SEARCH · Search NASA

Results for “deep data”

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.

SourceResearch connectionFrom yearThrough year

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

At least 361 records · Page 20

All sources NASA NTRS DOE OSTI

DOE OSTI2021

Data-Driven Multi-agent Deep Reinforcement Learning for Distribution System Decentralized Voltage Control with High Penetration of PVs

This paper proposes a novel model-free/data-driven centralized training and decentralized execution multi-agent deep reinforcement learning (MADRL) framework for distribution system voltage control with high penetration of PVs. The proposed MADRL can coordinate both the real and reactive power control of PVs with existing static var compensators and battery storage systems. Unlike the existing DRL-based voltage control methods, our proposed method does not rely on a system model during both the training and execution stages. This is achieved by developing a new interaction scheme between the surrogate modeling of the original system and the multi-agent soft actor critic (MASAC) MADRL algorithm. In particular, the sparse pseudo-Gaussian process with a few-shots of measurements is utilized to construct the surrogate model of the original environment, i.e., power flow model. This is a data-driven process and no model parameters are needed. Furthermore, the MASAC enabled MADRL allows to achieve better scalability by dividing the original system into different voltage control regions with the aid of real and reactive power sensitivities to voltage, where each region is treated as an agent. This also serves as the foundation for the centralized training and decentralized execution, thus significantly reducing the communication requirements as only local measurements are required for control. Comparative results with other alternatives on the IEEE 123-nodes and 342-nodes systems demonstrate the superiority of the proposed method.

14 SOLAR ENERGY↗

DOE OSTI2023

NETL Plastic Pipes Project (Final Report)

Plastic or composite pipelines have been the bane of the utility locating industry because they are neither conductive nor magnetic which are the properties traditionally used to locate buried utilities. Ground penetrating radar (GPR) is an effective geophysical tool for locating plastic/composite pipelines where resistive cover allows for adequate penetration of radar energy. However, GPR has limited applicability in areas where the soil cover is conductive due to significant clay and/or salt content. This study examines complementary near-surface geophysical methods that are potentially useful for locating buried plastic/composite pipelines, either singly or in combination. Specifically, this modeling study used computational numerical methods to forward model the response of GPR, resistivity, seismic, gravity gradiometry, and photoacoustic/thermoacoustic imaging methods to plastic/composite pipelines for various scenarios including: (1) pipe diameters ranging between 2 in. to 12 in.; (2) burial depths ranging between 3 ft. to 4 ft.; (3) various degrees in contrast in physical properties (i.e., electrical permittivity, elasticity, resistivity, density); and (4) various experimental acquisition choices (e.g., GPR radar and seismic source frequencies, electrode spacing). Numerical modeling performed herein reconfirmed that GPR is the preferred method for detecting/locating plastic pipelines. A caveat for GPR detection is that the material covering the plastic pipe (trench fill material and adjacent soil) must be sufficiently resistive to allow the two-way propagation to the required depth of investigation and back to the surface. GPR was the only method modeled in this study that can be used to directly detect plastic pipelines of 2-in.-diameter and larger when buried 3-ft-deep. GPR data processing and imaging also can determine pipe depth, pipe diameter, trench dimensions, and moisture conditions. Seismic modeling results suggest that direct detection of a 12-in.-diameter plastic pipe at 3-ft.-depth may be possible under favorable conditions; however, the associated signature would be weak (e.g., surface- to S-wave, backscattered surface-waves, and/or forward scattered surface-waves to S-wave). Direct pipe detection under field conditions with noise and strong lateral geologic heterogeneity is doubtful. Numerical modeling also suggests that plastic pipelines can be indirectly located by detecting the trench in which they are buried. GPR, direct current (DC) resistivity, and seismic methods have the potential to locate the pipeline trench if there is sufficient contrast between the trench-wall and trench-fill materials for the physical property being measured by each method (i.e., electrical permittivity for GPR; resistivity for DC resistivity; or density, compressional velocity, or shear velocity for seismic). Modeling also indicated that currently available (commercial) gravity gradiometers would be unable to directly detect/locate plastic pipelines ≤ 8-in.-diameter when buried 3-ft.-deep given the typical instrument noise floor for field surveying as well as the expected density variations due to geologic heterogeneity. The numerical modeling performed in this project did not identify a universal geophysical technology that can locate buried plastic pipelines in all parts of the United States (although GPR is suggested for all areas with resistive cover). However, the project results suggest that a towed land streamer simultaneously acquiring multiple geophysical data types including multi-offset GPR, multi-channel DC resistivity, seismic geophone- and/or distributed acoustic sensing (DAS), and potentially photoacoustic/thermoacoustic data would be an appropriate platform for locating buried plastic pipeline. Moreover, the complementary multiphysics data acquired by a towed land streamer would permit the use of joint and/or cooperative inversion frameworks for a more rigorous and consistent data interpretation.

