Search NASA⌕ Search

SEARCH · Search NASA

Results for “CLASSES”

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 487 records · Page 27
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Urban morphology and urban water demand evolution in the Los Angeles region

Detailed description of the dataset sources used in this study, the experimental workflow, and plotting for the paper figures provided at the associated GitHub Meta Repo: https://github.com/IMMM-SFA/Ferencz_et_al_2024_ERL The future water demand projections from this study are hypothetical future water demands that reflect the population and urban land cover changes represented by the scenarios considered. The intent and emphasis of this work is investigating the interactions between population change, evolution of urban morphology, and water demand. These projections are not meant to be likely future demands for specific water providers or the LA region and should not be interpreted as such. The folders contain input and output data for each step of the "Recreate my Experiment" workflow described in the associated GitHub meta-repository as well as data used for plotting Figures for the paper that this dataset supports. Description of each folder's contents and use: Step_1a: Inputs to the associated python script provided on the GitHub repo. Step_1b: Inputs (downscaled population rasters) used by the associated python script provided on the GitHub repo. Original 1-km squared rasters that were downscaled also provided. Step_1c: Urban growth projection rasters corresponding to SSP3 and SSP5 population scenarios are provided in separate subfolders as well as the water provider boundaries used for analysis. Outputs of data processing also provided. Associated python script provided on GitHub. Step_1d: Description of Inputs used by the QGIS Model Builder GUI that automates geospatial processing and clipping the of the high-resolution 60 cm land cover data for each urban land class footprint within a defined polygon boundary. The Model Builder is provided on the GitHub repo and can be used by QGIS. The outputs of this step are in "Clipped Provider Hi Res Landcover". If the user wants to use The Model Builder for different regions of LA or to test our outputs, they will need to download the hi resolution landcover raster listed in the Readme and in Ref [2] of the GitHub Page. Step_1e: All necessary inputs to generate average monthly demand over the 2017-2021 period and the minimum and maximum demands over the 2014-2021 for each water provider. Associated python scripts are on GitHub. Step 2: Output data about land cover metrics (areas and fractions) for each urban land class for each water provider. Associated python script on GitHub. Uses outputs from Step 1d "Clipped Provider Hi Res Landcover" Step 3: Both the Inputs for and Outputs from the urban projection raster analysis Python script on GitHub. The inputs are urban land class rasters for specific SSP and zoning scenarios (low, medium, high) from Step 1c. The outputs are rasters of urban pixels that were converted to a higher land class and the number of land class units that changed (Values of 1, 2, or 3). For example, a value of 2 could be LC 21 -> 23 or LC 22 -> 24. These maps are label "intensification." The other outputs are "urban growth" rasters showing the conversion of non urban to urban land, which are indicated by pixel values of 1. These are used for the urban growth change maps in Figure 3. Step 4: Output projections of indoor and outdoor annual and monthly demands for each water provider for the average, minimum, and maximum monthly demand scenarios for each of the four urban growth scenarios (SSP3 med, SSP5 low, SSP5 med, and SSP5 high). The outputs also include metrics on each water provider used for the demand sensitivity analysis presented in Figure 8. Outputs from Step 4 are used for Figures 4 - 8 of the paper. Figures: This folder has data used for plotting Figures 1 through 5, and 8. Data for Figures 6 and 7 are sourced directly from folders associated with the Processing and Analysis Steps 1 - 4. The GitHub meta repository provides descriptions of how each figure was made and the associated plotting scripts used.

