Machine learning modeling of irradiation embrittlement in low alloy steel of nuclear power plants

Lee, Gyeong-Geun; Kim, Min‐Chul; Lee, Bong-Sang

doi:10.1016/j.net.2021.06.014

Cited by 17 publications

(7 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When fluence is out of this range, its influence seems to be reduced. Specifically, the ICE plot shows a plateau beyond 1.24 × 10 24 n/m 2 : this result must be interpreted carefully because, as previously noted by Lee et al [27], extrapolation for tree-based ML models (such as GB) provides a constant value. Again, it is worth noting that both the analytical models proposed by the ASTM E900 [4] and by Kirk [28] describe the impact of fluence on the TTS through a logarithmic expression.…”

Section: Discussionsupporting

confidence: 60%

See 1 more Smart Citation

Prediction of the Transition-Temperature Shift Using Machine Learning Algorithms and the Plotter Database

Ferreño

Serrano

Kirk

et al. 2022

Metals

View full text Add to dashboard Cite

The long-term operating strategy of nuclear plants must ensure the integrity of the vessel, which is subjected to neutron irradiation, causing its embrittlement over time. Embrittlement trend curves used to predict the dependence of the Charpy transition-temperature shift, ΔT41J, with neutron fluence, such as the one adopted in ASTM E900-15, are empirical or semi-empirical formulas based on parameters that characterize irradiation conditions (neutron fluence, flux and temperature), the chemical composition of the steel (copper, nickel, phosphorus and manganese), and the product type (plates, forgings, welds, or so-called standard reference materials (SRMs)). The ASTM (American Society for Testing and Materials) E900-15 trend curve was obtained as a combination of physical and phenomenological models with free parameters fitted using the available surveillance data from nuclear power plants. These data, collected to support ASTM’s E900 effort, open the way to an alternative, purely data-driven approach using machine learning algorithms. In this study, the ASTM PLOTTER database that was used to inform the ASTM E900-15 fit has been employed to train and validate a number of machine learning regression models (multilinear, k-nearest neighbors, decision trees, support vector machines, random forest, AdaBoost, gradient boosting, XGB, and multi-layer perceptron). Optimal results were obtained with gradient boosting, which provided a value of R2 = 0.91 and a root mean squared error ≈10.5 °C for the test dataset. These results outperform the prediction ability of existing trend curves, including ASTM E900-15, reducing the prediction uncertainty by ≈20%. In addition, impurity-based and permutation-based feature importance algorithms were used to identify the variables that most influence ΔT41J (copper, fluence, nickel and temperature, in this order), and individual conditional expectation and interaction plots were used to estimate the specific influence of each of the features.

show abstract

Section: Discussionsupporting

confidence: 60%

“…Lee et al [27] previously developed an ETC using ML methods. They trained several regression models using the ASTM PLOTTER database described in Section 2.1.…”

Section: Discussionmentioning

confidence: 99%

Prediction of the Transition-Temperature Shift Using Machine Learning Algorithms and the Plotter Database

Ferreño

Serrano

Kirk

et al. 2022

Metals

View full text Add to dashboard Cite

show abstract

“…These sophisticated algorithms turn out to be often very powerful. The specific example in the nuclear materials field where this approach is being applied with some degree of success concerns correlations for RPV steel embrittlement versus neutron fluence and other variables [174][175][176].…”

Section: Advanced Modelling and Characterisationmentioning

confidence: 99%

“…One of the main problems with data-driven modelling procedures is that they are too often blind: the AI produces in most cases a sort of "black box" transfer function between input and output, a priori devoid of any physics (even though sometimes this procedure manages to improve also our physical understanding [174,175]). The more data are available, the higher are the chances that the procedure provides probative results, although it remains dangerous and unwarranted to rely on extrapolations [176]. In the case of RPV steels, a large amount of data is available from surveillance and MTR experiments, thus this approach is especially promising [174][175][176].…”

Section: Advanced Modelling and Characterisationmentioning

confidence: 99%

“…The more data are available, the higher are the chances that the procedure provides probative results, although it remains dangerous and unwarranted to rely on extrapolations [176]. In the case of RPV steels, a large amount of data is available from surveillance and MTR experiments, thus this approach is especially promising [174][175][176]. However, this situation is not necessarily common in the nuclear materials field.…”

Section: Advanced Modelling and Characterisationmentioning

confidence: 99%

See 1 more Smart Citation

Materials for Sustainable Nuclear Energy: A European Strategic Research and Innovation Agenda for All Reactor Generations

Malerba

Mazouzi²,

Bertolus

et al. 2022

Energies

View full text Add to dashboard Cite

Nuclear energy is presently the single major low-carbon electricity source in Europe and is overall expected to maintain (perhaps eventually even increase) its current installed power from now to 2045. Long-term operation (LTO) is a reality in essentially all nuclear European countries, even when planning to phase out. New builds are planned. Moreover, several European countries, including non-nuclear or phasing out ones, have interests in next generation nuclear systems. In this framework, materials and material science play a crucial role towards safer, more efficient, more economical and overall more sustainable nuclear energy. This paper proposes a research agenda that combines modern digital technologies with materials science practices to pursue a change of paradigm that promotes innovation, equally serving the different nuclear energy interests and positions throughout Europe. This paper chooses to overview structural and fuel materials used in current generation reactors, as well as their wider spectrum for next generation reactors, summarising the relevant issues. Next, it describes the materials science approaches that are common to any nuclear materials (including classes that are not addressed here, such as concrete, polymers and functional materials), identifying for each of them a research agenda goal. It is concluded that among these goals are the development of structured materials qualification test-beds and materials acceleration platforms (MAPs) for materials that operate under harsh conditions. Another goal is the development of multi-parameter-based approaches for materials health monitoring based on different non-destructive examination and testing (NDE&T) techniques. Hybrid models that suitably combine physics-based and data-driven approaches for materials behaviour prediction can valuably support these developments, together with the creation and population of a centralised, “smart” database for nuclear materials.

show abstract

Characterizing the flux effect on the irradiation embrittlement of reactor pressure vessel steels using machine learning

et al. 2023

View full text Add to dashboard Cite

Machine learning modeling of irradiation embrittlement in low alloy steel of nuclear power plants

Cited by 17 publications

References 10 publications

Prediction of the Transition-Temperature Shift Using Machine Learning Algorithms and the Plotter Database

Prediction of the Transition-Temperature Shift Using Machine Learning Algorithms and the Plotter Database

Materials for Sustainable Nuclear Energy: A European Strategic Research and Innovation Agenda for All Reactor Generations

Characterizing the flux effect on the irradiation embrittlement of reactor pressure vessel steels using machine learning

Contact Info

Product

Resources

About