42 ENGINEERING↗

DOE OSTI2022

Advances in Machine and Deep Learning for Modeling and Real-time Detection of Multi-Messenger Sources

We live in momentous times. The science community is empowered with an arsenal of cosmic messengers to study the universe in unprecedented detail. Gravitational waves, electromagnetic waves, neutrinos, and cosmic rays cover a wide range of wavelengths and timescales. Combining and processing these datasets that vary in volume, speed, and dimensionality requires new modes of instrument coordination, funding, and international collaboration with a specialized human and technological infrastructure. In tandem with the advent of large-scale scientific facilities, the last decade has experienced an unprecedented transformation in computing and signal-processing algorithms. The combination of graphics processing units, deep learning, and the availability of open source, high-quality datasets has powered the rise of artificial intelligence. This digital revolution now powers a multibillion dollar industry, with far-reaching implications in technology and society. In this chapter, we describe pioneering efforts to adapt artificial intelligence algorithms to address computational grand challenges in multi-messenger astrophysics. We review the rapid evolution of these disruptive algorithms, from the first class of algorithms introduced in early 2017 to the sophisticated algorithms that now incorporate domain expertise in their architectural design and optimization schemes. We discuss the importance of scientific visualization and extreme-scale computing in reducing time-to-insight and obtaining new knowledge from the interplay between models and data.

Artificial Intelligence↗

DOE OSTI2022

Differentiable, Learnable, Regionalized Process-Based Models With Multiphysical Outputs can Approach State-Of-The-Art Hydrologic Prediction Accuracy

Predictions of hydrologic variables across the entire water cycle have significant value for water resources management as well as downstream applications such as ecosystem and water quality modeling. Recently, purely data-driven deep learning models like long short-term memory (LSTM) showed seemingly insurmountable performance in modeling rainfall runoff and other geoscientific variables, yet they cannot predict untrained physical variables and remain challenging to interpret. Here, we show that differentiable, learnable, process-based models (called δ models here) can approach the performance level of LSTM for the intensively observed variable (streamflow) with regionalized parameterization. We use a simple hydrologic model HBV as the backbone and use embedded neural networks, which can only be trained in a differentiable programming framework, to parameterize, enhance, or replace the process-based model's modules. Without using an ensemble or post-processor, δ models can obtain a median Nash-Sutcliffe efficiency of 0.732 for 671 basins across the USA for the Daymet forcing data set, compared to 0.748 from a state-of-the-art LSTM model with the same setup. For another forcing data set, the difference is even smaller: 0.715 versus 0.722. Meanwhile, the resulting learnable process-based models can output a full set of untrained variables, for example, soil and groundwater storage, snowpack, evapotranspiration, and baseflow, and can later be constrained by their observations. Both simulated evapotranspiration and fraction of discharge from baseflow agreed decently with alternative estimates. The general framework can work with models with various process complexity and opens up the path for learning physics from big data.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2022

Towards automating structural discovery in scanning transmission electron microscopy *

