Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “feature selection”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 433 records · Page 24
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Development of a Convolutional Neural Network Classifier for Data Starved Spectra - 20199

The Institute for Clean Energy Technology (ICET) at Mississippi State University is exploring the utility of machine learning in augmenting its mobile radiation surveying platforms, which are currently being developed as means to survey depleted uranium contaminated areas in support of remediation and decommissioning efforts. Mobile survey platforms provide a means to efficiently scan large areas of interest while reducing human exposure to radiation and other hazards. The survey platforms can also be used for scanning for any gamma emitting isotope in addition to depleted uranium. The spectral data that the platforms collect may be data starved with relatively low counts and poorly defined spectral features depending on the speed of the platforms and scintillation detector selection. Such data-starved spectra are difficult to use for isotope identification, requiring advanced knowledge of the possible radionuclides that could be present and environmental factors that could attenuate signals or introduce background noise. These factors in combination with the volume of survey data increases the time it takes to perform analysis of survey data when the source type is unknown. There are a number of algorithms in the field of machine learning that can be used to classify data that would be challenging and time-consuming for a human to identify. Supervised machine learning algorithms train models based on extensive amounts of human-labeled training data. Once sufficiently trained, these models can be used to quickly make high-fidelity predictions on new data. Convolutional neural networks are machine learning algorithms that excel in learning representations of 'shapes'. They do this by taking numerical input data and convolving them with spatial feature detectors referred to as filters. These filters are incrementally adjusted to reduce the prediction error on the data during the backpropagation step of training. Discussed in this paper is the development of a convolutional neural network classifier (CNNC) that can utilize spectral survey data for source discrimination and isotope identification. Bench-top laboratory experiments data using LaBr{sub 3}(Ce) scintillation detectors were used to train and evaluate the performance of the developed CNNC. The CNNC is capable of discriminating a variety of gamma emitting source types, differentiating different forms of uranium (depleted vs. natural), and estimating the amount of uranium for a known geometry. The discussed CNNC may be useful in scenarios where survey systems are deployed in situations where hazardous radioactive material maybe present, but the type is unknown. When used in remediation applications the CNNC can be used to screen-out false positives, helping reduce remediation costs. (authors)

07 ISOTOPE AND RADIATION SOURCES↗
DOE OSTI2025

Conformational ensembles reveal the origins of serine protease catalysis

Enzymes exist in ensembles of states that encode the energetics underlying their catalysis. Conformational ensembles built from 1231 structures of 17 serine proteases revealed atomic-level changes across their reaction states. By comparing the enzymatic and solution reaction, we identified molecular features that provide catalysis and quantified their energetic contributions to catalysis. Serine proteases precisely position their reactants in destabilized conformers, creating a downhill energetic gradient that selectively favors the motions required for reaction while limiting off-pathway conformational states. The same catalytic features have repeatedly evolved in proteases and additional enzymes across multiple distinct structural folds. In conclusion, our ensemble-function analyses revealed previously unknown catalytic features, provided quantitative models based on simple physical and chemical principles, and identified motifs recurrent in nature that may inspire enzyme design.

Du, Siyuan [Stanford Univ., CA (United States)] (O↗
DOE OSTI2022

Effect of second Si–O vibrational overtones/combinations on quantifying water in silicate and silica minerals using infrared spectroscopy, and an experimental method for its removal

Infrared spectroscopy (IR) is the most widely used analytical tool to quantify trace water in silicate and silica minerals. A prerequisite for highly accurate IR measurements of trace water is a good understanding of the effect of the second Si–O vibrational overtones/combination bands (2nd Si–O VOCBs) on the water peaks. Silicate and silica minerals can be divided as isolated (Q 0 ), paired (Q 1 ), ring (Q 2 ), chain (Q 1 or Q 2 ), sheet (Q 3 ) and framework (Q 4 ) structures according to the polymerization of their SiO 4 tetrahedral units, and the 2nd Si–O VOCBs of these different structural types attain different vibrational features which are expected to affect the water peaks to different extents. Here, we selected olivine (Q 0 ) and α-quartz (Q 4 ) as two endmember-like structural examples, performed extensive IR measurements on both pristine and heat-treated thin sections prepared for these two minerals, and explored the vibrational features of the 2nd Si–O VOCBs. In this work, we have found that the 2nd Si–O VOCBs are well separated from the water peaks in olivine, but severely overlap with the water peaks in α-quartz, confirming the different roles that the 2nd Si–O VOCBs play in quantifying trace water in silicate and silica minerals with different structural polymerizations. To remove the influence of the 2nd Si–O VOCBs (or any other species rather than water), an experimental protocol has been successfully developed, as approved by some fundamental equations and verified by the data of α-quartz in the literature. This development should lead to significant accuracy improvement in quantifying trace amounts of water in Earth and planetary materials.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Tuning Metal–Organic Framework (MOF) Topology by Regulating Ligand and Secondary Building Unit (SBU) Geometry: Structures Built on 8-Connected M 6 (M = Zr, Y) Clusters and a Flexible Tetracarboxylate for Propane-Selective Propane/Propylene Separation

