressure boundary categorization approach (or methodology) is based on the Electric Power Research Institute (EPRI) report, EPRI-3002015999, Enhanced Risk-Informed Categorization Methodology for Pressure Boundary Components [28], as modified in the licensees supplement [5] (together, the EPRI document). Section 4.2 of the EPRI document describes a classification process consisting of 14 criteria for determining if a passive SSC is to be considered HSS. The EPRI document uses ten deterministic criteria (criteria 1 through 10) to categorize SSCs as HSS. In the EPRI document, proposed criteria 1 through 4 are identical to subparagraphs (2)(a)(1 through 5) in ASME Code Case N-716-1, Alternative Piping Classification and Examination Requirements, section XI, Division 1 (ASME Code Case N-716) [29], except for some editorial differences. In the EPRI document, proposed criteria 5 through 10 are new, and the NRC staff noted that neither the LAR, as supplemented, nor the EPRI document describe the origin and significance of these criteria. Criteria 7 and 8 are specific to pressurized-water reactors (PWRs) and, therefore, do not apply to this LAR. The proposed criteria also consist of a set of quantitative risk-based criteria (criteria 11, 12, and 13) for categorizing components as HSS. Criterion 14 covers piping and component supports. Items not captured by at least one of the proposed 14 criteria would default to LSS.
11 3.3.2 NRC Staff Review of Alternate Pressure Boundary Approach As the proposed approach to categorizing SSCs for use in a 10 CFR 50.69 application is novel, the NRC staff evaluated the LAR, as supplemented, to determine if it had an adequate technical basis for each of the 14 criteria and an adequate analysis of using the criteria, together, to satisfy the requirements of 10 CFR 50.69(c)(1)(i)-(v). The NRC staff finds that the licensee did not demonstrate that the proposed pressure boundary categorization methodology demonstrates that 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, as required by 10 CFR 50.69(c)(1)(ii); and provides reasonable assurance that DID is maintained, as required by 10 CFR 50.69(c)(1)(iii).
The review of the pilot study described in the LAR [1] and RAI response [5] showed that the proposed methodology was being applied to individual components rather than to whole systems and, thus, the licensee did not demonstrate that the proposed methodology meets the requirements of 10 CFR 50.69(c)(1)(v). The review of the treatment of pipe break uncertainty showed that the licensee did not provide sufficient justification to show that the proposed methodology appropriately assessed impact on overall CDF and LERF from SSCs categorized as RISC-3, as required by 10 CFR 50.69(c)(1)(iv). The NRC staffs rationale for this conclusion is discussed below.
3.3.3 Insufficient Technical Basis for Proposed Methodology The NRC staff evaluated the proposed methodology and supporting basis provided in the LAR [1], as supplemented [5], to determine if the technical basis for the proposed methodology is sufficient to demonstrate that the requirements in 10 CFR 50.69(c)(1)(ii) that 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 are met. The NRC staff determined that the LAR, as supplemented, does not have an adequate technical basis to demonstrate compliance with 10 CFR 50.69(c)(1)(ii).
Table 4-1 in the EPRI document lists the basis for the 14 criteria, but these bases are not specific and lack adequate technical justification. For example, the technical basis for some of the criteria are unsupported assertions such as consistent with LARs approved to date, and consistent with some of the insights from previous pressure boundary categorization efforts (for example, 10 CFR 50.69, RI-ISI [risk-informed inservice inspection], RI-RRA [risk-informed repair/replacement applications]). Consequently, the NRC staff requested additional justification for the 14 criteria in RAI-05.B [13], as follows:
Table 4-1 of EPRl 3002015999 [the EPRI document] appears to provide the basis for the list of predetermined HSS SSCs. The basis criteria appear to include consistency with the current (ANO-2) passive categorization process and industry insights from the Rl-lSl program. Item Nos. 8 through 10 appear to rely on engineering judgement and experience. It is unclear to the NRC staff if the proposed alternate method is sufficiently comprehensive to identify all the necessary HSS SSCs given the apparent reliance on insights, judgements, and experience.
Provide justification that the EPRl determination of predetermined HSS SSCs is comprehensive and adequate to identify all SSCs that should be HSS.
12 In its supplement [5], the licensee
[13], as follows:
As mentioned in the RAI, insights, judgments and experience from application of the existing methodology as RI-ISI applications, RI-repair/replacement applications and [10 CFR] 50.69 applications were used to support development of the predetermined set of HSS SSCs. This BWR [boiling-water reactor] and PWR experience includes application of the technology to a large number of Class 1, Class 2 and Class 3 systems, including non-safety related SSCs in a number of these systems. Additionally, knowledge from these applications informed the understanding on how and why pressure boundary failures can impact defense-in-depth (e.g., loss of inventory versus spatial effects).