Abstract Scanning transmission electron microscopy is now the primary tool for exploring functional materials on the atomic level. Often, features of interest are highly localized in specific regions in the material, such as ferroelectric domain walls, extended defects, or second phase inclusions. Selecting regions to image for structural and chemical discovery via atomically resolved imaging has traditionally proceeded via human operators making semi-informed judgements on sampling locations and parameters. Recent efforts at automation for structural and physical discovery have pointed towards the use of ‘active learning’ methods that utilize Bayesian optimization with surrogate models to quickly find relevant regions of interest. Yet despite the potential importance of this direction, there is a general lack of certainty in selecting relevant control algorithms and how to balance a priori knowledge of the material system with knowledge derived during experimentation. Here we address this gap by developing the automated experiment workflows with several combinations to both illustrate the effects of these choices and demonstrate the tradeoffs associated with each in terms of accuracy, robustness, and susceptibility to hyperparameters for structural discovery. We discuss possible methods to build descriptors using the raw image data and deep learning based semantic segmentation, as well as the implementation of variational autoencoder based representation. Furthermore, each workflow is applied to a range of feature sizes including NiO pillars within a La:SrMnO 3 matrix, ferroelectric domains in BiFeO 3 , and topological defects in graphene. The code developed in this manuscript is open sourced and will be released at github.com/nccreang/AE_Workflows .

47 OTHER INSTRUMENTATION↗

DOE OSTI2020

Candidate Periodically Variable Quasars from the Dark Energy Survey and the Sloan Digital Sky Survey

Periodically variable quasars have been suggested as close binary supermassive black holes. We present a systematic search for periodic light curves in 625 spectroscopically confirmed quasars with a median redshift of 1.8 in a 4.6 deg 2 overlapping region of the Dark Energy Survey Supernova (DES-SN) fields and the Sloan Digital Sky Survey Stripe 82 (SDSS-S82). Our sample has a unique 20-year long multi-color ( griz ) light curve enabled by combining DES-SN Y6 observations with archival SDSS-S82 data. The deep imaging allows us to search for periodic light curves in less luminous quasars (down to r ~23.5 mag) powered by less massive black holes (with masses ≳ 10 8.5 M ⊙ ) at high redshift for the first time. We find five candidates with significant (at >99.74% single-frequency significance in at least two bands with a global p-value of ~7 × 10 -4 –3× 10 -3 accounting for the look-elsewhere effect) periodicity with observed periods of ~3–5 years (i.e., 1–2 years in rest frame) having ~4–6 cycles spanned by the observations. If all five candidates are periodically variable quasars, this translates into a detection rate of |${\sim }0.8^{+0.5}_{-0.3}$| % or |${\sim }1.1^{+0.7}_{-0.5}$| quasar per deg 2 . Our detection rate is 4–80 times larger than those found by previous searches using shallower surveys over larger areas. This discrepancy is likely caused by differences in the quasar populations probed and the survey data qualities. Furthermore, we discuss implications on the future direct detection of low-frequency gravitational waves. Additionally, continued photometric monitoring will further assess the robustness and characteristics of these candidate periodic quasars to determine their physical origins.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2022

Real-Time Cavity Fault Prediction in CEBAF Using Deep Learning

Data-driven prediction of future faults is a major research area for many industrial applications. In this work, we present a new procedure of real-time fault prediction for superconducting radio-frequency (SRF) cavities at the Continuous Electron Beam Accelerator Facility (CEBAF) using deep learning. CEBAF has been afflicted by frequent downtime caused by SRF cavity faults. We perform fault prediction using pre-fault RF signals from C100-type cryomodules. Using the pre-fault signal information, the new algorithm predicts the type of cavity fault before the actual onset. The early prediction may enable potential mitigation strategies to prevent the fault. In our work, we apply a two-stage fault prediction pipeline. In the first stage, a model distinguishes between faulty and normal signals using a U-Net deep learning architecture. In the second stage of the network, signals flagged as faulty by the first model are classified into one of seven fault types based on learned signatures in the data. Initial results show that our model can successfully predict most fault types 200 ms before onset. We will discuss reasons for poor model performance on specific fault types.