Los Angeles↗
DOE OSTI2024

Risk Considerations of Transitioning CO2-EOR Field to CO2 storage Field: Case Study

In the United States (U.S.), carbon dioxide (CO2) injection wells at EOR sites are currently regulated as Class II wells under the U.S. Environmental Protection Agency’s (EPA) Underground Injection Control (UIC) program while dedicated geological CO2 storage (GCS) wells are considered Class VI wells. A CO2-EOR facility considering a transitioning from tertiary oil recovery to injecting CO2 for the primary purpose of long-term storage is required to obtain a Class VI permit where this transition poses an increased risk to underground sources of drinking water. This study considers how transitioning operations from CO2-EOR to storage can impact reservoir plume and pressure transient in the storage envelope, and how these changes could impact area of review and potential unwanted fluid migration. We developed a case study to assess subsurface response and leakage risks associated with a representative, hypothetical operation in a carbonate reservoir. This reservoir transitions from tertiary hydrocarbon recovery to dedicated GCS. Reservoir simulations were run for a set of credible CO2-EOR scenarios to estimate distributions of fluids phases and pressures throughout the model domain after CO2 flooding as well as forecasting the behavior of the reservoir after the transition to a dedicated storage phase. The evolutions from all simulated scenarios were used as the basis for leakage risk quantification using the National Risk Assessment Partnership’s Open-Source Integrated Assessment Model (NRAP-Open-IAM) with a novel reduced-order model to estimate time-dependent leakage of CO2, brine, and hydrocarbon fluids through potentially leaky wells. Results include a description of reservoir response, an estimate of the areal extent that could potentially be impacted by leakage to underground sources of drinking water, and estimates of the magnitude of potential leakage. Considerations for dedicated storage injection well selection, injectivity, and injection scheme performance and potential leakage risk are presented, with implications for risk assessment of well transition discussed. This study presents a risk-based workflow for the Class II to Class VI well transition. Integrating credible numerical simulation of viable CO2-EOR to dedicated CO2 storage with quantitative risk assessment tools, such as the NRAP-Open-IAM, will provide a valuable means to devise operational scenarios and inform decision-making related to storage benefit, leakage risk, and liability. Presented at the SPE/AAPG/SEG Carbon Capture Utilization and Storage Conference in Houston, TX, March 11-13, 2024.

Liu, Guoxiang↗
DOE OSTI2024

Molten Salt Reactor Safety Assessment - Three Approaches to A Common Objective

Safety adequacy assessment is central to nuclear power plant (NPP) licensing. Either NPP accident mitigation or prevention can result in adequate safety. Accident prevention, along with the the prevention of accident escalation, can be evaluated using either deterministic or probabilistic methods. Acceptable means to develop principal design criteria (PDCs) via probabilistic methods are provided in NRC Regulatory Guide 1.233 while advanced reactor design criteria are provided in NRC Regulatory Guide 1.232 (RG 1.232). Accident mitigation-based regulatory guidance for non-power reactors is provided in NUREG 1537. While RG 1.232 included class specific guidance for both sodium-cooled fast reactors and modular high-temperature gas-cooled reactors, it did not provide molten salt reactor (MSR) class specific guidance. Class specific design criteria more closely align with reactors in their class so will require less effort to employ to develop design specific PDCs. The American Nuclear Society has recently released a liquid-fueled MSR design safety standard (ANSI/ANS-20.2-2023) that provides class specific design criteria.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2025

Unconventional Spintronics from Chiral Perovskites

Spintronic devices typically employ heterostructures with ferromagnets which break time-reversal symmetry and have non-vanishing magnetization. With the growing class of materials that support spin-polarized carriers, current, and excitations, it is possible to envision emerging spintronic applications that are not limited to magnetoresistance. Here, chiral perovskites with no net magnetization are considered, where the space-inversion and mirror symmetries are broken to induce chiral structure. The known importance of these perovskites is further expanded by the demonstration of the chiral-induced spin selectivity (CISS). However, the generation of the spin-polarized carriers across the interface with these chiral perovskites remains to be fully understood. These first-principles studies for 2D PbBr4-based chiral perovskites provide their electronic structure and an orbital-based symmetry analysis, which allows to establish an effective Hamiltonian to elucidate the underlying origin of their chirality. The same analysis is used for the Edelstein effect, responsible for electrical generation of the nonequilibrium spin polarization in many materials, which in chiral perovskites can be a mechanism contributing to CISS. Furthermore, by examining optical properties of chiral perovskites and the opportunity to use them to realize tunable altermagnets, another class of zero-magnetization spintronic materials, a versatile materials platform is put forth for unconventional spintronics.