The portion of the methodology contained in Section 4.2 of 300201599 [the EPRI document] is essentially a two-step process. First, the plant is reviewed against criteria 1 through 10 to assign SSCs to HSS based upon the criteria contained in criteria 1 through 10. The second step is to determine, on a plant-specific basis, if there are additional SSCs that need to be added to the HSS scope. That is, all pressure boundary components (safety related and non-safety related) are reviewed against the criteria contained in criteria 11, 12 and 13. Any pressure boundary components (safety related or non-safety related) that exceed criteria contained in criteria 11, 12 or 13 are added to the HSS scope.
As such, criteria 1 through 13, taken in total, assure that all SSCs (safety related and non-safety related) have been assessed as to their risk significance.
Criteria 1 through 13 assure a robust list of HSS SSCs are identified and that only SSCs with a very small impact on plant risk are categorized as LSS.
The licensees RAI response [5], confirmed that the bases for the 14 criteria and the predetermined set of HSS SSCs consist of insights, judgments, and experience from the application of the existing methodology for RI-ISI applications, RI-RRA, and 10 CFR 50.69 applications, which the NRC staff evaluates the applicability of these sources in section 3.3.4 of this safety evaluation. The licensee did not provide any further technical justification explaining how the insights, judgments and experience were used to support its proposed methodology, which is a novel method for characterizing SSCs as HSS or LSS under 10 CFR 50.69.
Therefore, the licensee did not demonstrate how statements about insights, judgments, and experience were sufficient to demonstrate that 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, as required by 10 CFR 50.69(c)(1)(ii).
3.3.4 Applicability of RI-ISI, RI-RRA, and Previous LARs The NRC staff also evaluated the applicability of the sources (RI-ISI, RI-RRA, and 10 CFR 50.69 applications) of the insights, judgments, and experience mentioned in the licensees justification [5] for the proposed methodology. In addition to the staffs finding in section 3.3.3. of this safety evaluation, the NRC staff finds that the sources of the insights, judgments and experience discussed in the LAR, as supplemented, do not appear to be directly applicable to supporting the proposed methodology, as discussed below.
Proposed criteria 1 through 4, and 11 in the EPRI document are identical to subparagraphs (2)(a)(1 through 5) in ASME Code Case N-716 [29], with the exception of some
13 editorial differences. The LAR [1] and RAI response [5] cite RI-ISI and NRCs past acceptance of ASME Code Case N-716 for RI-ISI (via RG 1.147, Revision 19, Inservice Inspection Code Case Acceptability, ASME Section XI, Division 1 [30], which is incorporated into 10 CFR 50.55a by reference) as a justification for the proposed criteria used to categorize SSCs as HSS or LSS for 10 CFR 50.69.
While the NRC has approved the criteria described in ASME Code Case N-716 [29] to categorize components as HSS and LSS under RI-ISI programs, this approval does not extend to other usages of the terms. The approval of ASME Code Case N-716 considered the use of HSS and LSS within the context of the ASME Code Case and is not a generic approval of designating these systems as such for other uses outside of the ASME Code Case. RI-ISI programs use the terms HSS and LSS to prioritize components for inservice inspection, while 10 CFR 50.69 has a much larger scope (compared to RI-ISI) of alternative or reduced regulatory treatment, so transferring use of the LSS designation from ASME Code Case N-716 [29] for use in 10 CFR 50.69 is not equivalent. Consequently, the NRC staff has not previously approved these criteria for use in 10 CFR 50.69 applications. Therefore, the NRC staff finds that the proposed licensees basis presented in its LAR, as supplemented, for referencing LSS criteria from ASME Code Case N-716 [29] is not directly applicable to this LAR without additional technical justification.
The licensee also indicated in its RAI response [5] that it would use insights from RI-RRA.
Table 2 in RG 1.147 [30], states a condition that risk-informed repair/replacement ASME Code Cases N-660, Risk-Informed Safety Classification for Use in Risk-Informed Repair/Replacement Activities,Section XI, Division 1, and N-662-1, Alternative Repair/Replacement Requirements for Items Classified in Accordance with Risk-Informed Processes,Section XI, Division 1, must be applied only to ASME Code Classes 2 and 3 and non-Code Class pressure-retaining components and their associated supports. The NRC staff finds that the licensee did not justify how insights from these programs could be used for Class 1 SSCs. As with RI-ISI programs, the staff notes that the use of insights and experience from risk-informed repair/replacement code cases would not address the many other reduced regulatory treatments allowed by 10 CFR 50.69. The NRC staff finds that the licensee did not demonstrate that experience with RI-RRA is directly applicable to this LAR without additional technical justification.