Rahman, M.↗

NASA NTRS1977

Absolute flux density calibrations of radio sources: 2.3 GHz

A detailed description of a NASA/JPL Deep Space Network program to improve S-band gain calibrations of large aperture antennas is reported. The program is considered unique in at least three ways; first, absolute gain calibrations of high quality suppressed-sidelobe dual mode horns first provide a high accuracy foundation to the foundation to the program. Second, a very careful transfer calibration technique using an artificial far-field coherent-wave source was used to accurately obtain the gain of one large (26 m) aperture. Third, using the calibrated large aperture directly, the absolute flux density of five selected galactic and extragalactic natural radio sources was determined with an absolute accuracy better than 2 percent, now quoted at the familiar 1 sigma confidence level. The follow-on considerations to apply these results to an operational network of ground antennas are discussed. It is concluded that absolute gain accuracies within + or - 0.30 to 0.40 db are possible, depending primarily on the repeatability (scatter) in the field data from Deep Space Network user stations.

Freiley, A. J.↗

DOE OSTI2022

Achieving robust somatic mutation detection with deep learning models derived from reference data sets of a cancer sample

Accurate detection of somatic mutations is challenging but critical in understanding cancer formation, progression, and treatment. We recently proposed NeuSomatic, the first deep convolutional neural network-based somatic mutation detection approach, and demonstrated performance advantages on in silico data. In this study, we use the first comprehensive and well-characterized somatic reference data sets from the SEQC2 consortium to investigate best practices for using a deep learning framework in cancer mutation detection. Using the high-confidence somatic mutations established for a cancer cell line by the consortium, we identify the best strategy for building robust models on multiple data sets derived from samples representing real scenarios, for example, a model trained on a combination of real and spike-in mutations had the highest average performance. The strategy identified in our study achieved high robustness across multiple sequencing technologies for fresh and FFPE DNA input, varying tumor/normal purities, and different coverages, with significant superiority over conventional detection approaches in general, as well as in challenging situations such as low coverage, low variant allele frequency, DNA damage, and difficult genomic regions.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2020

Real-time Intrusion Detection for High-bandwidth Research Networks using Unsupervised Deep Learning

Nowadays, data generated by large-scale scientific experiments are on the scale of petabytes per month. These data are transferred through dedicated high-bandwidth networks across distributed collaboration sites for processing, storage, and analysis. Like general purpose networks, research networks experience attacks. However, monitoring traffic in such high-bandwidth network environments is challenging with conventional deep packet inspection (DPI)- based network intrusion detection systems (NIDS). Recently, machine learning (ML) algorithms have shown potential in detecting anomalous patterns with better generalizability and faster speed. Compared to the general purpose networks, traffic movement in research network is less volatile and more likely to benefit from the ML-based anomaly analysis. We present ThunderSecure, an efficient AI-powered system for detecting anomalies in 100G research networks. ThunderSecure implements a high-throughput packet processing infrastructure using multi-cores and GPUs. It takes time-series statistics from a variety of network measurements and uses 2D Convolutional Neural Network (CNN) to explore both spatial and temporal dependencies in network data stream. Patterns of science data traffic are learned with a one-class neural network, which blends an Adversarial Autoencoder (AAE) with Gaussian mixture density estimation. Testing traffic flows exhibiting significant deviations from the learned baseline of normality are marked as anomalies. We trained ThunderSecure on hundreds of billions of science data packets recorded from a 100G research network at Fermi National Accelerator Laboratory. The detection performance was evaluated on traffic captured from the same research network days and weeks after the training with different types of attack flows injected. To the best of our knowledge, this is the first ML work in the area of network anomaly detection that has been validated on such extreme scale datasets. Results show that ThunderSecure can recognize science data traffic that are captured long after the training and make nearly certain detection on those with anomalous flows injected, even in the case when the anomaly-to-normal mixing ratio is 0.1%.

Gong, Qian↗

DOE OSTI2018

Electron mass enhancement and magnetic phase separation near the Mott transition in double-layer ruthenates

Here, we present a detailed investigation of the specific heat of Ca 3 (Ru 1-x M x ) 2 O 7 (M = Ti, Fe, Mn) single crystals. Depending on the dopant and doping level, three distinct regions are present: a quasitwo-dimensional metallic state with antiferromagnetic (AFM) order formed by ferromagnetic bilayers (AFM-b), a Mott insulating state with G-type AFM order (G-AFM), and a localized state with a mixed AFM-b and G-AFM phase. Our specific heat data provide deep insights into the Mott transitions induced by Ti and Mn doping. We observed not only an anomalous large mass enhancement, but also an additional term in the specific heat, i.e., C ∝ T 2 , in the localized region. The C ∝ T 2 term is most likely due to long-wavelength excitations with both FM and AFM components. A decrease in the Debye temperature is observed in the G-type AFM region, indicating lattice softening associated with the Mott transition.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

