Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “classification models”

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 505 records · Page 28
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Leak Detection and Sensor Importance Within a Solvent Extraction Process Abstract

In anticipation of the Special Nuclear Material test bed (Beartooth), Idaho National Laboratory has developed a smaller, multi-sensor system for analyzing the solvent extraction process. These systems will allow for research into nuclear fuel processing operations. The multi-sensor system consists of a row of centrifugal contactors and allows for measurement sources that are not traditionally used in the solvent extraction process to be explored including temperature, vibration, acoustics, pH, color, flow, and motor current. Currently, the solvent extraction process is very labor intensive and requires vigilant operators to identify the occurrence of leaks, which can be common during startup or after any change to the system. This study aims to locate leaks using non-traditional measurement sources then to identify which signals were of greatest importance in making this classification using Local Interpretable Model-agnostic Explanations. These results can be used to help solvent extraction process operators detect leaks and to inform future test beds designers to which sensors contain relevant, actionable information in this scenario.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
NASA NTRS1985

Maximum likelihood classification of synthetic aperture radar imagery

Classification of synthetic aperture radar (SAR) images has important applications in geology, agriculture, and the military. A statistical model for SAR images is reviewed and a maximum likelihood classification algorithm developed for the classification of agricultural fields based on the model. It is first assumed that the target feature information is known a priori. The performance of the algorithm is then evaluated in terms of the probability of incorrect classification. A technique is also presented to extract the needed feature information from a SAR image; then both the feature extraction and the maximum likelihood classification algorithms are tested on a SEASAT-A SAR image.

Frost, V. S.↗
DOE OSTI2025

Unlocking Scholarly Insights: Leveraging Machine Learning Approaches for Citation Analysis and Intent Classification

Publicly funded organizations, notably institutions like the Los Alamos National Laboratory (LANL), are deeply vested in acquiring robust productivity metrics to gauge the entirety of their research output. Motivated by the imperative to enhance institutional productivity assessment, this study investigates the utilization of Large Language Models (LLM) such as BERT-based models, as well as local LLaMa-30b-instruct and Mixtral-8x7b-instruct architectures for classifying type of URL referenced resources in academic papers such as software, dataset, as well as authorship intent. Challenges in discerning resource types from context are highlighted, along with the potential of BERT and LLMs to address these challenges. Through comprehensive analysis, this research unveils a notable surge in documents featuring URL citations, indicative of the escalating importance of digital resources in scholarly publications. Moreover, citations to datasets and software demonstrate consistent growth over time, underscoring their increasing significance. Our findings also reveal that LANL authors contribute substantially to accessible science, comprising about 10% of dataset and software mentions in LANL

Large Language Models, BERT, citation classificati↗
DOE OSTI2021

NukeLM: Pre-Trained and Fine-Tuned Language Models for the Nuclear and Energy Domains

Natural language processing (NLP) tasks (text classification, named entity recognition, etc.) have seen amazing improvements over the last few years. This is due to models such as BERT that achieve deep knowledge transfer by using a large pre-trained model, then fine-tuning the model on specific tasks. The BERT architecture has shown even better performance on domain-specific tasks when the model is pre-trained using domain-relevant texts. Here, inspired by these recent advancements, we have developed NukeLM, a nuclear-domain BERT model pre-trained on 1.5 million abstracts from the DOE Office of Scientific and Technical Information (OSTI) database. This NukeLM model is then fine-tuned for the classification of research articles into either binary classes (related to the nuclear fuel cycle (NFC) or not) or multiple categories related to the subject of the article. We show that continued pre-training of a BERT-style architecture prior to fine-tuning results in greater performance in both article classification tasks. This information is critical for properly triaging manuscripts, a necessary task for better understanding citation networks that publish in the nuclear space and uncovering new areas of research in the nuclear (or nuclear relevant) domain.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
NASA NTRS2009

Lossless Compression of Classification-Map Data

A lossless image-data-compression algorithm intended specifically for application to classification-map data is based on prediction, context modeling, and entropy coding. The algorithm was formulated, in consideration of the differences between classification maps and ordinary images of natural scenes, so as to be capable of compressing classification- map data more effectively than do general-purpose image-data-compression algorithms. Classification maps are typically generated from remote-sensing images acquired by instruments aboard aircraft (see figure) and spacecraft. A classification map is a synthetic image that summarizes information derived from one or more original remote-sensing image(s) of a scene. The value assigned to each pixel in such a map is the index of a class that represents some type of content deduced from the original image data for example, a type of vegetation, a mineral, or a body of water at the corresponding location in the scene. When classification maps are generated onboard the aircraft or spacecraft, it is desirable to compress the classification-map data in order to reduce the volume of data that must be transmitted to a ground station.