The licensee also indicated in its RAI response [5] that it would use insights and experience from previous 10 CFR 50.69 license amendments; however, the proposed methodology in the EPRI document significantly differs from the ANO-2 method [22] of characterizing passive SSCs. The licensee did not explain how insights and experience from the use of the more rigorous ANO-2 method [22] or other NRC-approved methods would provide insights into the adequacy of the proposed approach, which differs significantly from those previously approved by the NRC staff.
Additionally, the LAR, as supplemented, does not clearly explain how criteria 5 through 10 can be justified by insights, judgements, and experience from RI-ISI, RI-RRA, and 10 CFR 50.69 applications. These criteria are not currently used in any approved RI-ISI program, RI-RRA, or 10 CFR 50.69 applications.
Based on the above, the NRC staff finds that the licensees basis of insights, judgments, and experience from the application of the existing methodology as RI-ISI, RI-RRA, and 10 CFR 50.69 applications does not demonstrate that the proposed methodology used to characterize SSC functional importance, reasonably reflects the current plant configuration and
14 operating practices, and applicable plant and industry operational experience. Therefore, the NRC staff concludes that the licensee did not demonstrate how the proposed methodology meets the requirements of 10 CFR 50.69(c)(1)(ii).
3.3.5 Evaluation of Pilot Study Results The NRC staff also evaluated the results of the Limerick-specific pilot study described in section 3.1.3 of enclosure 1 to the LAR [1] that used the proposed methodology. The pilot study results demonstrated that the proposed criteria would result in a large number of Class 1 components being categorized as LSS. The pilot study showed that allowing Class 1 items to be classified as LSS resulted in 341 SSCs, which had been categorized as HSS using the current ANO-2 method [22], to be categorized as LSS, using the process described in the EPRI document. As shown in the licensees supplement [5], these Class 1 SSCs included 2-inch main steam flow instrumentation, safeguard keep full 1.5-inch and 1-inch supply to reactor feedwater vessel instrumentation, and many instrumentation penetrations. Many individual small connections were also categorized as LSS in the pilot study, without any explanation of the individual or cumulative effects of these categorizations.
In the NRC staffs RAI [13] (RAI-01.B), the NRC staff requested justification for the licensees conclusion that the pilot studys categorization results using the proposed methodology are reasonable and consistent. The licensees response [5] was as follows:
The HSS portions of piping found in the current methodology were predominately the same as those found in the proposed EPRI methodology. There are two exceptions to this. What was different was the addition of BER [break exclusion region] piping not found in the current methodology. Constellation considers this is a conservative, acceptable addition to the scope of HSS piping. The reduction in HSS piping segments occurs with small instrument, drain, or supply piping like the example depicted below for Reactor Feedwater. In the figure below, the Safeguards Piping Fill system is connected to the 24" reactor feedwater system with 1" piping highlighted in brown. Failure of this piping would not prevent the plant from shutting down with normal makeup. Thus, it is reasonable that this class of piping be excluded from being HSS due to the nominal impact on the plant.
The NRC staff determined that the licensees statement (that the pilot study results using the proposed EPRI methodology were consistent with the current ANO-2 methodology [22]) in its RAI response [5] was not a sufficient justification because the ANO-2 methodology does not allow ASME Code Class 1 SSCs to be considered LSS, whereas the EPRI methodology allows many ASME Code Class 1 SSCs to be designated as LSS.
The licensees justification that the pilot studys categorization results are reasonable and consistent is because the failure of one 1-inch safeguards piping fill system pipe would not prevent the plant from shutting down with normal makeup. However, the licensee did not provide a basis to establish that this is a reasonable criterion for categorizing SSCs as LSS. The NRC staff notes that the licensee did not demonstrate in its RAI response [5] that the example was bounding or covered the large number and diversity of SSC sizes and systems affected by classifying ASME Code Class 1 SSCs as LSS.
The NRC staff also finds that the RAI response [5] does not address the reactor coolant boundarys function as a barrier to fission product release. The licensee did not provide an
15 explanation supporting the appropriateness of categorizing large numbers of such components as LSS, despite confirming that many of these components are ASME Code Class 1 (e.g., instrument, drain, or supply lines that may support a variety of normal and off-normal plant responses). Reducing the regulatory treatment for these SSCs can degrade their reliability, which may not maintain defense in depth. The licensee did not provide further substantive supporting details or information to justify these aspects of the proposed methodology.
Finally, the pilot study showed that the licensee had categorized several individual connections, instrumentation lines, and drain lines as LSS rather than categorizing entire systems or structures, contrary to the requirements in 10 CFR 50.69(c)(1)(v) that categorization be performed for entire systems and structures, not for selected components within a system or structure.
