Publications

Abstract not provided.

Abstract not provided.

Abstract not provided.

Thermal sprayed metal coatings are used in many industrial applications, and characterizing the structure and performance of these materials is vital to understanding their behavior in the field. X-ray Computed Tomography (CT) machines enable volumetric, nondestructive imaging of these materials, but precise segmentation of this grayscale image data into discrete material phases is necessary to calculate quantities of interest related to material structure. In this work, we present a methodology to automate the CT segmentation process as well as quantify uncertainty in segmentations via deep learning. Neural networks (NNs) are shown to accurately segment full resolution CT scans of thermal sprayed materials and provide maps of uncertainty that conservatively bound the predicted geometry. These bounds are propagated through calculations of material properties such as porosity that may provide an understanding of anticipated behavior in the field.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Radiation transport in stochastic media is a challenging problem type relevant for applications such as meteorological modeling, heterogeneous radiation shields, BWR coolant, and pebble-bed reactor fuel. A commonly cited challenge for methods performing transport in stochastic media is to simultaneously be accurate and efficient. Conditional Point Sampling (CoPS), a new method for transport in stochastic media, was recently shown to have accuracy comparable to the most accurate approximate methods for a common 1D benchmark set. In this paper, we use a pseudo-interface-based approach to extend CoPS to application in multi-D for Markovian-mixed media, compare its accuracy with published results for other approximate methods, and examine its accuracy and efficiency as a function of user options. CoPS is found to be the most accurate of the compared methods on the examined benchmark suite for transmittance and comparable in accuracy with the most accurate methods for reflectance and internal flux. Numerical studies examine accuracy and efficiency as a function of user parameters providing insight for effective parameter selection and further method development. Since the authors did not implement any of the other approximate methods, there is not yet a valid comparison for efficiency with the other methods.

Radiation transport in stochastic media is a problem found in a multitude of applications, and the need for tools that are capable of thoroughly modeling this type of problem remains. A collection of approximate methods have been developed to produce accurate mean results, but the demand for methods that are capable of quantifying the spread of results caused by the randomness of material mixing remains. In this work, the new stochastic media transport algorithm Conditional Point Sampling is expanded using Embedded Variance Deconvolution such that it can compute the variance caused by material mixing. The accuracy of this approach is assessed for 1D, binary, Markovian-mixed media by comparing results to published benchmark values, and the behavior of the method is numerically studied as a function of user parameters. We demonstrate that this extension of Conditional Point Sampling is able to compute the variance caused by material mixing with accuracy dependent on the accuracy of the conditional probability function used.

Abstract not provided.

Radiation transport in stochastic media is a problem found in a multitude of applications, and the need for tools that are capable of thoroughly modeling this type of problem remains. A collection of approximate methods have been developed to produce accurate mean results, but the demand for methods that are capable of quantifying the spread of results caused by the randomness of material mixing remains. In this work, the new stochastic media transport algorithm Conditional Point Sampling is expanded using Embedded Variance Deconvolution such that it can compute the variance caused by material mixing. The accuracy of this approach is assessed for 1D, binary, Markovian-mixed media by comparing results to published benchmark values, and the behavior of the method is numerically studied as a function of user parameters. We demonstrate that this extension of Conditional Point Sampling is able to compute the variance caused by material mixing with accuracy dependent on the accuracy of the conditional probability function used.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

The objective of this project is to investigate accuracy of error metrics in SCEPTRE and produce useful benchmarks, identify metrics that do not work well, identify metrics that do work well, and produce easy to reference results.