14 SOLAR ENERGY↗
DOE OSTI2024

RrA, an enzyme from Rhodospirillum rubrum , is a prototype of a new family of short‐chain L‐asparaginases

Abstract L‐Asparaginases (ASNases) catalyze the hydrolysis of L‐Asn to L‐Asp and ammonia. Members of the ASNase family are used as drugs in the treatment of leukemia, as well as in the food industry. The protomers of bacterial ASNases typically contain 300–400 amino acids (typical class 1 ASNases). In contrast, the chain of ASNase from Rhodospirillum rubrum , reported here and referred to as RrA, consists of only 172 amino acid residues. RrA is homologous to the N‐terminal domain of typical bacterial class 1 ASNases and exhibits millimolar affinity for L‐Asn. In this study, we demonstrate that RrA belongs to a unique family of cytoplasmic, short‐chain ASNases (scASNases). These proteins occupy a distinct region in the sequence space, separate from the regions typically assigned to class 1 ASNases. The scASNases are present in approximately 7% of eubacterial species, spanning diverse bacterial lineages. They seem to be significantly enriched in species that encode for more than one class 1 ASNase. Here, we report biochemical, biophysical, and structural properties of RrA, a member of scASNases family. Crystal structures of the wild‐type RrA, both with and without bound L‐Asp, as well as structures of several RrA mutants, reveal topologically unique tetramers. Moreover, the active site of one protomer is complemented by two residues (Tyr21 and Asn26) from another protomer. Upon closer inspection, these findings clearly outline scASNases as a stand‐alone subfamily of ASNases that can catalyze the hydrolysis of L‐Asn to L‐Asp despite the lack of the C‐terminal domain that is present in all ASNases described structurally to date.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

$\mathrm{RADICAL}$-Pilot and $\mathrm{PMIx}$/$\mathrm{PRRTE}$: Executing Heterogeneous Workloads at Large Scale on Partitioned $\mathrm{HPC}$ Resources

Execution of heterogeneous workflows on high-performance computing (HPC) platforms present unprecedented resource management and execution coordination challenges for runtime systems. Task heterogeneity increases the complexity of resource and execution management, limiting the scalability and efficiency of workflow execution. Re-source partitioning and distribution of tasks execution over portioned re-sources promises to address those problems but we lack an experimental evaluation of its performance at scale. Here this paper provides a performance evaluation of the Process Management Interface for Exascale (PMIx) and its reference implementation PRRTE on the leadership-class HPC plat-form Summit, when integrated into a pilot-based runtime system called RADICAL-Pilot. We partition resources across multiple PRRTE Distributed Virtual Machine (DVM) environments, responsible for launching tasks via the PMIx interface. We experimentally measure the work-load execution performance in terms of task scheduling/launching rate and distribution of DVM task placement times, DVM startup and termination overheads on the Summit leadership-class HPC platform. Integrated solution with PMIx/PRRTE enables using an abstracted, standardized set of interfaces for orchestrating the launch process, dynamic process management and monitoring capabilities. It extends scaling capabilities allowing to overcome a limitation of other launching mechanisms (e.g., JSM/LSF). Explored different DVM setup configurations provide insights on DVM performance and a layout to leverage it. Our experimental results show that heterogeneous workload of 65,500 tasks on 2048 nodes, and partitioned across 32 DVMs, runs steady with resource utilization not lower than 52%. While having less concurrently executed tasks resource utilization is able to reach up to 85%, based on results of heterogeneous workload of 8200 tasks on 256 nodes and 2 DVMs.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Drive Cycles, Battery Pack Scaling, and Usage Considerations for Long-Haul and Regional-Haul Electric Trucks