Hua, Xie↗
DOE OSTI2023

The LSST AGN Data Challenge: Selection Methods

Abstract Development of the Rubin Observatory Legacy Survey of Space and Time (LSST) includes a series of Data Challenges (DCs) arranged by various LSST Scientific Collaborations that are taking place during the project's preoperational phase. The AGN Science Collaboration Data Challenge (AGNSC-DC) is a partial prototype of the expected LSST data on active galactic nuclei (AGNs), aimed at validating machine learning approaches for AGN selection and characterization in large surveys like LSST. The AGNSC-DC took place in 2021, focusing on accuracy, robustness, and scalability. The training and the blinded data sets were constructed to mimic the future LSST release catalogs using the data from the Sloan Digital Sky Survey Stripe 82 region and the XMM-Newton Large Scale Structure Survey region. Data features were divided into astrometry, photometry, color, morphology, redshift, and class label with the addition of variability features and images. We present the results of four submitted solutions to DCs using both classical and machine learning methods. We systematically test the performance of supervised models (support vector machine, random forest, extreme gradient boosting, artificial neural network, convolutional neural network) and unsupervised ones (deep embedding clustering) when applied to the problem of classifying/clustering sources as stars, galaxies, or AGNs. We obtained classification accuracy of 97.5% for supervised models and clustering accuracy of 96.0% for unsupervised ones and 95.0% with a classic approach for a blinded data set. We find that variability features significantly improve the accuracy of the trained models, and correlation analysis among different bands enables a fast and inexpensive first-order selection of quasar candidates.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2022

System Modeling Frameworks for Wind Turbines and Plants: Review and Requirements Specifications

System modeling frameworks for wind turbines and plants are used by research groups and industry to design wind energy systems that take into account key trade-offs across performance, cost, and reliability at both the turbine and plant level. The frameworks are exercised using a variety of multi-disciplinary design, analysis and optimization (MDAO) methods. To improve inter-operability and foster collaboration, this report proposes a classification system for the frameworks along dimensions of model fidelity and scope. The classification system is first motivated with reviews the state-of-the-art in the development of software frameworks for integrated wind turbine and plant simulation. Within each major wind turbine and power plant subsystem, a matrix is developed for the disciplines used and the fidelity levels with which each discipline can be modeled. The existing frameworks are then classified according to the matrix. Next, an ontology is proposed that will allow for standardizing how data is transferred between the most common discipline-fidelity combinations used in the frameworks. A common representation of data creates the ability to 1) share system descriptions and analysis results, supporting more transparent benchmarks and comparison, and 2) integrate models together into workflows within and across organizations for improving the efficiency and performance of wind turbine and power plant design processes. Ultimately, this integration leads to better overall wind energy system designs with high performance and low costs.

17 WIND ENERGY↗
DOE OSTI2025

Leveraging large language models to automate the identification of healthcare access barriers for veterans

Objective: To develop and evaluate an automated system for identifying healthcare barriers focusing on transportation issues in veterans’ clinical notes using large language models (LLMs) and to assess the impact of different prompting strategies on classification performance and explanation consistency. Methods: We developed a hybrid system combining pattern matching for templated notes with LLM analysis for free-text notes. Using 2000 manually annotated clinical notes, we compared four prompting strategies (dual-role short, dual-role long, analysis-first, analysis-only) across Mistral-7B and Llama-3.1 models. We evaluated classification performance using standard metrics and assessed explanation consistency through embedding similarity analysis. Results: The analysis-first strategy achieved superior performance, with Mistral-7B reaching an F1 score of 0.914, outperforming traditional machine learning approaches (GBM: 0.786, BERT: 0.811). LLMs demonstrated higher explanation consistency within models (mean cosine similarity 0.887–0.908) compared to cross-model similarities (0.767–0.872). Pattern matching successfully handled 6.7% of templated notes deterministically. Mistral-7B showed greater internal consistency but higher abstention rates compared to Llama-3.1. Conclusion: Requiring LLMs to analyze evidence before classification improves both accuracy and explanation consistency for identifying transportation barriers in clinical notes. This approach enables automated barrier detection at scale while providing clinically relevant explanations, supporting both population-level healthcare planning and individual patient care decisions.

