ML18141A393
| ML18141A393 | |
| Person / Time | |
|---|---|
| Issue date: | 05/21/2018 |
| From: | Stephen Cumblidge Piping and NDE Branch |
| To: | |
| Rezai A, NRR-DMLR 415-1328 | |
| References | |
| Download: ML18141A393 (8) | |
Text
NRR Perspectives on Probabilistic Fracture Mechanics Calculations Stephen Cumblidge Materials Engineer NRR/DE/EPNB United States Nuclear Regulatory Commission (U.S.NRC)
Tel: +13014152823 Email: stephen.cumblidge@nrc.gov Rockville, MD May 21 2018
NRC Use of PFM Calculations The NRC has been using probabilistic risk assessments to guide regulatory decision making in several areas for decades The NRC has used the probabilistic fracture mechanics (PFM) code FAVOR to develop the alternate Pressurized Thermal Shock Rule (10CFR50.61a)
The NRC is cooperatively developing the Extremely Low Probability of Rupture (xLPR) Probabilistic fracture mechanics code to evaluate the leak before break methodology in systems with active Primary Water Stress Corrosion Cracking The NRC is open to evaluating and using new PFM codes in the future to assist in regulatory decision making, and recognizes the powerful insights provided by PFM 2
Benefits of PFM Calculations PFM Calculations can provide an overview of the probabilities of crack depths, leaks, or failure Deterministic or X
One can decide on a probability vs.
of an event as opposed to the binary criteria of a deterministic calculation PFM PFM and Deterministic Calculations can be effective independently or may be complementary 3
Confirmatory Calculations When the NRC staff performs a technical review, we check the important calculations performed in the submittal The NRC has experts in a variety of fields, including PFM, to allow for this type of confirmatory work PFM calculations present unique difficulties for performing confirmatory calculations 4
Novel PFM Codes Without access to the code or independent review of the code it can be very challenging to determine if the PFM calculations are adequate 5
NRC Review of a PFM Possible methods for verifying the code
- Having the code directly evaluated by NRC staff (Topical Report)
- Benchmarking the code against a code we have in house (xLPR, FAVOR)
- Benchmarking the code against deterministic calculations
- Independent peer review by a body with no conflict of interest 6
Issues With Previous PFM Codes Unusual assumptions (e.g. extremely low possibility of a circumferential flaw in a piping weld)
PFM did not account for degradation mechanisms seen in operational experience Correction factors with little to no technical justification 7
Path Forward The NRC is developing guidance on the use of PFM calculations in regulatory submittals This guidance will help to make submittals using PFM calculations more robust, will allow for more timely and predictable reviews, and make acceptance of PFM submittals easier 8