ML24319A134
| ML24319A134 | |
| Person / Time | |
|---|---|
| Issue date: | 11/21/2024 |
| From: | David Dijamco NRC/NRR/DNRL |
| To: | |
| References | |
| Download: ML24319A134 (1) | |
Text
ASME Code Inspection Relaxations Applications and Performance Monitoring DAVID DIJAMCO PRESENTATION TO THE ACRS FUELS, MATERIALS, & STRUCTURES SUBCOMMITTEE NOVEMBER 21, 2024
Topics
- PFM aspects staff focused on
- PFM acceptance criteria
- Audit of the PROMISE PFM computer code
- Sensitivity studies
- Criteria for plant-specific applications
- Performance monitoring
- Statistically determined inspection sample size
- Plant-specific applications
- Pressurizer (PZR) and steam generator (SG) vessel welds and nozzles
- Single/two-unit plant submittals and fleet submittals 2
Precedents for PFM with adequate performance monitoring (vessels)
- PFM BWRVIP-05 and BWRVIP-329-A (based on FAVOR analyses)
- Performance monitoring axial/longitudinal welds still being examined
- PFM WCAP-16168-A (based on FAVOR analyses)
- Performance monitoring coordinated fleet inspections that ensure regular stream of monitoring data
- Reduction of BWR vessel nozzle inspections (Code Case N-702-1)
- PFM BWRVIP-108 and BWRVIP-241 (based on VIPERNOZ)
- Performance monitoring 25% of nozzles still being inspected 3
PFM aspects staff focused on Acceptance criteria 1x10-6 failures/yr, consistent with the basis during the development of 10 CFR 50.61a, in which reactor pressure vessel (RPV) TWCF was conservatively assumed to be equivalent to an increase in CDF.
- Conservative because in reality an increase in RPV TWCF does not mean an equivalent increase in CDF
- Details are in NUREG-1806, Technical Basis for Revision of the Pressurized Thermal Shock (PTS) Screening Limit in the PTS Rule (10 CFR 50.61)
- Used for the PFM analyses in:
- EPRI report 3002015905 for PZR
- While PZRs and SGs are safety significant, they are not as safety significant as the RPV; therefore, staff finds 1x10-6 failures/yr appropriate.
4
PFM aspects staff focused on Audit of the PROMISE computer code
- PROMISE stands for Probabilistic Optimization of Inspection
- 2.5-day audit (ML20258A002); objective was for staff to understand how PFM principles were being applied, were they consistent with guidance
- Referred to RG 1.245 (guidance for PFM submittals)
- Inputs/models (probabilistic models, e.g., mean and standard deviation of distributed variables, but also non-probabilistic models, e.g., FEA, stress intensity factor solutions, ISI & exam coverage)
- Uncertainties
- Convergence
- Software V&V
- Sensitivity studies 5
PFM aspects staff focused on Audit of the PROMISE computer code (continued)
Key observations Software V&V was adequate Uncertainties adequately addressed Initial flaw distribution model was adequate ISI and examination coverage adequately modeled Performed adequate sensitivity studies 6
PFM aspects staff focused on PROMISE audit - V&V and Uncertainties
- Software V&V
- Followed ASME NQA standards and 10 CFR 50 Appendix B guidance
- Software V&V plan and V&V reports generated
- Plan contained testing of the various parts of the software, and that testing results were adequate and reflected in the reports
- Uncertainties
- Mean and standard deviation values of random variables (i.e., those with a probability distribution rather than a single value) were consistent with previously accepted values.
crack depth crack length fracture toughness crack growth rate crack growth threshold 7
PFM aspects staff focused on PROMISE audit - Initial Flaw Distribution, ISI & Exam Coverage
- Based on the Pressure Vessel Research Users Facility (PVRUF) unused RPV
- Developed from NDE of fabrication flaws in the vessel weld
- Consists primarily of small-surface breaking flaws
- Used in the BWRVIP-05-based submittals
- Staff ensured that ISI and examination coverage (of the weld volume) were modeled since these are key aspects of ASME Code,Section XI, examinations.
- ISI model: implemented through a probability of detection (POD) curve at times of inspections
- Examination coverage model: implemented by allowing modeled postulated flaw to grow for a number of realizations proportional to coverage missed 8
PFM aspects staff focused on PROMISE audit - Sensitivity studies Staff ensured that sensitivity studies (SS) were performed for the critical parameters of stress and fracture toughness.
SS on stress up to more than 2 times base case stress levels, and on fracture toughness up less than half of base case fracture toughness were performed and showed that acceptance criteria of 1x10-6 failures/yr was met.
9 From RG 1.245:
PFM aspects staff focused on Criteria for plant-specific applications
- EPRI reports were based representative/conservative geometric configurations, transients/cycles based on survey of PWRs
- Thus, the need for criteria for the following parameters in plant-specific applications:
- Geometry
- Materials
- Loading conditions (thus stress) and cycles 10
PFM aspects staff focused on Criteria for plant-specific applications (continued) 11 EPRI Report 3002014590 EPRI Report 3002015906 SGs PZRs EPRI Report 3002015905 Staff also evaluates plant-specific inspection history: number of ISIs and examination volume coverage.
Performance monitoring Supports RIDM in three primary ways What about the other 3 aspects of RIDM: safety margins, defense-in-depth, and compliance with regulations?