DOE OSTI2018

Microstructure evolution and mechanical performance of nickel based superalloy C1023 at elevated temperatures

Highlights: • Growth and strengthening mechanisms of carbide in superalloy C1023 were investigated. • Carbide dendrite primary arm grows selectively based on the vertex of blocky carbonitride. • Monolayer or multilayer carbide dendrite arm has certain crystal relationship with γ′ and γ. • Plastic deformation of γ hinders interior crack propagating from MC to matrix at 650 °C. • Oxidation of GBs significantly weakens the tensile strength of the superalloy at 850 °C. - Abstract: Microstructure evolution of polycrystalline nickel based superalloy C1023 casting was investigated by comparing as cast superalloy and heat treated superalloy in the paper. Influences of strengtheners and carbides on the tensile behaviors of heat treated superalloy at different temperatures (20 °C, 650 °C, and 850 °C) were studied. Resulting data from deep corrosion suggested that growth mode of scripted carbide included nucleation on the vertex of blocky carbonitride, expansion of dendrite arm in C-rich area, and enlargement of dendrite cap. Decomposition of MC carbide generated numerous octahedral M{sub 23}C{sub 6} carbide as well as nanoscale Cr-Mo-rich nuclei. Diffusion of Mo from γ to precipitates such as γ', MC, and M{sub 23}C{sub 6}, occurred during the heat treatment while irregular γ' (4 nm–500 nm) transformed into cuboidal particles. At 650 °C, no significant deterioration of tensile strength was found while plastic deformation ability of γ base was improved. Oxidation of grain boundaries (GBs) dominated the crack behavior of the superalloy at 850 °C.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

DOE OSTI2024

Enhancing molecular design efficiency: Uniting language models and generative networks with genetic algorithms

This study examines the effectiveness of generative models in drug discovery, material science, and polymer science, aiming to overcome constraints associated with traditional inverse design methods relying on heuristic rules. Generative models generate synthetic data resembling real data, enabling deep learning model training without extensive labeled datasets. They prove valuable in creating virtual libraries of molecules for material science and facilitating drug discovery by generating molecules with specific properties. While generative adversarial networks (GANs) are explored for these purposes, mode collapse restricts their efficacy, limiting novel structure variability. To address this, we introduce a masked language model (LM) inspired by natural language processing. Although LMs alone can have inherent limitations, we propose a hybrid architecture combining LMs and GANs to efficiently generate new molecules, demonstrating superior performance over standalone masked LMs, particularly for smaller population sizes. This hybrid LM-GAN architecture enhances efficiency in optimizing properties and generating novel samples.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Determination of the strong coupling and its running from measurements of inclusive jet production

The value of the strong coupling α S is determined in a comprehensive analysis at next-to-next-to-leading order accuracy in quantum chromodynamics. The analysis uses double-differential cross section measurements from the CMS Collaboration at the CERN LHC of inclusive jet production in proton-proton collisions at centre-of-mass energies of 2.76, 7, 8, and 13 TeV, combined with inclusive deep-inelastic data from HERA. The value α S (m z ) = 0.1176 $^{+0.0014}_{-0.0016}$ is obtained at the scale of the Z boson mass. By using the measurements in different intervals of jet transverse momentum, the running of α S is probed for energies between 100 and 1600 GeV.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2020

The importance of temporal and spatial incoherence in quantitative interpretation of 4D-STEM

Recent developments in pixelated detectors, when combined with aberration correction of probe forming optics have greatly enhanced the field of scanning electron diffraction. Differential phase contrast is now routine and deep learning has been proposed as a method to extract maximum information from diffraction patterns. This study examines the effects of temporal and spatial incoherence on convergent beam electron diffraction patterns and demonstrates that simple center of mass measurements cannot be naively interpreted. The inclusion of incoherence in deep learning data sets is also discussed.