Electrifying Class-8 heavy-duty trucks presents a promising opportunity to enhance energy efficiency and reduce freight transport costs. Battery electric trucks (BETs), once considered niche, are gaining traction due to advancements in battery technology and cost reductions. However, accurately predicting battery lifespan under realistic usage conditions remains a key challenge. Understanding battery failure mechanisms and their links to design, operation, and management is essential for developers and fleet operators. This study introduces a method to develop simplified, lab-testable dynamic stress test (DST) cycles for regional and long-haul Class-8 BETs, derived from real-world diesel truck usage. These DSTs enable benchmarking of battery technologies, identification of aging stressors, and optimization of battery design, life, and cost. The approach supports evaluation of key metrics such as levelized cost of driving and total cost of ownership, aiding fair comparisons and adoption decisions. We also propose feasible battery pack sizes that meet current driving demands with strategic charging, and a method to scale pack-level DSTs to cell-level cycles for lab-based testing. These tools facilitate tradeoff analysis across battery chemistries, pack sizing, and charging strategies, while offering means to get insights into battery aging under realistic conditions-ultimately supporting informed BET deployment decisions.

25 ENERGY STORAGE↗
DOE OSTI2019

Non-targeted discovery-based analysis for gas chromatography with mass spectrometry: A comparison of peak table, tile, and pixel-based Fisher ratio analysis

The ability to discover minute differences between samples or sample classes for gas chromatography coupled to mass spectrometry (GC-MS) can be a challenging endeavor, especially when those differences are not a priori . Fisher ratio (F-ratio) analysis is an apt technique to probe the differences between GC-MS chromatograms. F-ratio analysis is a supervised, non-targeted, discovery-based method that compares two different samples (or sample classes) to reduce the GC-MS dataset into a hit list composed of class distinguishing compounds. Three different F-ratio techniques, peak table, tile, and pixel-based were used to “discover” nine non-native analytes that were spiked into gasoline at four different nominal concentrations of 250, 85, 25, 5 parts-per-million (ppm). For the tile and pixel-based F-ratio calculations, a novel methodology is introduced to improve the sensitivity of the F-ratio calculations while reducing false positives. Furthermore, we use a combinatorial technique using null class comparisons, termed null distribution analysis, to determine a statistical F-ratio cutoff for analysis of the hit lists. The pixel-based algorithm was the most sensitive method and was able to “discover” all nine spiked analytes at a nominal concentration of 250 ppm albeit with one false positive interspersed towards the bottom of the hit list. The pixel-based software was also able to “discover” more of the spiked analytes at the lower concentrations with seven of the spiked analytes “discovered” at 85 ppm, four of the spiked analytes “discovered” at 25 ppm, and one analyte “discovered” at 5 ppm.

47 OTHER INSTRUMENTATION↗
DOE OSTI2021

Simultaneous attenuation of trace organics and change in organic matter composition in the hyporheic zone of urban streams

Trace organic compounds (TrOCs) enter rivers with discharge of treated wastewater. These effluents can contain high loads of dissolved organic matter (DOM). In a 48 h field study, we investigated changes in molecular composition of seven DOM compound classes (FTICR-MS) and attenuation of 17 polar TrOCs in a small urban stream receiving treated wastewater. Correlations between TrOCs and DOM were used to identify simultaneous changes in surface water and the hyporheic zone. Changes in TrOC concentrations in surface water ranged between a decrease of 29.2% for methylbenzotriazole and an increase of 152.2% for the transformation product gabapentin-lactam. In the hyporheic zone, only decreasing TrOC concentrations were observed, ranging from 4.9% for primidone to 93.8% for venlafaxine . TrOC attenuation coincided with a decline of molecular diversity of easily biodegradable DOM compound classes while molecular diversity of poorly biodegradable DOM compound classes increased. This concurrence indicates similar or linked attenuation pathways for biodegradable DOM and TrOCs. Strong correlations between TrOCs and DOM compound classes as well as high attenuation of TrOCs primarily occurred in the hyporheic zone. This suggests high potential for DOM turnover and TrOC mitigation in rivers if hyporheic exchange is sufficient.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Long-range structural defects by pathogenic mutations in most severe glucose-6-phosphate dehydrogenase deficiency