Healthcare access barriers↗
DOE OSTI2023

Attention-Augmented Parametric Kernel Graph Neural Network (APKGNN) for Node Classification

We present a new graph neural network, the Attention-based Parametric-Kernel augmented Graph Neural Network (APKGNN), developed for node classification tasks. Despite extensive work on modeling multi-faceted relationships between connected nodes of a graph, the effect of attention on edge features mapped to relationships has not yet been analyzed through learning representation. This study derives such an attention vector by first calculating node features corresponding to endpoints of an edge and then aggregating these with extracted local intrinsic patches of a given graph to generate augmented local patch vectors. This process uses a parametric kernel based on Gaussian mixture models (GMMs) to embed local neighborhoods of the graph in local patches. The patch vectors then convolve with the above node features to produce an updated node representation. We show that this new learning representation (APKGNN) achieves higher node classification accuracy on tasks - both standard benchmarks (Cora, PubMed, Citeseer) and new experimental short text corpora where nodes correspond to text documents and words. This implementation of the GNN convolution layer outperforms state-of-the-art (SOTA) algorithms, achieving higher training, validation, and test accuracy by a significant margin on three standard benchmark data sets under both SOTA experimental settings and those for new testbeds.

Bose, Avishek↗
NASA NTRS1996

Modeling Global Biogenic Emission of Isoprene: Exploration of Model Drivers

Vegetation provides the major source of isoprene emission to the atmosphere. We present a modeling approach to estimate global biogenic isoprene emission. The isoprene flux model is linked to a process-based computer simulation model of biogenic trace-gas fluxes that operates on scales that link regional and global data sets and ecosystem nutrient transformations Isoprene emission estimates are determined from estimates of ecosystem specific biomass, emission factors, and algorithms based on light and temperature. Our approach differs from an existing modeling framework by including the process-based global model for terrestrial ecosystem production, satellite derived ecosystem classification, and isoprene emission measurements from a tropical deciduous forest. We explore the sensitivity of model estimates to input parameters. The resulting emission products from the global 1 degree x 1 degree coverage provided by the satellite datasets and the process model allow flux estimations across large spatial scales and enable direct linkage to atmospheric models of trace-gas transport and transformation.

Alexander, Susan E.↗
DOE OSTI2022

Uncertainty Analysis in Multi‐Sector Systems: Considerations for Risk Analysis, Projection, and Planning for Complex Systems

Abstract Simulation models of multi‐sector systems are increasingly used to understand societal resilience to climate and economic shocks and change. However, multi‐sector systems are also subject to numerous uncertainties that prevent the direct application of simulation models for prediction and planning, particularly when extrapolating past behavior to a nonstationary future. Recent studies have developed a combination of methods to characterize, attribute, and quantify these uncertainties for both single‐ and multi‐sector systems. Here, we review challenges and complications to the idealized goal of fully quantifying all uncertainties in a multi‐sector model and their interactions with policy design as they emerge at different stages of analysis: (a) inference and model calibration; (b) projecting future outcomes; and (c) scenario discovery and identification of risk regimes. We also identify potential methods and research opportunities to help navigate the tradeoffs inherent in uncertainty analyses for complex systems. During this discussion, we provide a classification of uncertainty types and discuss model coupling frameworks to support interdisciplinary collaboration on multi‐sector dynamics (MSD) research. Finally, we conclude with recommendations for best practices to ensure that MSD research can be properly contextualized with respect to the underlying uncertainties.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

