Machine learning to predict refractory corrosion during nuclear waste vitrification

2021-04-01
Smith-Gray, Natalie J.
Sargın, Irmak
Beckman, Scott
McCloy, John
The goal of this study was to determine the effects of model nuclear waste glass composition on the corrosion of Monofrax(R) K-3 refractory, using machine learning (ML) methods for data investigation and modeling of published borosilicate glass composition data and refractory corrosion performance. First, statistical methods were used for exploration of the data, and the list of features (model terms) was determined. Several model types were explored, and the Bayesian Ridge type was the most promising due to low mean average error and mean standard error as well as high R-2 value. Parameters and model results using previously identified model features and those from this study are compared. ML methods appear to give results at least as good as previously available models for describing the effects of glass composition on refractory corrosion.

Suggestions

Theoretical and experimental investigation of water-gas shift reaction over supported copper/iron oxide catalysts
Yalçın, Özgen; Önal, Işık; Department of Chemical Engineering (2017)
Density functional theory calculations were carried out to investigate the role of the chromium and copper, which were demonstrated to act as textural and catalytic promoters, respectively, for the Fe3O4-Cr2O3-CuO catalyst system by in situ experimental studies at the atomic scale. There is a minor effect of Cr on the dissociative adsorption of H2O, but no effect on CO adsorption on the Feoct2 termination of Fe3O4 (111) surface indicating that Cr does not act as a chemical promoter. Copper promotion of the ...
Robust design of lithium extraction from boron clays by statistical design and analysis of experiments
Büyükburç, Atıl; Köksal, Gülser; Department of Industrial Engineering (2003)
In this thesis, it is aimed to design lithium extraction from boron clays using statistical design of experiments and robust design methodologies. There are several factors affecting extraction of lithium from clays. The most important of these factors have been limited to a number of six which have been gypsum to clay ratio, roasting temperature, roasting time, leaching solid to liquid ratio, leaching time and limestone to clay ratio. For every factor, three levels have been chosen and an experiment has be...
Numerical investigation of coupled heat and mass transfer inside the adsorbent bed of an adsorption cooling unit
Solmus, Ismail; Rees, D. Andrew S.; Yamali, Cemil; Baker, Derek Keıth; KAFTANOĞLU, BİLGİN (2012-05-01)
In this study, the influence of several design parameters on the transient distributions of temperature, pressure and amount adsorbed in the radial direction of a cylindrical adsorbent bed of an adsorption cooling unit using silica gel/water have been investigated numerically. For this purpose, a transient one-dimensional local thermal non-equilibrium model that accounts for both internal and external mass transfer resistances has been developed using the local volume averaging method. For the conditions in...
Durability of PEM fuel cell electrocatalysts prepared by microwave irradiation technique
Sayin, Elif Seda; BAYRAKÇEKEN YURTCAN, Ayşe; Eroğlu, İnci (2012-11-01)
In this study, the durability of the PEM fuel cell electrocatalysts was investigated by using cyclic voltammetry (CV) and rotating disk electrode (RDE) techniques. Accelerated degradation tests (ADTs) were applied to the commercial catalysts and the electrocatalysts prepared by microwave irradiation technique in order to determine the platinum dissolution/agglomeration and carbon corrosion characteristics. The parameters examined for the commercial catalysts were carbon to Nafion (C/N) ratio in the catalyst...
Geotechnical Assessment of Compacted Sand Bentonite Mixtures to be Utilized in Underground Nuclear Waste Repositories and Barrier Design
Akgün, Haluk; Koçkar, Mustafa Kerem (2016-07-01)
This study assesses the mechanical and hydrological performance of compacted sand bentonite mixtures to be utilized in sealing underground nuclear waste repositories. In order to assess the performance of this material and to obtain an optimum sand bentonite mixture, a variety of laboratory tests, namely, compaction, falling head permeability, swelling, unconfined compression and shear strength tests were performed on sand bentonite mixtures possessing bentonite contents ranging from 5% to 15%. Sand bentoni...
Citation Formats
N. J. Smith-Gray, I. Sargın, S. Beckman, and J. McCloy, “Machine learning to predict refractory corrosion during nuclear waste vitrification,” MRS ADVANCES, vol. 6, no. 4-5, pp. 131–137, 2021, Accessed: 00, 2022. [Online]. Available: https://hdl.handle.net/11511/96130.