Topology evolution originating from variations of linker and SBU (Secondary Building Unit) geometries could largely enrich the chemistry of metal–organic frameworks (MOFs). Here we report the synthesis and characterization of three MOF structures built on the same organic linker, N,N,N',N'-Tetrakis(4-carboxyphenyl)-1,4-phenylenediamine (tcppda) and similar 8-connected M 6 (M = Zr or Y) clusters. The three compounds, HIAM-402, HIAM-403, and HIAM-311, feature 4,8-connected sqc, scu, and flu topology, respectively. Detailed structural analysis revealed that different geometries of the inorganic M 6 SBUs as well as the organic linker have led to the formation of distinct MOF nets. In particular, HIAM-402 features exceptional framework stability and high porosity and acts as a propane-selective adsorbent for the discrimination of propane and propylene. Its balanced adsorption selectivity (S propane/propylene = 1.43) and capacity (Q propane = 133.3 cm 3 /g, 298 K and 1 bar) endow it with the capability of separating propane and propylene mixtures and one-step production of highly pure propylene (purity >99.9%), as validated by column breakthrough measurements, with the presence of moisture or propyne. As a result, ab initio calculations further confirm that the propane-selective behavior of HIAM-402 is a result of its higher binding energy toward propane compared to that of propylene.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Rapid screening of secondary aromatic metabolites in Populus trichocarpa leaves

Abstract Background High-throughput metabolomics analytical methodology is needed for population-scale studies of bioenergy-relevant feedstocks such as poplar ( Populus sp.). Here, the authors report the relative abundance of extractable aromatic metabolites in Populus trichocarpa leaves rapidly estimated using pyrolysis-molecular beam mass spectrometry (py-MBMS). Poplar leaves were analyzed in conjunction with and validated by GC/MS analysis of extracts to determine key spectral features used to build PLS models to predict the relative composition of extractable aromatic metabolites in whole poplar leaves. Results The Pearson correlation coefficient for the relative abundance of extractable aromatic metabolites based on ranking between GC/MS analysis and py-MBMS analysis of the Boardman leaf set was 0.86 with R 2 = 0.76 using a simplified prediction approach from select ions in MBMS spectra. Metabolites most influential to py-MBMS spectral features in the Clatskanie set included the following compounds: catechol, salicortin, salicyloyl-coumaroyl-glucoside conjugates, α-salicyloylsalicin, tremulacin, as well as other salicylates, trichocarpin, salicylic acid, and various tremuloidin conjugates. Ions in py-MBMS spectra with the highest correlation to the abundance of extractable aromatic metabolites as determined by GC/MS analysis of extracts, included m/z 68, 71, 77, 91, 94, 105, 107, 108, and 122, and were used to develop the simplified prediction approach without PLS models or a priori measurements. Conclusions The simplified py-MBMS method is capable of rapidly screening leaf tissue for relative abundance of extractable aromatic secondary metabolites to enable prioritization of samples in large populations requiring comprehensive metabolomics that will ultimately inform plant systems biology models and advance the development of optimized biomass feedstocks for renewable fuels and chemicals.

09 BIOMASS FUELS↗
DOE OSTI2020

A greedy algorithm for computing eigenvalues of a symmetric matrix with localized eigenvectors

Here, we present a greedy algorithm for computing selected eigenpairs of a large sparse matrix $H$ that can exploit localization features of the eigenvector. When the eigenvector to be computed is localized, meaning only a small number of its components have large magnitudes, the proposed algorithm identifies the location of these components in a greedy manner, and obtains approximations to the desired eigenpairs of $H$ by computing eigenpairs of a submatrix extracted from the corresponding rows and columns of $H$. Even when the eigenvector is not completely localized, the approximate eigenvectors obtained by the greedy algorithm can be used as good starting guesses to accelerate the convergence of an iterative eigensolver applied to $H$. We discuss a few possibilities for selecting important rows and columns of $H$ and techniques for constructing good initial guesses for an iterative eigensolver using the approximate eigenvectors returned from the greedy algorithm. We demonstrate the effectiveness of this approach with examples from nuclear quantum many-body calculations and many-body localization studies of quantum spin chains.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Prescreening-Based Subset Selection for Improving Predictions of Earth System Models With Application to Regional Prediction of Red Tide