Based upon its review as discussed above, the NRC staff finds that the licensees pilot study results did not show that using the proposed methodology would provide reasonable assurance to meet the requirement of 10 CFR 50.69(c)(1)(ii) that 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 licensee did not provide a sufficient technical basis for designating hundreds of ASME Code Class 1 SSCs as LSS, using the proposed process, as seen in the pilot study results, which could degrade the reactor coolant pressure boundary. Thus, the licensee did not demonstrate that the proposed methodology would maintain defense in depth, which is contrary to the requirement of 10 CFR 50.69(c)(1)(iii). Also, the pilot study showed that the licensee categorized individual small SSCs using the proposed methodology as opposed to entire systems, contrary to the requirements of 10 CFR 50.69(c)(1)(v). Therefore, the NRC staff finds that the licensee did not demonstrate that the proposed categorization methodology meets 10 CFR 50.69(c)(1)(ii), (iii), and (v).
3.3.6 Inadequacy of Quantitative Risk-Based Criteria (Criteria 11, 12, and 13)
The proposed methodology in the EPRI document would have the licensee compare the risk of an SSC to the quantitative risk-based criteria 11, 12, and 13 if the SSC is not captured as HSS in the ten deterministic criteria of the proposed methodology. The quantitative risk-based criteria use CDF and LERF (with conditional core damage probability (CCDP) and conditional large early release probability (CLERP), respectively) to categorize SSCs as HSS. The NRC staff reviewed the use of quantitative risk-based criteria 11, 12, and 13 of the EPRI document.
The licensee currently uses [20] the method for categorizing passive components under 10 CFR 50.69 that the NRC staff approved for ANO-2 [22]. In the ANO-2 method [22], all ASME Code Class 1 SSCs are designated as HSS, and the ASME Code Class 2 and 3 SSCs are evaluated assuming a failure probability of one. The NRC staffs safety evaluation for Amendment Nos. 230 and 193 [20] states:
Categorizing solely based on consequences, which measures the safety-significance of the pipe given that it ruptures, is conservative compared to including the rupture frequency in the categorization. The categorization will not be affected by changes in frequency arising from changes to the treatment.
Uncertainties associated with consequences of failure (e.g., failure probabilities of one) are relatively small, compared to uncertainties associated with frequencies of failure. Consequently, assessing consequences of failure provides a higher degree of assurance that any effects of an
16 approval on plant configuration and operating practice will be adequately reflected. In addition, assessing consequences of failure also provides assurance that the results of any changes are appropriately considered when demonstrating that the proposed methodology will maintain DID.
However, in contrast to the ANO-2 method [22], the proposed methodology described in the EPRI document allows some ASME Code Class 1 SSCs to be categorized as LSS using CDF/LERF, which, unlike the CCDP/CLERP, does not assume a failure frequency of one.
These changes represent a less conservative method and a lower level of rigor (using CDF instead of CCDP) compared to Limericks currently accepted 10 CFR 50.69 process [20],
because the uncertainty associated with changes in frequencies is larger than those uncertainties for a consequence analysis and may be significant.
Criteria 11, 12, and 13 of the proposed methodology in the EPRI document use the internal flooding PRA model to derive contribution to CDF and LERF from a specific passive SSC, the CCDP, and the conditional large early release probability (CLERP). The proposed methodology would assign SSCs as LSS if the licensee determines that the SSC contribution to CDF is less than 10-6 per year, with some modifications if the CCDP for the SSC is between 1 or 0.01. It also would assign SSCs as LSS if the SSC contribution to LERF is less than 10-7 per year, with some modifications if the CLERP is between 0.1 and 0.01. Any SSC with a contribution to CDF less than 10-8 and a contribution to LERF less than 10-9 would be categorized LSS.
As described in ASME/American Nuclear Society (ANS) Standard ASME/ANS RA-Sa-2009, Addenda to ASME/ANS RA-S-2008, Standard for Level 1/Large Early Release Frequency Probabilistic Risk Assessment for Nuclear Power Plant Applications [31], the objective of the internal flooding PRA is to assess the impact of an internal flood as the cause of either an accident or a system failure. It analyzes fluid sources within the plant that could flood plant locations or create adverse conditions (e.g., spray, elevated temperature, humidity, pressure, pipe whip, jet impingement) that could damage mitigative plant equipment. There are failure modes of passive components that could lead to other plant consequences, outside of flooding, such as failing an instrument line that could initiate a plant transient. Smaller pipes usually are not considered a source of flooding and, therefore, would be screened out of the internal flooding PRA, thus leading to LSS categorization under the proposed methodology without further examination.