- Safety margins and defense-in-depth: primarily have to do with design; design parameters (material properties and operating characteristics) and multiple means to accomplish safety functions are not changing
- Compliance with regulations: licensees seek an alternative to ASME Code requirements pursuant to 10 CFR 50.55a(z)(1)evaluated by staff 12
- Direct evidence of presence and/or extent of degradation
- Validation/confirmation of continued adequacy of analyses
- Timely method to detect novel/unexpected degradation
Performance monitoring (continued)
Illustration of interval extension
- Fewer inspections with interval extension. The question is: what inspection sample size is acceptable?
13
Performance monitoring (continued)
Statistically determined sample
- Quantitative sampling calculation can be derived from statistical calculation (next two slides)
- Binomial distribution
- Monte carlo analysis
- At the conceptual level, the objective is to determine the sample size (in our case # of inspections) from a population of like objects that gives x%
probability of success outcome (detection of degradation/cracking),
assuming a certain p% of the population has characteristic for "success" outcome (degraded/cracked).
- Staff described details in Rudland, David L. and Widrevitz, Dan, PVP2023-105203, Statistical Approach to Developing a Performance Monitoring Program 14
Performance monitoring (continued)
Binomial distribution The binomial distribution is frequently used to model the number of successes in a sample of size n drawn with replacement from a population of a certain size Can be used to find # of inspections needed to find a crack Independent of population size 15
,, =
1
=
k= number of successes (cracks found) n=number of trials (inspections) p= probability of success on an individual trial (% of population cracked)
If k=0 then this is the probability of no successes is:
, = 1 and therefore, the probability of at least one success is:
1,
Performance monitoring (continued)
Monte carlo (MC) analysis Same concept can be applied with an MC analysis More general, allows maximum flexibility in the analysis Binomial response can be recreated Works for better for small populations 16 Sample a weld population with x%
cracked Loop on weld population cracked?
Sample Inspection of y% of welds inspected?
count=count+1 MC Loop, n realizations p=count/n yes yes no no Done
?
yes no Done
?
yes no
Performance monitoring (continued)
Should the statistics be applied at weld level or whole component level?
Weld level 17
Performance monitoring (continued)
Should the statistics be applied at weld level or whole component level?
Component level: inspection of the whole component means inspecting the suite of welds required to be inspected for that component (PZR in our example).
18 PRESSURIZER
Performance monitoring (continued)
Example of statistical calculation for PZRs (1 of 2)
Objective:
Determine inspection sample size for performance monitoring of PZRs 19 Population size =
61 PZRs (61 PWRs in US, one PZR ea) 5% of population is degraded/cracked 90% probability of detecting at least one crack in the population BINOMIAL/MONTE CARLO 25% of population of PZRs
Performance monitoring (continued)
Example of statistical calculation for PZRs (2 of 2)
- Submittal with 1 unit requesting three 10-year intervals
- 3 PZR inspections required by ASME Code
- 25% sample = 1 PZR for performance monitoring sample (rounded up)
- Submittal with 10 units requesting three 10-year intervals
- 30 PZR inspections required by ASME Code
- 25% sample = 8 PZRs for performance monitoring sample (rounded up) 20
Performance monitoring (continued)
Timing of inspections
- Inspections performed later during the requested extended interval more impactful (but time from last inspection cant be too long).
- Later inspections have more chance of detecting degradation (if present) than earlier inspections since the degradation has had time to develop to a level that is detectable.
21
Plant-specific applications Submittals using the EPRI reports as technical basis
- Applications (i.e., submittals) have been coming pursuant to 10 CFR 50.55a(z)(1) requesting to extend ISI intervals, referring to the EPRI reports as technical basis.
- Staff approach on evaluating these:
- PFM consistent with the technical basis reports, especially that the submittal meets the plant-specific criteria covered earlier
- EPRI report 3002015905 for PZRs
- Performance monitoring is adequate 22
Plant-specific applications (continued)
Single or two-unit plant submittals
- These submittals are for one or two-unit plants proposing to extend the ASME Code required 10-year ISI interval to up to three 10-year ISI intervals.
- They refer to the EPRI reports for the PFM the technical basis and provide an adequate performance monitoring plan.
23
Plant-specific applications (continued)
Fleet submittals
- These submittals are for more multiple plants (thus for multiple units) proposing to extend the ASME Code required 10-year ISI interval to up to three 10-year ISI intervals; tech basis for PFM also the EPRI reports.
- Proposed performance monitoring gets interesting since now you have different alignment of ISI intervals of the various plants.
24
Plant-specific applications (continued)
Fleet submittals Proposed Performance Monitoring Sample 25 v
v w
x x
y z
Plant-specific applications (continued)
Fleet submittals Example of how the staff confirms that the proposed sample size for performance monitoring is adequate.
26 Calculation of total ASME Code required PZR inspections Using statistics, sample size needed is 0.25 x 14 = 4 PZRs (rounded up)
Calculation of PZR equivalents Total no. of PZRs in proposed monitoring sample is = 1.0 (from above) + 3 (from prev slide) = 4
Guidance?
- There have been fifteen or so submittals for PZRs and SGs since the first submittals.
- Similar approach taken for other components. Examples:
- Heat exchanger vessels
- Reactor closure head studs, but with DFM as technical basis instead of PFM
- These clearly bring up the question, is the staff developing guidance?
27
Questions?
28