Glucose-6-phosphate dehydrogenase (G6PD) deficiency is the most common blood disorder, presenting multiple symptoms, including hemolytic anemia. It affects 400 million people worldwide, with more than 160 single mutations reported in G6PD. The most severe mutations (about 70) are classified as class I, leading to more than 90% loss of activity of the wild-type G6PD. The crystal structure of G6PD reveals these mutations are located away from the active site, concentrating around the noncatalytic NADP + -binding site and the dimer interface. However, the molecular mechanisms of class I mutant dysfunction have remained elusive, hindering the development of efficient therapies. To resolve this, we performed integral structural characterization of five G6PD mutants, including four class I mutants, associated with the noncatalytic NADP + and dimerization, using crystallography, small-angle X-ray scattering (SAXS), cryogenic electron microscopy (cryo-EM), and biophysical analyses. Comparisons with the structure and properties of the wild-type enzyme, together with molecular dynamics simulations, bring forward a universal mechanism for this severe G6PD deficiency due to the class I mutations. We highlight the role of the noncatalytic NADP + -binding site that is crucial for stabilization and ordering two β-strands in the dimer interface, which together communicate these distant structural aberrations to the active site through a network of additional interactions. This understanding elucidates potential paths for drug development targeting G6PD deficiency.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Designing multicomponent hydrides with potential high T c superconductivity

While hydrogen-rich materials have been demonstrated to exhibit high T c superconductivity at high pressures, there is an ongoing search for ternary, quaternary, and more chemically complex hydrides that achieve such high critical temperatures at much lower pressures. First-principles searches are impeded by the computational complexity of solving the Eliashberg equations for large, complex crystal structures. Here, we adopt a simplified approach using electronic indicators previously established to be correlated with superconductivity in hydrides. This is used to study complex hydride structures, which are predicted to exhibit promisingly high critical temperatures for superconductivity. In particular, we propose three classes of hydrides inspired by the Fm$\overline{3}$ m RH 3 structures that exhibit strong hydrogen network connectivity, as defined through the electron localization function. The first class [RH 11 X 3 Y] is based on a Pm$\overline{3}$m structure showing moderately high T c , where the T c estimate from electronic properties is compared with direct Eliashberg calculations and found to be surprisingly accurate. The second class of structures [(RH 11 ) 2 X 6 YZ] improves on this with promisingly high density of states with dominant hydrogen character at the Fermi energy, typically enhancing T c . The third class [(R 1 H 11 )(R 2 H 11 )X 6 YZ] improves the strong hydrogen network connectivity by introducing anisotropy in the hydrogen network through a specific doping pattern. These design principles and associated model structures provide flexibility to optimize both T c and the structural stability of complex hydrides.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2021

A search for ultrahigh-energy neutrinos associated with astrophysical sources using the third flight of ANITA

The ANtarctic Impulsive Transient Antenna (ANITA) long-duration balloon experiment is sensitive to interactions of ultrahigh-energy (E>10 18 eV) neutrinos in the Antarctic ice sheet. The third flight of ANITA, lasting 22 days, began in December 2014. We develop a methodology to search for energetic neutrinos spatially and temporally coincident with potential source classes in ANITA data. This methodology is applied to several source classes: the potential IceCube-identified neutrino sources TXS 0506+056 and NGC 1068, flaring high-energy blazars reported by the Fermi All-Sky Variability Analysis, gamma-ray bursts, and supernovae. Among searches within the five source classes, one candidate was identified as associated with SN 2015D, although not at a statistically significant level. Finally, we proceed to place upper limits on the source classes. We further comment on potential application of this methodology to more sensitive future instruments.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Formation of S0s in extreme environments Ⅲ: the role of environment in the formation pathways

