Future code development to adopt a risk-informed design methodology will require improved accuracy of fatigue initiation predictions. The ability to account for through wall strain gradients in plant components, particularly over the first 3 mm of wall thickness is one area where conservatism can be reduced. This is due to extant design fatigue curves being derived from strain controlled membrane loading tests where the 25% load drop definition for end of test equates to approximately a 3 mm crack. Being able to define initiation fatigue curves for much shorter crack depths would enable fatigue crack growth methods to then predict the additional cycles, taking into consideration the strain gradient, required before the defined end of life crack size is reached. The R5 procedure provides a method to adjust an existing Stress-Life (SN) curve representing an initiation crack depth, to a smaller depth. This method was developed for materials at higher temperature and for a CO2 cooling environment, thus its validity was uncertain for application to a Pressurised Water Reactor (PWR) plant. This paper details the development of best estimate and design basis SN curves and environmental fatigue enhancement factors (Fen) for crack initiation to a depth of 250 μm. It is concluded that the general methodology in R5 was found, through this work, to adequately describe fatigue initiation lives for stainless steel in a PWR environment when augmented with a crack size dependent Fen equation and with modified coefficients determined through regression. The resulting method is similar to R5 in that an SN curve can be derived for any crack size, however the current data set only provides validation down to a crack size of 250 μm, as striations at shorter depths were not visible with existing methods.
Even with improvements to remove excessive conservatisms, current fatigue assessment approaches can result in high Cumulative Usage Factors (CUFs) for some analyses. In order to improve plant availability from these assessments and mitigate future changes to design codes, an improvement in understanding in this area is desirable. Hence the proposal for a Life Assessment Methodology (LAM) was created. The LAM is a concept for an approach based on modelling each stage of fatigue life to predict total fatigue life, as a means of minimising conservatism in an assessment, where necessary. It should also be capable of incorporating statistical methods to assign reliability figures to calculated plant lives. This paper describes the proposed definition of the LAM and how a proof of concept version of the LAM was developed to assess the Bettis Bechtel Stepped Pipe (BBSP) test. The results were presented with two seeded cases (fixed inputs) and a range of lives corresponding to associated Target Reliabilities (TRs). The Best Estimate (BE) and TR associated lives produced were based on using the latest methods available for calculating Fatigue Initiation (FI) and Fatigue Crack Growth (FCG), whereas the seeded Effective Strain Range (ESR) comparison case used current deterministic assessment methods. The results for the case study concluded that there is a benefit to pursuing the development of the LAM when compared to traditional assessment methods. It highlighted and quantified the conservatism present in traditional assessment methods for these cases as well as the need to understand the required TR for a specific component as this can have a large effect on the predicted life. With further refinements to the method, a more realistic and robust output of the total fatigue life distribution (for specific cases) would be obtained, which in turn would allow us to better quantify the conservatism associated with a TR.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.