Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “classification”

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 1,657 records · Page 92
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

The Dark Machines Anomaly Score Challenge: Benchmark Data and Model Independent Event Classification for the Large Hadron Collider

We describe the outcome of a data challenge conducted as part of the Dark Machines (https://www.darkmachines.org) initiative and the Les Houches 2019 workshop on Physics at TeV colliders. The challenged aims to detect signals of new physics at the Large Hadron Collider (LHC) using unsupervised machine learning algorithms. First, we propose how an anomaly score could be implemented to define model-independent signal regions in LHC searches. We define and describe a large benchmark dataset, consisting of >1 billion simulated LHC events corresponding to 10\, fb^{-1} 10 f b − 1 of proton-proton collisions at a center-of-mass energy of 13 TeV. We then review a wide range of anomaly detection and density estimation algorithms, developed in the context of the data challenge, and we measure their performance in a set of realistic analysis environments. We draw a number of useful conclusions that will aid the development of unsupervised new physics searches during the third run of the LHC, and provide our benchmark dataset for future studies at https://www.phenoMLdata.org. Code to reproduce the analysis is provided at https://github.com/bostdiek/DarkMachines-UnsupervisedChallenge.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2020

Classification of all $\mathcal{N}\geq 3$ moduli space orbifold geometries at rank 2

We classify orbifold geometries which can be interpreted as moduli spaces of four-dimensional \mathcal{N}\geq 3 𝒩 ≥ 3 superconformal field theories up to rank 2 (complex dimension 6). The large majority of the geometries we find correspond to moduli spaces of known theories or discretely gauged version of them. Remarkably, we find 6 geometries which are not realized by any known theory, of which 3 have an \mathcal{N}=2 𝒩 = 2 Coulomb branch slice with a non-freely generated coordinate ring, suggesting the existence of new, exotic \mathcal{N}=3 𝒩 = 3 theories.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2015

Tissue Classification

The project began as a e ort to support InLight and Lumidigm. With the sale of the companies to a non-New Mexico entity, the project then focused on supporting a new company Medici Technologies. The Small Business (SB) is attempting to quantify glucose in tissue using a series of short interferometer scans of the nger. Each scan is produced from a novel presentation of the nger to the device. The intent of the project is to identify and, if possible, implement improved methods for classi cation, feature selection, and training to improve the performance of predictive algorithms used for tissue classi cation.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2017

Seismic classification through sparse filter dictionaries

We tackle a multi-label classi cation problem involving the relation between acoustic- pro le features and the measured seismogram. To isolate components of the seismo- grams unique to each class of acoustic pro le we build dictionaries of convolutional lters. The convolutional- lter dictionaries for the individual classes are then combined into a large dictionary for the entire seismogram set. A given seismogram is classi ed by computing its representation in the large dictionary and then comparing reconstruction accuracy with this representation using each of the sub-dictionaries. The sub-dictionary with the minimal reconstruction error identi es the seismogram class.

58 GEOSCIENCES↗
DOE OSTI2016

Classification of background suppression profiles for low background RPM data

Suppression of the ambient gamma background radiation by traffic structure and cargo is a well-understood and studied effect for deployed radiation portal monitors (RPM). For effective analysis of measured RPM profiles with the objective of inferring the spatial characteristics of radiation sources, it is important to account for the effects of background suppression. In this report we analyze background suppression for a test dataset from vehicle RPMs at a sample port and estimate the distributions of suppression amplitudes and shapes. Cluster analysis of standardized and normalized profiles is used to obtain the dominant suppression shapes in the sample field data. We determine that a large fraction of non-alarm RPM occupancies are represented by a small number of suppression shapes. This fraction increases when the signal-to-noise ratio of an occupancy profile is improved by the addition of signals for multiple RPM detectors located at the same height. The calculated suppression shapes from RPM data can be used along with source models in the process of spatial profile analysis both in the field or offline. This background suppression analysis is an important step in improving the effectiveness of the RPM profile analysis methodology which is currently being investigated and may lead to methods that reduce the number of secondary inspections as well as to decision support tools that aid operators in evaluating RPM data that do not contain spectral information.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2016

MODAL CLASSIFICATION: Processing and Anomaly Detection in a Noisy Vibrational Environment

n this report we summarize the development of a model-based detection scheme capable of the on-line detection of structural anomalies applying both system identification methods to extract a modal model and state estimation/detection methods to both detect an anomaly and locate all affected sensor positions specified by measurement location. Background information evolving from a simple single input/single output (SISO) mechanical system introducing the state-space formulation which leads directly to a multiple input/multiple output (MIMO) state-space formulation is discussed.From that representation the modal state-space model is introduced through an eigenvalue-eigenvector formulation leading to a similarity transformation and complex modes. Next the sampled-data state-space system is introduced leading to discrete state-space systems which provide the basis for the system identification methods to follow.