Furthermore, the staff notes that some passive SSC failures are considered as initiating events in the internal events PRA, such as main steam line break or LOCA. The licensee indicated in its RAI response [5] that the internal events PRA would also be used, but the licensee did not provide additional details. The internal events PRA generally does not explicitly model passive SSC failure outside of considering the initiating events needed to be modeled. Therefore, the NRC staff determined that the licensee did not demonstrate that the criteria in the proposed process considers consequences from pipe ruptures (for example, failures in passive components not modeled in PRA and protective measures); passive SSC failures (e.g., smaller piping may be screened out of the internal flooding PRA because it is not a source of flooding);
and deterministic cases of lower contribution to overall CDF and LERF with relatively high consequence of a single failure, which could miscalculate the significance of SSCs.
The NRC staff finds that the licensee did not support its proposed thresholds for determining whether pressure boundary components are HSS or LSS. For example, the licensee did not provide a justification for using a CCDP greater than 0.01 or a CLERP greater than 0.01 for HSS designation. The NRC staff also finds that the threshold for contribution to CDF of 10-6 or
17 LERF of 10-7 does not appear to capture any SSCs as HSS because individual contributions to CDF/LERF are generally lower than 10-6 CDF and 10-7 LERF.
In RAI-01.D [13], the NRC staff requested the licensee to provide the number and percentage of SSCs determined to be HSS based on the ten deterministic criteria and, separately, criteria 11 through 14 per section 4.2 of the EPRI document. The licensees response [5] stated, All (100%) of the HSS SSCs were determined to be such from the prescribed HSS criteria 1 through 10. There are no SSCs that meet the 11 through 14 criteria. The thresholds established in criteria 11 through 13 for categorizing SSCs would result in few or no SSCs being quantitatively assessed for risk. Because the quantitative risk criteria were applied to hundreds of diverse SSCs in the pilot study and classified each SSC as LSS, the licensee did not demonstrate that the thresholds are able to accurately categorize SSCs as HSS or LSS.
Finally, some SSCs not modeled in the PRA do not appear to be assessed outside of criteria 1 through 10 in the EPRI document. If these components are not captured by criteria 1 through 10, then the proposed methodology would default them to the LSS categorization and, therefore, HSS SSCs may be mischaracterized as LSS.
Based on the above, the NRC staff finds that the licensee did not demonstrate that the proposed process considers certain consequences and did not provide sufficient justification for the use of the proposed risk thresholds, which could result in the proposed risk criteria misclassifying HSS SSCs as LSS. Therefore, the NRC staff determined that, contrary to 10 CFR 50.69(c)(1)(ii), the licensee did not demonstrate how the proposed methodology determines 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.
3.3.7 Uncertainties and Changes in Pipe Break Frequencies The use of PRA in criteria 11, 12, and 13 in the EPRI document, which rely on frequency risk insights, introduces pipe break frequencies based on current operating experience into the categorization process and, therefore, add a potential significant source of uncertainty. The NRC staff reviewed this methodology against 10 CFR 50.69(c)(1)(iv), which requires the proposed methodology to include evaluations that provide reasonable confidence that for SSCs categorized as RISC-3, sufficient safety margins are maintained and that any potential increases in CDF and LERF resulting from changes in treatment permitted by implementation of 10 CFR 50.69(b)(1) and (d)(2) are small.
The previously approved method for passive categorization [20] was based on the consequences of the passive SSC failures by categorizing components with a CCDP greater than 10-4 as HSS. In contrast, for HSS categorization, the proposed alternate approach considers the CDF and LERF contribution, pipe break frequency, and a larger CCDP threshold.
However, the LAR, as supplemented, does not include any substantive consideration of this source of uncertainty.
The NRC staff notes that if passive components were categorized to LSS per the proposed methodology and subject to the further reduced regulatory treatment allowed under 10 CFR 50.69, pipe break frequencies may change because of the different potential treatment effects of RI-ISI versus 10 CFR 50.69. The LAR, as supplemented, did not adequately address how the proposed methodology would consider the impact on safety margins and any potential increases in CDF or LERF from these changes in pipe break frequencies. Specifically, the
18 licensee did not demonstrate how the proposed methodology addresses the expected increase in pipe break frequency, the effects on risk from reducing regulatory treatment per 10 CFR 50.69 for ASME Class 1 SSCs, or SSCs with a high CCDP. The NRC staff notes that this is in contrast, for example, to the categorization process described in NEI 00-04 [21] for active SSCs, which includes an evaluation of the impact on risk from changes in reliability from changes in treatment for active RISC-3 SSCs, to meet the requirements of 10 CFR 50.69(c)(1)(iv).
The NRC finds that the LAR did not explain or justify how the licensee would assess the impact on overall CDF and LERF from all candidate passive RISC-3 SSCs for passive SSCs categorized using the proposed alternate approach, which is contrary to the requirements in 10 CFR 50.69(c)(1)(iv) that requires the licensee to include evaluations that provide reasonable confidence that for SSCs categorized as RISC-3, sufficient safety margins are maintained and that any potential increases in CDF and LERF resulting from changes in treatment permitted by implementation of 10 CFR 50.69(b)(1) and (d)(2) are small.