36 MATERIALS SCIENCE↗

DOE OSTI2025

A Mass Conservation Relaxed (MCR) LSTM Model for Streamflow Simulation Across CONUS

The recent development of the physics-aware Mass-Conserving Long Short-Term Memory network (MC-LSTM) provides an alternative to other data-driven Deep Learning (DL) models in hydrology. Mass-Conserving Long Short-Term Memory incorporates mass conservation directly into the LSTM architecture. Despite the theoretical advancements, studies have reported a surprisingly limited performance of the MC-LSTM in streamflow simulation. We hypothesize that such a limitation is due to the unrealistic mass conservation scheme in MC-LSTM, which overlooks unobserved incoming water fluxes beyond precipitation. As an attempt to verify this hypothesis, we propose a Mass Conservation Relaxed LSTM (MCR-LSTM), which incorporates a bi-directional mass relaxation (MR) component to account for potential incoming water fluxes beyond precipitation. We train and test the proposed MCR-LSTM model across 531 watersheds in the contiguous United States (CONUS) against three baseline models: the Sacramento Soil Moisture Accounting, LSTM, and MC-LSTM. Our results show that MCR-LSTM outperforms MC-LSTM despite its underperformance compared to LSTM. Specifically, MCR-LSTM's advantage over MC-LSTM is mainly seen in the Plains and Western U.S., where the newly incorporated MR component better simulates water loss and suggests the likely existence of additional incoming water fluxes beyond precipitation, respectively. The novelty and contribution of this study are twofold: firstly, it introduces an alternative physics-aware DL tool (i.e., MCR-LSTM) in hydrology with higher accuracy in specific regions compared to MC-LSTM. Secondly, it provides a diagnosis of regions where strict, precipitation-based mass conservation constraints may be unrealistic in streamflow simulation.

deep learning↗

DOE OSTI2024

RNA language models predict mutations that improve RNA function

Structured RNA lies at the heart of many central biological processes, from gene expression to catalysis. RNA structure prediction is not yet possible due to a lack of high-quality reference data associated with organismal phenotypes that could inform RNA function. We present GARNET (Gtdb Acquired RNa with Environmental Temperatures), a new database for RNA structural and functional analysis anchored to the Genome Taxonomy Database (GTDB). GARNET links RNA sequences to experimental and predicted optimal growth temperatures of GTDB reference organisms. Using GARNET, we develop sequence- and structure-aware RNA generative models, with overlapping triplet tokenization providing optimal encoding for a GPT-like model. Leveraging hyperthermophilic RNAs in GARNET and these RNA generative models, we identify mutations in ribosomal RNA that confer increased thermostability to the Escherichia coli ribosome. The GTDB-derived data and deep learning models presented here provide a foundation for understanding the connections between RNA sequence, structure, and function.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2023

Microbial carbon use efficiency promotes global soil carbon storage

Soils store more carbon than other terrestrial ecosystems. How soil organic carbon (SOC) forms and persists remains uncertain, which makes it challenging to understand how it will respond to climatic change. It has been suggested that soil microorganisms play an important role in SOC formation, preservation and loss. Although microorganisms affect the accumulation and loss of soil organic matter through many pathways, microbial carbon use efficiency (CUE) is an integrative metric that can capture the balance of these processes. Although CUE has the potential to act as a predictor of variation in SOC storage, the role of CUE in SOC persistence remains unresolved. Here we examine the relationship between CUE and the preservation of SOC, and interactions with climate, vegetation and edaphic properties, using a combination of global-scale datasets, a microbial-process explicit model, data assimilation, deep learning and meta-analysis. We find that CUE is at least four times as important as other evaluated factors, such as carbon input, decomposition or vertical transport, in determining SOC storage and its spatial variation across the globe. In addition, CUE shows a positive correlation with SOC content. Our findings point to microbial CUE as a major determinant of global SOC storage. Understanding the microbial processes underlying CUE and their environmental dependence may help the prediction of SOC feedback to a changing climate.

63 RADIATION, THERMAL, AND OTHER ENVIRON. POLLUTAN↗