42 ENGINEERING↗
DOE OSTI2019

Carbon Life Cycle Analysis of CO 2 -EOR for Net Carbon Negative Oil (NCNO) Classification (Final Report)

This final report summarizes the work that was conducted to achieve the project’s general objective of developing a clear and repeatable methodology to determine whether the oil produced in a conventional CO 2 enhanced oil recovery (CO 2 -EOR) operation can be classified as Net Carbon Negative Oil (NCNO). The report also summarizes the reservoir mass accounting methodology and the monitoring, verification, and accounting (MVA) methodology, which are critical elements that support the general objective and were conceived as project goals. Because the purpose is to classify the EOR product (crude oil) and not the EOR project itself, and as the rate of crude oil production varies significantly with time, the methodology uses a novel dynamic carbon lifecycle analysis (d-LCA) that links instant energy demand and associated greenhouse gas (GHG) emissions to instant operational performance. The EOR operational performance is assessed through CO 2 utilization rates, which relate usage of CO 2 to oil production This dynamic method provides a better understanding of the evolution of the environmental impact (CO 2 emissions) and mitigation (geologic CO 2 storage) associated with an expanded carbon capture, utilization and storage (CCUS) system, from start to closure of operations. The dynamic interplay between operational and environmental performance forms the basis of our CCUS technology analysis. The LCA part of the work is meant to complement previous NETL LCA work, and so it focused on EOR/storage efficiency, its variability through time, and the impact of this variability on LCA results. The environmental impact of the CCUS system was assessed for 3 system boundaries: (1) gate-to-gate, (2) gate-to-grave, and (3) cradle-to-grave. The environmental impact was also assessed considering the electricity displacement from other generation facilities. Another goal of our study was to use results to make recommendations on EOR field development strategies that are conducive to producing NCNO. Because field operational strategies have a significant impact on reservoir engineering parameters that affect both CO 2 storage and oil production (e.g., sweep efficiency, flood conformance, fluid saturation distribution), we conducted a scenario analysis that assesses the operational and environmental performance of four common and novel CO 2 -EOR field development strategies. Results show that all CO 2 -EOR evaluated scenarios start operating with a negative carbon footprint and, years into the project, transition into operating with a positive carbon footprint. Having achieved a result that proves that all CO 2 -EOR operations produce NCNO during the first years of production is critical in the context of the urgency of climate change mitigation. Transition points are significantly different in each scenario. ii In a cradle-to-grave CCUS system, if the capture facility displaces a conventional power source and credits are added for displacement, the adverse impact caused by the system expansion not only is reversed but transformed into a significant environmental benefit. The effect of the displacement more than compensates the emissions associated with the electricity generation requirements for the production of CO 2 , as much more electricity will be displaced. Results provide an understanding of the evolution of the system’s net carbon balance of all four field development strategies studied. This information can be useful to CO 2 -EOR operators seeking value in storing more CO 2 through a carbon credit program (e.g., the 45Q federal carbon credit program). Most importantly, this study serves as confirmation that CO 2 -EOR can be operationally designed to both enhance oil production and reduce greenhouse gas emissions to the atmosphere.

02 PETROLEUM↗
DOE OSTI2018

Toward Uncertainty Quantification for Supervised Classification

Our goal is to develop a general theoretical basis for quantifying uncertainty in supervised machine learning models. Current machine learning accuracy-based validation metrics indicate how well a classifier performs on a given data set as a whole. However, these metrics do not tell us a model's efficacy in predicting particular samples. We quantify uncertainty by constructing probability distributions of the predictions made by an ensemble of classifiers. This report details our initial investigations into uncertainty quantification for supervised machine learning. We apply an uncertainty analysis to the problem of malicious website detection. Machine learning models can be trained to find suspicious characteristics in the text of a website's Uniform Resource Locator (URL). However, given the vast numbers of URLs and the ever changing tactics of malicious actors, it will always be possible to find sets of websites which are outliers with respect to a model's hypothesis. Therefore, we seek to understand a model's per-sample reliability when classifying URL data. Acknowledgements This work was funded by the Sandia National Laboratories Laboratory Directed Research and Development (LDRD) program.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Classification of Methane Emissions from Industrial Meters, Vintage vs Modern Plastic Pipe, and Plastic-lined Steel and Cast-Iron Pipe (Final Report)

The project focus was to improve the characterization of methane (CH 4 ) emissions currently detailed in the U.S. Environmental Protection Agency (EPA) Greenhouse Gas Inventory (GHGI) for categories of assets within the Natural Gas (NG) distribution system to isolate what is driving inflation or deflation behind current, nationwide, aggregated Emission Factors (EFs). These specific categories included industrial/commercial NG customer meters, modern and vintage plastic pipe, along with plastic-lined steel and cast-iron pipe. CH 4 emission rate data was gathered during 25 sampling campaigns at industrial/commercial meter sites, vintage and modern plastic pipe sites, as well as plastic-lined steel and cast-iron sites located within NG distribution networks.

