Search NASA⌕ Search

SEARCH · Search NASA

Results for “Deep potential”

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 361 records · Page 20
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

DLIO: A DATA-CENTRIC BENCHMARK FOR DEEP LEARNING APPLICATIONS

SF-22-136 Deep learning has been shown as a successful method for various tasks, and its popularity results in numerous open-source deep learning software tools. Deep learning has been applied to a broad spectrum of scientific domains such as cosmology, particle physics, computer vision, fusion, and astrophysics. Scientists have performed a great deal of work to optimize the computational performance of deep learning frameworks. However, the same cannot be said for I/O performance. As deep learning algorithms rely on big-data volume and variety to effectively train neural networks accurately, I/O is a significant bottleneck on large-scale distributed deep learning training. DLIO, is a novel representative benchmark suite built based on the I/O profiling of the selected workloads. DLIO can be utilized to accurately emulate the I/O behavior of modern deep learning applications. Using DLIO, application developers and system software solution architects can identify potential I/O bottlenecks in their applications and guide optimizations to boost the I/O performance leading to lower training times. The storage vendor can also use DLIO as a guide for designing and optimize the storage and filesystem targeting at deep learning application.

ZHENG, HUIHUO↗
DOE OSTI2020

Potential for Transport of Cesium as Bio-colloids in a High Ionic Strength System

The Waste Isolation Pilot Plant (WIPP) is a deep geologic repository for long-term disposal of transuranic, radioactive waste that is a byproduct from the nation's nuclear defense program. Most likely release scenario → human intrusion (potentially through drilling) (US DOE 1995; 1996). Brine may proceed through the Rustler formation (the most transmissive layer) (Perkins et al., 1999). There is a need to investigate halophilic microbes including their mobility in the Rustler formation and the potential for enhancing transport of cesium as a bio-colloid. Objectives: To quantify the change in mobility of Cs by interaction with microbes (Chromohalobacter) that may be present near Waste Isolation Pilot Plant. What is Chromohalobacter?: a halophilic microbe, isolated from near the WIPP site, able to thrive in the high salt concentrations relevant to the repository. Does the microbe growth phase effect uptake of Cs? Experiments conducted with actively growing and stationary phase Chromohalobacter. Uptake may differ if Cs{sup +} is mistaken for K{sup +} in an actively growing (Log) phase versus stationary phase microbes (not actively reproducing). Materials: 1'' Teflon column packed with 1 gram of dolomite [355-500 μm, CaMg(CO{sub 3}){sub 2}] connected to a syringe pump and fraction collector via Teflon tubing. Chromohalobacter (100 million cells/mL) and dolomite collected near the WIPP. Synthetic brine - 15% NaCl (w/v ∼2.78 M) + 3 mM NaHCO{sub 3}. Methodology: Negative Control 5,000 ppb Cs initially injected into mini-columns with brine. Injection solution was then switched to only brine. Cs Experiments with Stationary Phase Chromo: Set 1: Viable stationary Chromo injected into minicolumns after reaction with 200 ppb Cs. Injection solution was then switched to only brine (without Cs or microbes). Set 2: Stressed stationary Chromo initially injected into mini-columns after reaction with 200 ppb Cs. Then, switched to brine only. Cs Experiments with Log phase Chromo: Set 1: Chromo injected into mini-columns after reaction with 5,000 ppb Cs. Injection then switched to only brine (without Cs or microbes). Set 2: Chromo grown with 5,000 ppb Cs and spiked with additional Cs before injection into columns. Preliminary Conclusions: Cs was not taken up or mobilized by Chromohalobacter in the actively growing or stationary phases or under stressed conditions. Competition may have occurred between Cs and K{sup +} at lower ionic strength, significantly greater concentrations of K{sup +} were present. Results suggest that Chromohalobacter can selectively uptake K{sup +} in the presence of Cs{sup +}. Future Research: K{sup +} and Cs{sup +} uptake will be monitored at high ionic strength in K{sup +} depleted media to consider uptake of Cs in the absence of K.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2023

Bayesian, Multifidelity Operator Learning for Complex Engineering Systems–A Position Paper

Abstract Deep learning has significantly improved the state-of-the-art in computer vision and natural language processing, and holds great potential to design effective tools for predicting and simulating complex engineering systems. In particular, scientific machine learning seeks to apply the power of deep learning to scientific and engineering tasks, with operator learning (OL) emerging as a particularly effective tool. OL can approximate nonlinear operators arising in complex engineering systems, making it useful for simulating, designing, and controlling those systems. In this position paper, we provide a comprehensive overview of OL, including its potential applications to complex engineering domains. We cover three variations of OL approaches: deterministic OL for modeling nonautonomous systems, OL with uncertainty quantification (UQ) capabilities, and multifidelity OL. For each variation, we discuss drawbacks and potential applications to engineering, in addition to providing a detailed explanation. We also highlight how multifidelity OL approaches with UQ capabilities can be used to design, optimize, and control engineering systems. Finally, we outline some potential challenges for OL within the engineering domain.