We present the ensemble method of prescreening-based subset selection to improve ensemble predictions of Earth system models (ESMs). In the prescreening step, the independent ensemble members are categorized based on their ability to reproduce physically-interpretable features of interest that are regional and problem-specific. The ensemble size is then updated by selecting the subsets that improve the performance of the ensemble prediction using decision relevant metrics. We apply the method to improve the prediction of red tide along the West Florida Shelf in the Gulf of Mexico, which affects coastal water quality and has substantial environmental and socioeconomic impacts on the State of Florida. Red tide is a common name for harmful algal blooms that occur worldwide, which result from large concentrations of aquatic microorganisms, such as dinoflagellate Karenia brevis, a toxic single celled protist. We present ensemble method for improving red tide prediction using the high resolution ESMs of the Coupled Model Intercomparison Project Phase 6 (CMIP6) and reanalysis data. The study results highlight the importance of prescreening-based subset selection with decision relevant metrics in identifying non-representative models, understanding their impact on ensemble prediction, and improving the ensemble prediction. These findings are pertinent to other regional environmental management applications and climate services. Additionally, our analysis follows the FAIR Guiding Principles for scientific data management and stewardship such that data and analysis tools are findable, accessible, interoperable, and reusable. As such, the interactive Colab notebooks developed for data analysis are annotated in the paper. This allows for efficient and transparent testing of the results’ sensitivity to different modeling assumptions. Moreover, this research serves as a starting point to build upon for red tide management, using the publicly available CMIP, Coordinated Regional Downscaling Experiment (CORDEX), and reanalysis data.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Prediction by Convolutional Neural Networks of CO 2 /N 2 Selectivity in Porous Carbons from N 2 Adsorption Isotherm at 77 K

Abstract Porous carbons are an important class of porous materials with many applications, including gas separation. An N 2 adsorption isotherm at 77 K is the most widely used approach to characterize porosity. Conventionally, textual properties such as surface area and pore volumes are derived from the N 2 adsorption isotherm at 77 K by fitting it to adsorption theory and then correlating it to gas separation performance (uptake and selectivity). Here the N 2 isotherm at 77 K was used directly as input (representing feature descriptors for the porosity) to train convolutional neural networks to predict gas separation performance (using CO 2 /N 2 as a test case) for porous carbons. The porosity space for porous carbons was explored for higher CO 2 /N 2 selectivity. Porous carbons with a bimodal pore‐size distribution of well‐separated mesopores (3–7 nm) and micropores (<2 nm) were found to be most promising. This work will be useful in guiding experimental research of porous carbons with the desired porosity for gas separation and other applications.

Wang, Song↗
DOE OSTI2020

Prediction by Convolutional Neural Networks of CO 2 /N 2 Selectivity in Porous Carbons from N 2 Adsorption Isotherm at 77 K

Abstract Porous carbons are an important class of porous materials with many applications, including gas separation. An N 2 adsorption isotherm at 77 K is the most widely used approach to characterize porosity. Conventionally, textual properties such as surface area and pore volumes are derived from the N 2 adsorption isotherm at 77 K by fitting it to adsorption theory and then correlating it to gas separation performance (uptake and selectivity). Here the N 2 isotherm at 77 K was used directly as input (representing feature descriptors for the porosity) to train convolutional neural networks to predict gas separation performance (using CO 2 /N 2 as a test case) for porous carbons. The porosity space for porous carbons was explored for higher CO 2 /N 2 selectivity. Porous carbons with a bimodal pore‐size distribution of well‐separated mesopores (3–7 nm) and micropores (<2 nm) were found to be most promising. This work will be useful in guiding experimental research of porous carbons with the desired porosity for gas separation and other applications.