03 NATURAL GAS↗
DOE OSTI2019

Initial Development of an Improved Creep-Fatigue Design Method that Avoids the Separate Evaluation of Creep and Fatigue Damage and Eliminates the Requirement for Stress Classification

This report describes a method for capturing the effect of strain range, temperature, hold time, and elastic follow up on the cyclic life of a material undergoing uniaxial load. Developing design correlations of this type requires methods for extrapolating experimental data as a function of hold time and follow up factor. Various extrapolation methods are developed and discussed here and recommendations are developed for the final approaches to be adopted for design use. The report also describes comparisons between the new creep-fatigue design method, described here, and current creep-fatigue design methods in the ASME Boiler and Pressure Vessel Code and associated nuclear Code Cases. These comparisons demonstrate several advantages of the new method: it is easier to execute and is less over conservative than many current approaches. This report, along with a companion on extending these results to multiaxial load and on combining load cycles, represents the accumulation of several years of work on the development of this new approach to creep-fatigue design, combining an elastic perfectly plastic analysis method with simplified model tests measuring the effect of elastic follow up on creep-fatigue cyclic life. The conclusions to this report describe a path forward to developing the required design data for high temperature advanced reactor structural materials and guiding the adoption of the new method in the ASME Boiler and Pressure Vessel Code and related standards.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Multiclass classification experiments

We seek a multiclass classifier that can satisfy the following requirements. 1. Predict among at least three classes. 2. Handle a moderately large number of correlated predictors. 3. Handle mixed categorical and continuous predictors. 4. Handle missing values (in some way). 5. Be trained with about n = 1,000 responses, and 6. Quantify prediction uncertainty. Based on 5. above, it is assumed that accuracy of prediction is paramount (relative to, say, runtime). All methods considered in this document take on the order of minutes (most take seconds) on a 2.8 GHz Quad-Core Intel Core i7 processor with 16GB RAM. Secondly, quantification of uncertainty is assumed to be a secondary task, since, with this small data size, resampling methods (i.e. bootstrapping) are possible. Further studies should quantify the feasibility and accuracy of resampling methods, and/or other methods, for quantifying uncertainty. R was used for all software.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Applying Machine Learning to the Classification of DC-DC Converters (Milestone 2 Deliverable Report)

Since extending the Autodetector to a convolutional neural network (CNN) machine learning classifier model, an effort has been executed to demonstrate its ability to distinguish not only a switching DC-DC converter as high voltage, but identify the make and model of a converter on which it was trained. This was achieved by collecting data in a noisy environment, pre-processing the time domain data to obtain composite images using a method that improves upon that of the prior research, then validating a trained CNN model to an accuracy of 100% on a selected candidate converter.

42 ENGINEERING↗
DOE OSTI2020

Data product development for cold cloud and precipitation process analysis/Snow regime classifications from the NSA snow product

This grant funded work focused first on developing a data product for cold-cloud precipitation and second, under a continuation, on examining distinct snowfall regimes occurring at the North Slope Alaska (NSA) facility. The overarching goal under the first phase was to provide a data product that quantifies the vertically- and temporally-resolved properties of cold, precipitating clouds and that will enable future analyses of the processes that control their formation and evolution. The data product derives from observations at the NSA Utqiagvik (BRW) and Oliktok Point (OLI) sites, its algorithm utilizing radar observations supplemented with in situ and other collocated remote sensing measurements to constrain an optimal estimation retrieval.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Applying Machine Learning to the Classification of DC-DC Converters. NA-22 Final Report

Tools are now available that enable measurement electromagnetic radiation (EMR) from active electronics in an item. This radiation may be intended WIFI or cellular network links, for example or unintended such as the switching noise generated by DC-to-DC converters. It would be extremely valuable to have the capability to discriminate between the low-voltage DC-to-DC converters or other digital noise prevalent in most modern electronics, versus the high-voltage DC-to-DC converters used in utility firesets. Previous work performed under a Sandia Laboratory Directed Research and Development (LDRD) project on Charge State Detection using a deep neural network has been continued in this effort. A state-of-the-art supervised machine learning algorithm has not only been extended to discriminate between low and high voltage converters but has been validated in determining a converters make and model.

42 ENGINEERING↗