Machine learning predictions of irradiation embrittlement in reactor pressure vessel steels

Abstract

Irradiation increases the yield stress and embrittles light water reactor (LWR) pressure vessel steels. In this study, we demonstrate some of the potential benefits and risks of using machine learning models to predict irradiation hardening extrapolated to low flux, high fluence, extended life conditions. The machine learning training data included the Irradiation Variable for lower flux irradiations up to an intermediate fluence, plus the Belgian Reactor 2 and Advanced Test Reactor 1 for very high flux irradiations, up to very high fluence. Notably, the machine learning model predictions for the high fluence, intermediate flux Advanced Test Reactor 2 irradiations are superior to extrapolations of existing hardening models. The successful extrapolations showed that machine learning models are capable of capturing key intermediate flux effects at high fluence. Similar approaches, applied to expanded databases, could be used to predict hardening in LWRs under life-extension conditions.

A machine learning model for flank wear prediction in face milling of Inconel 718

“…The 5-fold CV test was repeated 20 times in this work, and the CV RMSE was given as the average of the foldaverage RMSEs. Details of the CV methods and the cross-plot analyses can be found elsewhere [22].…”

“…Recently, deep learning (DL) and machine learning (ML) model are considered as powerful methods to decipher and explore the complex underlying physics of the materials science and engineering [18], including quality prediction in manufacturing [19], effective charge in electromigration effect [20], dielectric constant and dissipation factor in low temperature co-red ceramics [21], irradiation embrittlement in steel [22] etc. More relevantly, direct tool wear detection of physical vapor deposition (PVD)-coated carbide inserts by using a convolution neural network (CNN) model using image features is one of the approach to characterize failure occurrence [23].…”

Machine learning in nuclear materials research

“…Distinct from the semi-empirical models just discussed, there have been five models of RPV embrittlement using standard ML approaches. 94,[110][111][112][113] The earliest work came from Obraztsov et al 94 in 2006, who used a surveillance database of DBTT shifts DTx (we denote this temperature shift DTx as it was not clear from our available references how it was measured) for 41 main metal and weld-seam materials in the VVER-440 vessels. Features included a dozen alloy elements, fluence, power plant number, and a binary coding of main metal or weld seam.…”

“…We would further add that these uncertainty estimates are themselves uncertain and need validation. However, the authors note that these types of models are quick to generate and can serve as checks on new data and physics-based models by highlighting where there are disagreements and thus indicating more caution and double-checking is needed, which suggests an important application for their use.Recently Liu et al112 have completed a similar study to Mathew et al111 using an extended version of the full test reactor IVAR database that adds a number of high flux, high fluence test reactor irradiations from ATR1 irradiations, giving a total of 1501 data points in what they call the IVAR+ database. Liu, et al target Dsy and use composition variables (Cu, Ni, Mn, Si, P, C), irradiation temperature, fluence, and a function of flux and fluence called effective fluence as features.…”