Wang, Song↗
DOE OSTI2025

Hydrodynamic modeling of plasma channel systems for laser plasma accelerators

Structured plasma channels are an essential technology for driving high-gradient, plasma-based acceleration and control of electron and positron beams for advanced concepts accelerators. Laser and gas technologies can permit the generation of long plasma columns known as hydrodynamic, optically-field-ionized (HOFI) channels, which feature low on-axis densities and steep walls. By carefully selecting the background gas and laser properties, one can generate narrow, tunable plasma channels for guiding high intensity laser pulses. Here, we present on the development of simulations of HOFI channels using the FLASH code, a publicly available radiation hydrodynamics code. We explore sensitivities of the channel evolution to laser profile, intensity, and background gas conditions, and identify relevant scalings with laser intensity through a range of practical channel delays.

Accelerators↗
DOE OSTI2024

Label-Free In Situ Chemical Characterization of Amyloid Plaques in Human Brain Tissues

The accumulation of amyloid plaques and increased brain redox burdens are neuropathological hallmarks of Alzheimer’s disease. Altered metabolism of essential biometals is another feature of Alzheimer’s, with amyloid plaques representing sites of disturbed metal homeostasis. Despite these observations, metal-targeting disease treatments have not been therapeutically effective to date. A better understanding of amyloid plaque composition and the role of the metals associated with them is critical. To establish this knowledge, the ability to resolve chemical variations at nanometer length scales relevant to biology is essential. Here, we present a methodology for the label-free, nanoscale chemical characterization of amyloid plaques within human Alzheimer’s disease tissue using synchrotron X-ray spectromicroscopy. Our approach exploits a C–H carbon absorption feature, consistent with the presence of lipids, to visualize amyloid plaques selectively against the tissue background, allowing chemical analysis to be performed without the addition of amyloid dyes that alter the native sample chemistry. Using this approach, we show that amyloid plaques contain elevated levels of calcium, carbonates, and iron compared to the surrounding brain tissue. Chemical analysis of iron within plaques revealed the presence of chemically reduced, low-oxidation-state phases, including ferromagnetic metallic iron. The zero-oxidation state of ferromagnetic iron determines its high chemical reactivity and so may contribute to the redox burden in the Alzheimer’s brain and thus drive neurodegeneration. Ferromagnetic metallic iron has no established physiological function in the brain and may represent a target for therapies designed to lower redox burdens in Alzheimer’s disease. Additionally, ferromagnetic metallic iron has magnetic properties that are distinct from the iron oxide forms predominant in tissue, which might be exploitable for the in vivo detection of amyloid pathologies using magnetically sensitive imaging. We anticipate that this label-free X-ray imaging approach will provide further insights into the chemical composition of amyloid plaques, facilitating better understanding of how plaques influence the course of Alzheimer’s disease.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

High-Resolution Full-Field Structural Microscopy of the Voltage-Induced Filament Formation in VO 2 -Based Neuromorphic Devices

In order to make neuromorphic functions in memristive devices more efficient, information about the structural properties of filament formation at the micro- and mesoscopic scales is necessary. Despite extensive research on VO 2 , a key material due to its filament formation, local operando structural measurements remain challenging and often involve destructive specimen preparation and long rastering times, greatly limiting the scope of experimental studies. Utilizing dark-field X-ray microscopy (DFXM), a fullfield imaging modality, structural signatures of the filament formation process operando are revealed in VO 2 devices. DFXM experiments illustrate that rutile filaments contain isolated monoclinic clusters, indicating structural nonuniformity interior to the filament. The formation of the rutile phase beneath device electrodes was shown to precede filament development, followed by the formation of filament paths guided by nucleation sites within the device. Finally, a medium-term (<30 min) memory mechanism is observed in VO 2 , mediated by sites within the device gap that tend to switch at significantly lower voltages after electrical cycling, a tendency that persists through a brief thermal reset. High spatial resolution, large field-of-view, structure selectivity, and fast signal acquisition of DFXM provided insight into structural features of the filamentary channel and surrounding regions during voltage cycling.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2023

PeakDecoder enables machine learning-based metabolite annotation and accurate profiling in multidimensional mass spectrometry measurements

Multidimensional measurements using state-of-the-art separations and mass spectrometry provide advantages in untargeted metabolomics analyses for studying biological and environmental bio-chemical processes. However, the lack of rapid analytical methods and robust algorithms for these heterogeneous data has limited its application. Here, we develop and evaluate a sensitive and high-throughput analytical and computational workflow to enable accurate metabolite profiling. Our workflow combines liquid chromatography, ion mobility spectrometry and data-independent acquisition mass spectrometry with PeakDecoder, a machine learning-based algorithm that learns to distinguish true co-elution and co-mobility from raw data and calculates metabolite identification error rates. We apply PeakDecoder for metabolite profiling of various engineered strains of Aspergillus pseudoterreus, Aspergillus niger, Pseudomonas putida and Rhodosporidium toruloides. Results, validated manually and against selected reaction monitoring and gas-chromatography platforms, show that 2683 features could be confidently annotated and quantified across 116 microbial sample runs using a library built from 64 standards.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Agentic artificial intelligence for multistage physics experiments at a large-scale user facility particle accelerator