Based on the above, the NRC staff finds that the licensee did not demonstrate that the LAR, as supplemented, meets 10 CFR 50.69(c)(1)(iv) because the licensee did not provide sufficient justification to show that the proposed methodology would assess the impact on overall CDF and LERF from SSCs categorized as RISC-3.
3.3.8 Conclusion for Proposed Pressure Boundary Categorization Approach Based on its review, the NRC staff finds that the licensees proposed pressure boundary categorization methodology does not meet 10 CFR 50.69(c)(1)(ii), (iii), (iv), and (v) for the reasons discussed in sections 3.3.3 through 3.3.7 of this safety evaluation. Therefore, the NRC staff concludes that the licensees proposed alternate pressure boundary approach is not acceptable for the licensees categorization process under 10 CFR 50.69.
3.4 Proposed Changes to License Conditions The licensee proposed a license condition that would allow the use of alternative DID and pressure boundary approaches for categorizing SSCs in the application of 10 CFR 50.69.
Based on the NRC staffs conclusions in sections 3.2.7 and 3.3.8 of this safety evaluation, the NRC staff denies the incorporation of these two proposed methods into the 10 CFR 50.69 license condition.
4.0 CONCLUSION
The Commission has concluded, based on the considerations discussed above, that there is not reasonable assurance that such activities would be conducted in compliance with the Commissions regulations. Therefore, the Commission denies the proposed alternative DID and pressure boundary approaches for categorizing SSCs in the licensees application of 10 CFR 50.69.
19
5.0 REFERENCES
[1] Rafferty-Czincila, S., Exelon Generation Company, LLC, letter to U.S. Nuclear Regulatory Commission, Application to Implement an Alternate Defense-in-Depth Categorization Process, an Alternate Pressure Boundary Categorization Process, and an Alternate Seismic Tier 1 Categorization Process in Accordance with the Requirements of 10 CFR 50.69, Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors, March 11, 2021 (ML21070A412).
[2] Helker, D. P., Exelon Generation Company, LLC, letter to U.S. Nuclear Regulatory Commission, Supplement - Application to Implement an Alternate Defense-in-Depth Categorization Process, an Alternate Pressure Boundary Categorization Process, and an Alternate Seismic Tier 1 Categorization Process in Accordance with the Requirements of 10 CFR 50.69, Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors, May 5, 2021 (ML21125A215).
[3] Helker, D. P., Exelon Generation Company, LLC, letter to U.S. Nuclear Regulatory Commission, Supplement - Application to Implement an Alternate Defense-in-Depth Categorization Process, an Alternate Pressure Boundary Categorization Process, and an Alternate Seismic Categorization Process in Accordance with the Requirements of 10 CFR 50.69, Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors, December 15, 2021 (ML21349B364).
[4] Helker, D. P., Constellation Energy Generation, LLC, letter to U.S. Nuclear Regulatory Commission, Supplement - Application to Implement an Alternate Defense-in-Depth Categorization Process, an Alternate Pressure Boundary Categorization Process, and an Alternate Seismic Categorization Process in Accordance with the Requirements of 10 CFR 50.69, Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors, February 14, 2022 (ML22045A480).
[5] Helker, D. P., Constellation Energy Generation, LLC, letter to U.S. Nuclear Regulatory Commission, Response to Request for Additional Information Application to Implement an Alternate Defense-in-Depth Categorization Process, an Alternate Pressure Boundary Categorization Process, and an Alternate Seismic Categorization Process in Accordance with the Requirements of 10 CFR 50.69, Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors, June 30, 2022 (ML22182A400).
[6] Klett, A., U.S. Nuclear Regulatory Commission, letter to Rhoades, D. P., Constellation Energy Generation, LLC, Limerick Generating Station, Units 1 and 2 - Issuance of Amendment Nos. 261 and 223 Re: Alternative Seismic Approach for Risk-Informed Categorization and Treatment of Structures, Systems, and Components and Denial of Proposed Alternative Defense-in-Depth and Pressure Boundary Component Categorization Processes (EPID L-2021-LLA-0042), May 17, 2023 (ML23089A124).
[7] Sreenivas, V., U.S. Nuclear Regulatory Commission, letter to Rhoades, D. P., Exelon Generation Company, LLC, Limerick Generating Station Units 1 and 2 - Audit Plan in Support of Review of License Amendment Request to Implement an Alternative Defense-in-Depth Categorization Process, an Alternate Pressure Boundary Categorization Process, and an Alternate Seismic Tier 1 Categorization Process in Accordance with the Requirements of 10 CFR 50.69, Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors, (EPID L-2021-LLA-0042), October 1, 2021 (ML21263A248).