It is well established that there are at least two main channels to form lenticular (or S0) galaxies. The first, which we name ‘faded spiral’ scenario, includes quenching events that led to consumption or removal of gas from a spiral progenitor. The second, which we call ‘merger’ scenario, includes merger-like events and interactions between galaxies. Each scenario leaves characteristic signatures in the newly formed lenticular galaxy. However, the conditions that trigger one mechanism over another are still unknown. This paper is the third of a series aimed at understanding the role of the environment in the formation of lenticular galaxies. In this study, we combine the kinematics, morphology, and properties of the stellar populations of 329 S0s from the SAMI and MaNGA surveys in order to highlight the role of the environment in the process. We divide the S0s into two classes (A and B) according to their global properties, that we can associate to the products of a faded spiral scenario (class A) or a merger scenario (class B). Here we then study how the various classes are distributed within different environments. Our study reveals that the ‘faded spiral’ pathway is the most efficient channel to produce S0s, and it becomes more efficient as the mass of the group or cluster or local density of galaxies increase. The merger pathway is also a viable channel, and its efficiency becomes higher with decreasing local density or environment mass.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Constant-Depth Preparation of Matrix Product States with Adaptive Quantum Circuits

Adaptive quantum circuits, which combine local unitary gates, midcircuit measurements, and feedforward operations, have recently emerged as a promising avenue for efficient state preparation, particularly on near-term quantum devices limited to shallow-depth circuits. Matrix product states (MPS) comprise a significant class of many-body entangled states, efficiently describing the ground states of one-dimensional gapped local Hamiltonians and finding applications in a number of recent quantum algorithms. Recently, it has been shown that the Affleck-Kennedy-Lieb-Tasaki state—a paradigmatic example of an MPS—can be exactly prepared with an adaptive quantum circuit of constant depth, an impossible feat with local unitary gates alone due to its nonzero correlation length [Smith , PRX Quantum 4, 020315 (2023)]. In this work, we broaden the scope of this approach and demonstrate that a diverse class of MPS can be exactly prepared using constant-depth adaptive quantum circuits, outperforming theoretically optimal preparation with unitary circuits. We show that this class includes short- and long-ranged entangled MPS, symmetry-protected topological (SPT) and symmetry-broken states, MPS with finite Abelian, non-Abelian, and continuous symmetries, resource states for MBQC, and families of states with tunable correlation length. Moreover, we illustrate the utility of our framework for designing constant-depth sampling protocols, such as for random MPS or for generating MPS in a particular SPT phase. We present sufficient conditions for particular MPS to be preparable in constant time, with global on-site symmetry playing a pivotal role. Altogether, this work demonstrates the immense promise of adaptive quantum circuits for efficiently preparing many-body entangled states and provides explicit algorithms that outperform known protocols to prepare an essential class of states. Published by the American Physical Society 2024

Smith, Kevin C. (ORCID:0000000223971518)↗
DOE OSTI2020

Evolutionary Variation in MADS Box Dimerization Affects Floral Development and Protein Abundance in Maize

Interactions between MADS-box transcription factors are critical in the regulation of floral development, and shifting MADS-box protein-protein interactions are predicted to have influenced floral evolution. However, precisely how evolutionary variation in protein-protein interactions affects MADS-box protein function remains unknown. To assess the impact of changing MADS-box protein-protein interactions on transcription factor function, we turned to the grasses, where interactions between B-class MADS-box proteins vary. We tested the functional consequences of this evolutionary variability using maize as an experimental system. We found that differential B-class dimerization was associated with subtle, quantitative differences in stamen shape. In contrast, differential dimerization resulted in large-scale changes to downstream gene expression. Differential dimerization also affected B-class complex composition and abundance, independent of transcript levels. This indicates that differential B-class dimerization affects protein degradation, revealing an important consequence for evolutionary variability in MADS-box interactions. Our results highlight complexity in the evolution of developmental gene networks - changing protein-protein interactions could affect not only the composition of transcription factor complexes, but also their degradation and persistence in developing flowers. Here, our results also show how coding change in a pleiotropic master regulator could have small, quantitative effects on development.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Global Explainability of A Deep Abstaining Classifier for Cancer Pathology Reports