We present a language-model-driven agentic artificial intelligence (AI) system to autonomously execute multistage physics experiments on a production synchrotron light source. Implemented at the Advanced Light Source particle accelerator, the system translates natural language user prompts into structured execution plans that combine archive data retrieval, control-system channel resolution, automated script generation, controlled machine interaction, and analysis. In a representative machine physics task, we show that preparation time was reduced by 2 orders of magnitude relative to manual scripting even for a system expert, while operator-standard safety constraints were strictly upheld. Core architectural features, plan-first orchestration, bounded tool access, and dynamic capability selection, enable transparent, auditable execution with fully reproducible artifacts. These results establish a blueprint for the safe integration of agentic AI into accelerator experiments and demanding machine physics studies, as well as routine operations, with direct portability across accelerators worldwide and, more broadly, to other large-scale scientific infrastructures.

Accelerator/storage ring control systems↗
DOE OSTI2023

Social network structure and the spread of complex contagions from a population genetics perspective

Ideas, behaviors, and opinions spread through social networks. If the probability of spreading to a new individual is a non-linear function of the fraction of the individuals’ affected neighbors, such a spreading process becomes a “complex contagion”. This non-linearity does not typically appear with physically spreading infections, but instead can emerge when the concept that is spreading is subject to game theoretical considerations (e.g. for choices of strategy or behavior) or psychological effects such as social reinforcement and other forms of peer influence (e.g. for ideas, preferences, or opinions). Here we study how the stochastic dynamics of such complex contagions are affected by the underlying network structure. Motivated by simulations of complex contagions on real social networks, we present a framework for analyzing the statistics of contagions with arbitrary non-linear adoption probabilities based on the mathematical tools of population genetics. The central idea is to use an effective lower-dimensional diffusion process to approximate the statistics of the contagion. This leads to a tradeoff between the effects of ”selection” (microscopic tendencies for an idea to spread or die out), random drift, and network structure. Our framework illustrates intuitively several key properties of complex contagions: stronger community structure and network sparsity can significantly enhance the spread, while broad degree distributions dampen the effect of selection compared to random drift. Finally, we show that some structural features can exhibit critical values that demarcate regimes where global contagions become possible for networks of arbitrary size. Our results draw parallels between the competition of genes in a population and memes in a world of minds and ideas. Our tools provide insight into the spread of information, behaviors, and ideas via social influence, and highlight the role of macroscopic network structure in determining their fate.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

Design and Analysis of High Heat Flux Plasma-Facing Components for NSTX Upgrade

The National Spherical Tokamak Experiment (NSTX) upgrade (NSTX-U) requirements lead to enhanced heat loads on plasma-facing components (PFCs) especially in the divertor regions, where normal heat flux density can reach up to 8 MW/m 2 . For these high heat flux (HHF) regions, the PFCs were redesigned, to use castellations which relieved the thermal stresses associated with high incident heat flux. Another design feature of HHF PFCs is the absence of front face mounting holes which create localized areas of high temperature and thermal stress concentrations. Optimized shaping of the front face of the HHF PFCs eliminates regions where the front face is perpendicular to the heat flux direction during normal helicity operation and spreads the heat load. A unique mechanism was designed to connect PFC tiles to the NSTX-U center stack casing using locking features accessible from the low heat flux regions. Isotropic graphite was selected as the HHF PFC tile material. Initial tile design parameters were accessed using analytical expressions for pulsed heat flux loading. A working prototype of the locking mechanism was created during the initial stages of the design to prove the concept performance. This article presents an overview of the divertor HHF PFC design and the results of the thermal and structural analyses performed using ANSYS software. The results of the analyses cover normal operating conditions and disruptions which impose electromagnetic (EM) loads from eddy and halo currents. The 3-D, transient, nonlinear analyses took into account the temperature-dependent properties of the materials, friction interfaces between the parts, and variable electric properties of the parts and interfaces. The results confirm that the HHF PFC tiles remain within the allowable limits for the loads defined by the NSTX-U Recovery Project. In addition, tolerance stack up analyses were performed to ensure tile performance in the worst possible assembly configuration. Finally, the design process was completed successfully, and the NSTX HHF PFC tiles are currently in the production phase.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2022

