Uncertainty Sampling of Weld Residual Stress Fields in Probabilistic Analysis: Part II — Examples

Abstract

NRC Standard Review Plan (SRP) 3.6.3 describes Leak-Before-Break (LBB) assessment procedures that can be used to assess compliance with the 10CFR50 Appendix A, GDC-4 requirement that primary system pressure piping exhibit an extremely low probability of rupture. SRP 3.6.3 does not allow for assessment of piping systems with active degradation mechanisms, such as Primary Water Stress Corrosion Cracking (PWSCC) which is currently occurring in systems that have been granted LBB approvals. US NRC staff, working cooperatively with the Electric Power Research Institute through a memorandum of understanding, conducted a multi-year project that focused on the development of a viable method and approach to address the effects of PWSCC in primary piping systems approved for LBB. This project, called eXtremely Low Probability of Rupture (xLPR) [1], defined the requirements necessary for a modular-based probabilistic fracture mechanics assessment tool to directly assess compliance with the regulations. Using the lessons learned from the pilot study, the production version of this code, designated as Version 2.0, focused on those primary piping systems previously approved for LBB. In this version the appropriate fracture mechanics-based models are employed to model the physical cracking behavior and a variety of computational options are provided to characterize, categorize and propagate problem uncertainties. One of the most influential uncertainty on risk in the xLPR code is the one associated with weld residual stresses (WRS). WRS plays a key role in both crack initiation and crack growth. PWSCC is mainly driven by tensile stresses, whose major contributors are the tensile weld residual stresses that develop during fabrication of the piping system. Handling the uncertainty involved with WRS within a probabilistic framework is quite challenging. A companion paper presents the selected approach to represent uncertainty within the framework of the xLPR code while respecting a set of requirements in term of smoothness of profile, efficiency of (potential) importance sampling and (for axial WRS) equilibrium. This paper illustrate with examples the implementation of the described methods into xLPR v2.0.