20
[8] Klett, A., U.S. Nuclear Regulatory Commission, email to Stewart, G. H., Exelon Generation Company, LLC, Supplement to Limerick 50.69 Audit Plan dated October 1, 2021 (L-2021-LLA-0042), October 20, 2021 (ML21295A036).
[9] Klett, A., U.S. Nuclear Regulatory Commission, email to Stewart, G. H., Exelon Generation Company, LLC, Audit Plan Supplement for LIM 50.69 LAR (L-2021-LLA-0042),
January 24, 2022 (ML22028A183).
[10] Klett, A., U.S. Nuclear Regulatory Commission, email to Stewart, G. H., Exelon Generation Company, LLC, Audit Plan Supplement for LIM 50.69 LAR (L-2021-LLA-0042),
February 2, 2022 (ML22034A014).
[11] Klett, A., U.S. Nuclear Regulatory Commission, letter to Rhoades, D. P., Constellation Energy Generation, LLC, Limerick Generating Station, Units 1 and 2 - Summary of Regulatory Audit in Support of Alternate Processes for Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors (EPID L-2021-LLA-0042), Month 24, 2023 (ML23079A188).
[12] Sreenivas, V., U.S. Nuclear Regulatory Commission, email to Stewart, G. H. (et al), Exelon Generation Company, LLC, Limerick Application to Modify 50.69 Categorization to Implement an Alternate Defense-In-Depth Categorization Process, an Alternate Pressure Boundary Categorization Process, and an Alternate Seismic Tier 1 Categorization, April 20, 2021 (ML21111A031).
[13] Klett, A., U.S. Nuclear Regulatory Commission, email to Stewart, G. H., Constellation Energy Generation, LLC, NRC Request for Additional Information - Limerick License Amendment Request (L-2021-LLA-0042), May 13, 2022 (ML22136A003).
[14] Klett, A., U.S. Nuclear Regulatory Commission, letter to Rhoades, D. P., Constellation Energy Generation, LLC, Limerick Generating Station, Units 1 and 2 - Summary of May 3, 2022, Public Observation Meeting with Constellation Energy Generation, LLC Re. Proposed License Amendment Request to Modify 10 CFR 50.69 License Conditions (EPID L-2021-LLA-0042), June 24, 2022 (ML22168A192).
[15] Klett, A., U.S. Nuclear Regulatroy Commission, letter to Rhoades, D. P., Constellation Energy Generation, LLC, Limerick Generating Station, Units 1 and 2 - Summary of June 24, 2022, Public Observation Meeting with Constellation Energy Generation, LLC Re. Proposed License Amendment Request to Modify 10 CFR 50.69 License Conditions (EPID L-2021-LLA-0042), August 25, 2022 (ML22227A111).
[16] Klett, A., U.S. Nuclear Regulatory Commission, letter to Rhoades, D. P., Constellation Energy Generation, LLC, Limerick Generating Station, Units 1 and 2 - Summary of February 23, 2023, Public Observation Meeting with Constellation Energy Generation, LLC Re. Proposed License Amendment Request to Modify Title 10 of the Code of Federal Regulations, Section 50.69 License Conditions (EPID L-2021-LLA-0042), March 28, 2023 (ML23076A280).
[17] U.S. Nuclear Regulatory Commission, Monthly Notice; Applications and Amendments to Facility Operating Licenses and Combined Licenses Involving No Significant Hazards Considerations, Office of the Federal Register, Volume 86, Page 43690 (86 FR 43686).
[18] U.S. Nuclear Regulatory Commission, Monthly Notice; Applications and Amendments to Facility Operating Licenses and Combined Licenses Involving No Significant Hazards Considerations, Office of the Federal Register, Volume 87, Page 9649 (87 FR 9647),
February 22, 2022.
[19] U.S. Nuclear Regulatory Commission, Monthly Notice; Applications and Amendments to Facility Operating Licenses and Combined Licenses Involving No Significant Hazards
21 Considerations, Office of the Federal Register, Volume 87, Page 15458 (87 FR 15458),
March 18, 2022.
[20] Sreenivas, V., U.S. Nuclear Regulatory Commission, letter to Hanson, B. C., Exelon Generation Company, LLC, Limerick Generating Station, Units 1 And 2 - Issuance of Amendment Nos. 230 and 193 to Adopt Title 10 of the Code of Federal Regulations Section 50.69, Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors, (CAC NOS. MF9873 AND MF9874; EPID L-2017-LLA-0275), July 31, 2018 (ML18165A162).
[21] Nuclear Energy Institute, NEI 00-04, Revision 0, 10 CFR 50.69 SSC Categorization Guideline, July 2005 (ML052910035).