Comparative genomic and transcriptomic analyses of trans-kingdom pathogen Fusarium solani species complex reveal degrees of compartmentalization

Background: The Fusarium solani species complex (FSSC) comprises fungal pathogens responsible for mortality in a diverse range of animals and plants, but their genome diversity and transcriptome responses in animal pathogenicity remain to be elucidated. We sequenced, assembled and annotated six chromosome-level FSSC clade 3 genomes of aquatic animal and plant host origins. We established a pathosystem and investigated the expression data of F. falciforme and F. keratoplasticum in Chinese softshell turtle (Pelodiscus sinensis) host. Results: Comparative analyses between the FSSC genomes revealed a spectrum of conservation patterns in chromosomes categorised into three compartments: core, fast-core (FC), and lineage-specific (LS). LS chromosomes contribute to variations in genomes size, with up to 42.2% of variations between F. vanettenii strains. Each chromosome compartment varied in structural architectures, with FC and LS chromosomes contain higher proportions of repetitive elements with genes enriched in functions related to pathogenicity and niche expansion. We identified differences in both selection in the coding sequences and DNA methylation levels between genome features and chromosome compartments which suggest a multi-speed evolution that can be traced back to the last common ancestor of Fusarium. We further demonstrated that F. falciforme and F. keratoplasticum are opportunistic pathogens by inoculating P. sinensis eggs and identified differentially expressed genes also associated with plant pathogenicity. These included the most upregulated genes encoding the CFEM (Common in Fungal Extracellular Membrane) domain. Conclusions: The high-quality genome assemblies provided new insights into the evolution of FSSC chromosomes, which also serve as a resource for studies of fungal genome evolution and pathogenesis. This study also establishes an animal model for fungal pathogens of trans-kingdom hosts.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

FY21 Progress Report: SRNL Analysis of ICCWR LCM and WAMS data for Corrosion and Cracking

The development of algorithms for machine learning and data analysis for the 3013 Surveillance Program is a collaborative effort by the Savannah River National Laboratory (SRNL) and the University of South Carolina (USC). For corrosion detection, Laser Confocal Microscope (LCM) or Wide Area 3D Measurement System (WAMS) data is extracted from large binary files, with software written to convert the data to physical attributes (e.g., height, color and grayscale values; all as functions of a location in a plane projection). A user-friendly Matlab Graphical User Interface (GUI) that reads data from either LCM or WAMS files was developed to integrate input data with software developed for processing and evaluation. The GUI can selectively download binary data, interrogate data attributes, label data, flag significant features, execute Machine Learning (ML) algorithms, output parameters for trained ML algorithms, report ML model accuracy with respect to labeled data, and generate graphical representations for various analyses. Features can be called out by user-specified thresholds, manual labeling or machine learning algorithms when they have been completed. The ability to rapidly label data is important because of the volume of data required for training machine learning algorithms. The GUI has the flexibility to allow addition of improved ML algorithms, methods for data visualization, and statistical computations. Statistical analyses via the GUI include areas of pits within a defined range of pit depths, correlations between Red-Green-Blue (RGB) or grayscale intensity and relative surface height, covariances between values associated with features, and feature histograms. The development of supervised machine learning algorithms, however, has been hindered by a lack of training data. The machine learning algorithms for crack identification are being refined but require improvements to the true positive rate for crack detection. This shortcoming is an artifact of the limited training data currently available, perhaps more so than the structure of the neural networks. At present, the best results are had from a consensus over an ensemble of randomly generated Deep Neural Network (DNN) or Convolutional Neural Network (CNN) algorithms. Although the consensus accuracy method has yielded optimum true positive and true negative rates in excess of 80%, additional validation testing is necessary. In addition to the suite of LCM data that was initially used, and which represents the majority of the work presented in this report, WAMS image data was also reviewed at a preliminary level. The review included a comparison between image resolution and dynamic range for each method. WAMS (ZON file) image data was found to have a pixel pitch of 3.69μm compared to 1 μm for the LCM (vk4 file) data, which implies a lower resolution for the WAMS images. Conversely, the ratio of dynamic range of the WAMS data to the LCM data was approximately 41:20 for height data, suggesting that information from WAMS should more accurately determine the depth of pits. At present, the significance of the greater dynamic range of the WAMS data relative to the LCM data has not yet been evaluated.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