We present a global explainability method to characterize sources of errors in a real-world multitask deep abstaining classifier (DAC), in the context of cancer histology prediction. Our multitask classifier, currently deployed for automated annotation of cancer pathology reports from NCI-SEER registries, was trained and evaluated on 1.04 million hand-annotated samples and makes simultaneous predictions of cancer site, subsite, histology, laterality, and behavior for each report. The DAC framework enables the model to abstain on ambiguous reports and confusing classes to achieve the target accuracy on the retained (non-abstained) samples, but at the cost of decreased coverage. Requiring 97% accuracy on the histology task caused our model to retain only 22% of all samples, mostly the less ambiguous and common classes. Local explainability with the GradInp technique provided a computationally efficient way of obtaining contextual reasoning for hundreds of thousands of individual predictions. Our method, involving dimensionality reduction of approximately 13000 aggregated local explanations (ALE), offers a tractable path to true global explainability. It enabled identification of sources of errors in histology classification, globally, as hierarchical complexity among classes, label noise, insufficient information, and conflicting evidence. This suggests several strategies for iterative improvement of our DAC, including well-designed exclusion criteria, focused annotation, and reduced penalties for errors involving hierarchically related classes.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Characterizing Quantum Classifier Utility in Natural Language Processing Workflows

Quantum Natural Language Processing (QNLP) develops natural language processing (NLP) models for deployment on quantum computers. We explore feature and data prototype selection techniques to address challenges posed by encoding high dimensional features. Our study builds quantum circuit classifiers that includes classical feature pre-processing, quantum embedding and quantum model training. The quantum models are built on 4 or 6 qubits and the quantum neural network (QNN) uses the established bricklayer design. We compare the dependence of model performance (in terms of accuracy and F1 scores) on feature length, embedding gates and parameterized unitary design. We compare the performance of quantum machine learning models to classical convolution neural network model (CNN) on binary and multi-class classification tasks using two datasets of synthetic features and labels. The first is the ECP-CANDLE P3B3 dataset a corpus of synthetically generated cancer pathology reports. The second dataset is extracted from well-known benchmark dataset (MADELON) - features are generated with a combination of informative, repeated and uninformative features. Both datasets are used for binary classification and multi-class classification with 3 classes. We observe robust, accurate performance from all models on the binary classification tasks, but multiclass classification is a challenge for the quantum models-there is a notable decrease in accuracy when using 3 classes. Overall the performance is comparable in terms of recall and accuracy between QNNs and CNNs, even with large datasets. These results provide a point of comparison between quantum and classical models on real-world datasets.

Hamilton, Kathleen↗
DOE OSTI2020

Time Series Classification for Locating Forced Oscillation Sources

Here, this article presents a machine learning based time-series classification method for using synchrophasor measurements to locate the source of forced oscillation (FO) for fast disturbance removal. First, multivariate time series (MTS) matrices are constructed by the most informative measurements selected by sequential feature selection from each power plant. Then, the Mahalanobis matrix is trained such that the Mahalanobis distance between the MTSs from the same class (i.e., with the same FO source location) are minimized and from different classes (i.e., with different FO source locations) are maximized. This allows MTSs to be classified by classifiers with class membership corresponding to the location of each FO source. To meet the runtime requirements of online matching, class templates are constructed to reduce data size and improve matching efficiency. To account for uncertainty in identifying the exact beginning of an FO event, dynamic time warping is used to align the out-of-sync MTSs. IEEE 39bus and WECC 179bus systems are used for algorithm development and validation. Simulation results demonstrate that the algorithm meets online operation runtime requirement with high accuracy using misaligned data sets.

42 ENGINEERING↗