[22] Markley, M. T., U.S. Nuclear Regulatory Commission, letter to Vice President, Operations, Entergy Operations, Inc., Arkansas Nuclear One, Unit 2 - Approval of Request for Alternative ANO2-R&R-004, Revision 1, Request to Use Risk-Informed Safety Classification and Treatement for Repair/Replacement Activities in Class 2 and3 Moderate and High Energy Systems, (TAC MD5250), April 22, 2009 (ML090930246).
[23] U.S. Nuclear Regulatory Commission, Risk-Informed Categorization and Treatment of Structures, Systems and Components for Nuclear Power Reactors, Office of the Federal Register, Volume 69, Page 68008 (69 FR 68008), November 22, 2004.
[24] U.S. Nuclear Regulatory Commission, Regulatory Guide 1.174, Revision 3, An Approach for Using Probabilistic Risk Assessment in Risk-Informed Decisions on Plant-Specific Changes to the Licensing Basis, January 2018 (ML17317A256).
[25] Powell, M., Pressurized Water Reactor Users Group, letter to U.S. Nuclear Regulatory Commission, PWR Owners Group, Risk Management Committee, For Information Only - Transmittal of PWROG-20015-NP Revision 1, Alternate 10 CFR 50.69 Defense-in-Depth Categorization Process, March 23, 2021 (ML21084A097).
[26] U.S. Nuclear Regulatory Commission, NUREG-1855, Guidance on the Treatment of Uncertainties Associated with PRAs in Risk-Informed Decisionmaking, Final Report, March 2017 (ML17062A466).
[27] U.S. Nuclear Regulatory Commission, Regulatory Guide 1.200, Revision 2, An Approach for Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk-Informed Activities, March 2009 (ML090410014).
[28] Electric Power Research Institute, EPRI-3002015999, Enhanced Risk-Informed Categorization Methodology for Pressure Boundary Components, November 2019 (ML21082A171).
[29] American Society of Mechanical Engineers, ASME Code Case N-716-2, Alternative Piping Classification and Examination Requirements,Section XI, Division 1, November 7, 2016.
[30] U.S. Nuclear Regulatory Commission, Regulatory Guide 1.147, Revision 19, Inservice Inspection Code Case Acceptability, ASME Section XI, Division 1, October 2019 (ML19128A244).
[31] American Society of Mechanical Engineers and the American Nuclear Society, ASME/ANS, ASME/ANS RA Sa-2009, Addenda to ASME/ANS RA S 2008, Standard for Level 1/Large Early Release Frequency Probabilistic Risk Assessment for Nuclear Power Plant Applications, Addendum A to RA S 2008, February 2009.
[32] U.S. Nuclear Regulatory Commission and Pacific Northwest National Laboratory, NUREG/CR-6936 (PNNL-16186), Probabilities of Failure and Uncertainty Estimate Information for Passive Components - A Literature Review, May 2007 (ML071430371).
22
[33] Electric Power Research Institute, EPRI TR-111880-NP, Piping System Failure Rates and Rupture Frequencies for Use in Risk-Informed In-service Inspection Applications, Final Report, September 1999 (ML003776638).
[34] U.S. Nuclear Regulatory Commission, NUREG-2122, Glossary of Risk-Related Terms in Support of Risk-Informed Decisionmaking, November 2013 (ML13311A353).
6.0 ABBREVIATIONS ANO-2 Arkansas Nuclear One, Unit 2 ANS American Nuclear Society ASME American Society of Mechanical Engineers BER break exclusion region BWR boiling-water reactor CCDP conditional core damage probability CCF common cause failure CDF core damage frequency CEG Constellation Energy Generation, LLC CFR Code of Federal Regulations CLERP conditional large early release probability DID defense-in-depth or defense in depth EPRI Electric Power Research Institute HSS high safety-significant (or high safety significance)
IDP integrated decision-making panel LAR license amendment request LERF large early release frequency LSS low safety-significant (or low safety significance)
NEI Nuclear Energy Institute NRC U.S. Nuclear Regulatory Commission NSHC no significant hazards consideration PRA probabilistic risk assessment PWR pressurized-water reactor PWROG Pressurized Water Reactor Owners Group RAI request for additional information RG regulatory guide RI risk-informed RI-ISI risk-informed inservice inspection RI-RRA risk-informed repair/replacement applications RISC risk-informed safety class SoC statement of considerations SSC structure, system, and component 7.0 PRINCIPAL CONTRIBUTORS Jeff Circle, Office of Nuclear Reactor Regulation Mihaela Biro, Office of Nuclear Reactor Regulation Steve Cumblidge, Office of Nuclear Reactor Regulation Dan Widrevitz, Office of Nuclear Reactor Regulation Audrey Klett, Office of Nuclear Reactor Regulation