An Introduction to Word Embeddings and Language Models

Language models have advanced at a phenomenal pace over the past decade. This document provides a short introduction to terminology, word embeddings (aka low-dimensional representations), and popular large-scale language models (LMs). Word embeddings are used to represent words as numerical vectors and are context-independent, meaning a word can only have a single representation (e.g., club can only be club sandwich, not golf club ). Language models can determine the probability of a given sequence of words occurring in a sentence and can provide context to distinguish between words and phrases that sound similar. LMs are context-dependent (e.g., club can be club sandwich or golf club ) and largely fall in two main classes – autoregressive and autoencoding models. Autoregressive models are pretrained on the classic language modeling task: guess the next token having read all the previous ones. Those models can be fine-tuned and achieve great results on many tasks, the most natural application is text generation. A typical example of such models is GPT, but others include GPT-2, GPT-3, CTLR, TRANSFORMER-XL, REFORMER, XLNET. Autoencoding models are pretrained by corrupting the input tokens in some way and trying to reconstruct the original sentence. They can be fine-tuned and achieve great results on many tasks such as text generation, but their most natural application is sentence classification or token classification. A typical example of such models is BERT, but others include ROBERTA, ALBERT, XML, XML-ROBERTA, FLAUBERT AND LONGFORMER.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1986

Spectral evolution of young stellar objects

An evolutionary sequence, from protostars to pre-main-sequence stars, is suggested for the classification of young stellar objects. This sequence is derived by comparing the predictions of the theoretical models of Adams and Shu with the morphological classification scheme of Lada and Wilking. We first define the spectral index in the near- and mid-infrared, n is identical to d log(nu F sub nu)/d log nu, and then interpret the class of sources with negative spectral indices as protostars. The inferred mass infall rates for these objects are generally consistent with the measured gas temperatures of approx. 35 K in Ophiuchus, and of approx. 10 K in Taurus. Fitting the data requires us to adopt cloud rotation rates in Ophiuchus which are typically an order of magnitude greater than in Taurus, and we speculate on the mechanistic origin for this difference. Next, we consider a subclass of T Tauri stars with near- and mid-infrared excesses and positive or zero spectral indices. We find that the objects with the steeper indices can be understood as the post-infall products from the collapse of rotating cloud cores, where the infrared excesses arise from the simple reprocessing of visible stellar photons in optically thick but spatially thin disks. The sources with flatter spectra may require massive accretion disks. Given the existence of protostars and naked star/disk systems, there is a natural interpretation of another subclass of T Tauri stars, those with two peaks in their emergent spectral energy distributions. These are readily explained as intermediate cases in which dust envelopes still surround the stars and disks. Finally, we find that the theory can be extended to explain the spectral energy distribution of FU Orionis, a famous outburst source. Our model suggests that FU Orionis has a disk, but it offers no discrimination between the competing ideas that the outburst took place on the star or in the disk.

Adams, F. C.↗
DOE OSTI2021

Deep Learning for Automated Detection and Identification of Migrating American Eel Anguilla rostrata from Imaging Sonar Data

Adult American eels (Anguilla rostrata) are vulnerable to hydropower turbine mortality during outmigration from growth habitat in inland waters to the ocean where they spawn. Imaging sonar is a reliable and proven technology for monitoring of fish passage and migration; however, there is no efficient automated method for eel detection. We designed a deep learning model for automated detection of adult American eels from sonar data. The method employs convolution neural network (CNN) to distinguish between 14 images of eels and non-eel objects. Prior to image classification with CNN, background subtraction and wavelet denoising were applied to enhance sonar images. The CNN model was first trained and tested on data obtained from a laboratory experiment, which yielded overall accuracies of >98% for image-based classification. Then, the model was trained and tested on field data that were obtained near the Iroquois Dam located on the St. Lawrence River; the accuracy achieved was commensurate with that of human experts.

13 HYDRO ENERGY↗
DOE OSTI2021

The agglomeration and dispersion dichotomy of human settlements on Earth

Human settlements on Earth are scattered in a multitude of shapes, sizes and spatial arrangements. These patterns are often not random but a result of complex geographical, cultural, economic and historical processes that have profound human and ecological impacts. However, little is known about the global distribution of these patterns and the spatial forces that creates them. This study analyses human settlements from high-resolution satellite imagery and provides a global classification of spatial patterns. We find two emerging classes, namely agglomeration and dispersion. In the former, settlements are fewer than expected based on the predictions of scaling theory, while an unexpectedly high number of settlements characterizes the latter. To explain the observed spatial patterns, we propose a model that combines two agglomeration forces and simulates human settlements’ historical growth. Our results show that our model accurately matches the observed global classification (F1: 0.73), helps to understand and estimate the growth of human settlements and, in turn, the distribution and physical dynamics of all human settlements on Earth, from small villages to cities.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Neural architecture codesign for fast physics applications

