Weibull and Bootstrap-Based Data-Analytics Framework for Fatigue Life Prognosis of the Pressurized Water Nuclear Reactor Component Under Harsh Reactor Coolant Environment

Journal Article · 05 November 2019 · Journal of Nondestructive Evaluation, Diagnostics and Prognostics of Engineering Systems

DOI: https://doi.org/10.1115/1.4045162 · OSTI ID:1579939

Park, Jae Phil ^[1]; Mohanty, Subhasish ^[2]; Bahn, Chi Bum ^[1]; Majumdar, Saurin ^[2]; Natesan, Krishnamurti ^[2]

---

1. Pusan National Univ., Busan (South Korea)
2. Argonne National Lab. (ANL), Lemont, IL (United States)

In general, the fatigue life of a safety critical pressure component is estimated using best-fit fatigue life curves (S-N curves). These curves are estimated based on underlying in-air condition fatigue test data. The best-fitting approach requires a large safety factor to accommodate the uncertainty associated with large scatter in fatigue test data. In addition to this safety factor, reactor component fatigue life prognostics requires an additional correction factor that in general is also estimated deterministically. This additional factor known as the environmental correction factor F_en is to cater the effect of the harsh coolant environment that severely reduces the life of these components. The deterministic F_en factor may also lead to further conservative estimation of fatigue life leading to unnecessary early retirement of costly reactor components. To address the above-mentioned issues, we propose a data-analytics framework which uses Weibull and Bootstrap probabilistic modeling techniques for explicitly quantifying the uncertainty/scatter associated with fatigue life rather than estimating the lives based on a best-fit based deterministic approach. We assume the proposed probabilistic approach would provide the first hand information for assessing the maximum and minimum effects of pressurized water reactor water on the reactor component. In the discussed approach, in addition to the probabilistic fatigue curves, we suggest using a probabilistic environment correction factor F_en. Here, we assume the probabilistic fatigue curve and F_en would capture the S-N data scatter associated with the bulk effect of material grades, surface finish, strain rate, etc. on the material/component fatigue life.

• Research Organization: Argonne National Laboratory (ANL), Argonne, IL (United States)
• Sponsoring Organization: National Research Foundation of Korea (NRF); Korea Institute of Energy Technology Evaluation and Planning (KETEP); USDOE Office of Nuclear Energy (NE), Nuclear Reactor Technologies (NE-7). Light Water Reactor Sustainability Program
• Grant/Contract Number: AC02-06CH11357
• OSTI ID: 1579939
• Journal Information: Journal of Nondestructive Evaluation, Diagnostics and Prognostics of Engineering Systems, Vol. 3, Issue 1; ISSN 2572-3901
• Publisher: ASMECopyright Statement
• Country of Publication: United States
• Language: English