We develop a pipeline to streamline neural architecture codesign for physics applications to reduce the need for ML expertise when designing models for novel tasks. Our method employs neural architecture search and network compression in a two-stage approach to discover hardware efficient models. This approach consists of a global search stage that explores a wide range of architectures while considering hardware constraints, followed by a local search stage that fine-tunes and compresses the most promising candidates. We exceed performance on various tasks and show further speedup through model compression techniques such as quantization-aware-training and neural network pruning. We synthesize the optimal models to high level synthesis code for FPGA deployment with the hls4ml library. Additionally, our hierarchical search space provides greater flexibility in optimization, which can easily extend to other tasks and domains. We demonstrate this with two case studies: Bragg peak finding in materials science and jet classification in high energy physics, achieving models with improved accuracy, smaller latencies, or reduced resource utilization relative to the baseline models.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
NASA NTRSDate not supplied

Satellite-Derived Imagery and Transfer Learning: A Novel Technique for Land Cover Classification

Land cover classification is a continuing research topic due to its relevance to land use and land cover changes from impacts such as climate change, agriculture, urbanization, and hazardous weather. Simple access to frequently changing land cover classifications could provide knowledge and decision support to various researchers and agencies across the globe for each of above-mentioned and related influences. This research aims to provide a novel technique for land cover classification of remote sensing imagery; harnessing Artificial Neural Networks and transfer learning (TL). Knowledge sharing techniques within machine learning are typically utilized when training datasets are sparse or transitions between data modalities is required. In this case, two data modalities, multi and hyperspectral data, are considered for knowledge transfer. The large number of continuous spectral bands available from hyperspectral sensors typically provide increased sophisticated land classification capability compared to more traditional multispectral imagery with limited discrete spectral bands. However, large-scale, frequent access to hyperspectral imagery is relatively limited. The proposed classification technique would therefore prove useful in regions in which hyperspectral data are not readily available for classification but multispectral data are. Image segmentation models are trained on each multi and hyperspectral datasets. Knowledge sharing techniques are then applied to each model to understand what knowledge, if any, is gained when moving between the data modalities. The datasets utilized for training and testing a U-Net model include the European Space Agency’s multispectral imager Sentinel-2 and the hyperspectral German Aerospace Center’s Earth Sensing Imaging Spectrometer (DESIS). Initially, only two classes, land and water, are classified for simplicity. However, more complex classes can be added if knowledge sharing is successful. The workflow for image classification through supervised image segmentation with a U-Net model will be discussed along with metrics calculated before and after TL for both Sentinel-2 and DESIS data are applied. Additionally, future steps to advance the sophistication of this technique as well as other applicable methodologies will be explored.

Emily Foshee↗
DOE OSTI2024

United States Multi-Sector Dynamics land use and land cover base maps to support Human-Earth System Modeling

Datasets are land use and land cover (LULC) rasterized base maps at 30-m resolution for the conterminous United States (CONUS) for the years 2008, 2011, 2016, and 2019. Separate base maps are provided where LULC classifications are thematically congruent with Community Land Model (CLM), Land Use Harmonization (LUH2), and Global Change Analysis Model (GCAM), and a detailed decomposition of all combined land classes into a Multisector Dynamics (MSD) LULC product. Base maps were developed using empirically derived satellite (National Land Cover Dataset, MODIS) and combined observation datasets (Crop Data Layer, Protected Areas Database) and represent the most up-to-date accurate information on LULC in the CONUS. The four datasets encompass four different landcover classification systems: MSD Layers - The raw landcover classes obtained from reclassifying NLCD and USDA Crop data layers into a respective landcover class GCAM Layers - The MSD classes mosaiced, reclassified, and combined into the respective GCAM landcover classes CLM Layers - Similar process to GCAM layers, but mosaiced, reclassified, and combined MSD layers to their respective PFT classes LUH2 Layers - Similar process to both GCAM and CLM Layers, but mosaiced, reclassified and combined the MSD layers to align with the respective states

Food↗