Computer Science↗
DOE OSTI2022

Deep Learning Based Superconducting Radio-Frequency Cavity Fault Classification at Jefferson Laboratory

This work investigates the efficacy of deep learning (DL) for classifying C100 superconducting radio-frequency (SRF) cavity faults in the Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Lab. CEBAF is a large, high-power continuous wave recirculating linac that utilizes 418 SRF cavities to accelerate electrons up to 12 GeV. Recent upgrades to CEBAF include installation of 11 new cryomodules (88 cavities) equipped with a low-level RF system that records RF time-series data from each cavity at the onset of an RF failure. Typically, subject matter experts (SME) analyze this data to determine the fault type and identify the cavity of origin. This information is subsequently utilized to identify failure trends and to implement corrective measures on the offending cavity. Manual inspection of large-scale, time-series data, generated by frequent system failures is tedious and time consuming, and thereby motivates the use of machine learning (ML) to automate the task. This study extends work on a previously developed system based on traditional ML methods (Tennant and Carpenter and Powers and Shabalina Solopova and Vidyaratne and Iftekharuddin, Phys. Rev. Accel. Beams, 2020, 23, 114601), and investigates the effectiveness of deep learning approaches. The transition to a DL model is driven by the goal of developing a system with sufficiently fast inference that it could be used to predict a fault event and take actionable information before the onset (on the order of a few hundred milliseconds). Because features are learned, rather than explicitly computed, DL offers a potential advantage over traditional ML. Specifically, two seminal DL architecture types are explored: deep recurrent neural networks (RNN) and deep convolutional neural networks (CNN). We provide a detailed analysis on the performance of individual models using an RF waveform dataset built from past operational runs of CEBAF. In particular, the performance of RNN models incorporating long short-term memory (LSTM) are analyzed along with the CNN performance. Furthermore, comparing these DL models with a state-of-the-art fault ML model shows that DL architectures obtain similar performance for cavity identification, do not perform quite as well for fault classification, but provide an advantage in inference speed.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

High Temperature Stability of Aqueous Foams for Potential Application in Enhanced Geothermal System (EGS)

Enhanced Geothermal Systems (EGS) are typically constructed by injecting high-pressure water into the deep hot dry rocks (HDR) under carefully controlled conditions to create new or re-open existing fractures, which usually utilizes an immense quantity of water. Alternatively, aqueous foam-based fracturing fluids may offer potential advantages over conventional water-based fracturing fluid, including reduced consumption of water and environmental impact. Although foam-based fracturing has shown promising results in oil and gas industries, its applicability to EGS is unknown. Foams are complex mixtures of liquids with gaseous phases that are thermodynamically unstable. Their stability will decrease over time due to liquid drainage, bubble coarsening, and coalescence. Therefore, it is essential to stabilize foam fluids at high temperatures for EGS related applications such as fracking of HDRs.This paper is focused on investigating high-temperature stability of N2-in-water or N2 foams containing different surfactants and stabilizing agents. Foam stability was characterized by the half-life, which is defined as the time a foam decreases to 50% of its original height due to drainage. Surfactants with various concentrations dissolved in deionized water were investigated, including alfa olefin sulfonate (AOS), sodium dodecyl sulfonate (SDS), TergitolTM (NP – 40), and cetyltrimethylammonium chloride (CTAC). Also, the effects of other stabilizing agents, including guar gum, bentonite clay, crosslinker, graphene oxide (GO) dispersion on foam stability were tested. Initial results showed that foam stability decreased dramatically as temperature increased. On the other hand, foams became relatively more stable as pressure increased. Certain stabilizing agents, such as guar gum and crosslinker, can enhance foam stability at a wide range of temperatures. Our results indicate that it may be possible to obtain stable foams at high-temperature, high-pressure conditions with appropriate stabilizing agents.

Thakore, Virensinh↗
DOE OSTI2025

Quantum Transfer Learning to Boost Dementia Detection

Dementia is a devastating condition with profound implications for individuals, families, and healthcare systems. Early and accurate detection of dementia is critical for timely intervention and improved patient outcomes. While classical machine learning and deep learning approaches have been explored extensively for dementia prediction, these solutions often struggle with high-dimensional biomedical data and large-scale datasets, quickly reaching computational and performance limitations. To address this challenge, quantum machine learning (QML) has emerged as a promising paradigm, offering faster training and advanced pattern recognition capabilities. This work aims to demonstrate the potential of quantum transfer learning (QTL) to enhance the performance of a weak classical deep learning model applied to a binary classification task for dementia detection. Besides, we show the effect of noise on the QTL-based approach, investigating the reliability and robustness of this method. Using the OASIS 2 dataset, we show how quantum techniques can transform a suboptimal classical model into a more effective solution for biomedical image classification, highlighting their potential impact on advancing healthcare technology.

Bhowmik, Sounak [University of Tennessee, Knoxvill↗
DOE OSTI2025

Toward Complete Merger Identification at Cosmic Noon with Deep Learning

As we enter the era of large imaging surveys such as $\textit{Roman}$, Rubin, and $\textit{Euclid}$, a deeper understanding of potential biases and selection effects in optical astronomical catalogs created with the use of ML-based methods is paramount. This work focuses on a deeper understanding of the performance and limitations of deep learning-based classifiers as tools for galaxy merger identification. We train a ResNet18 model on mock Hubble Space Telescope CANDELS images from the IllustrisTNG50 simulation. Our focus is on a more challenging classification of galaxy mergers and nonmergers at higher redshifts $1

Schechter, Aimee [Colorado U.] (ORCID:000000017120↗
DOE OSTI2024

Simultaneous Two- and Three-Photon Deep Imaging of Autofluorescence in Bacterial Communities

The intrinsic fluorescence of bacterial samples has a proven potential for label-free bacterial characterization, monitoring bacterial metabolic functions, and as a mechanism for tracking the transport of relevant components through vesicles. The reduced scattering and axial confinement of the excitation offered by multiphoton imaging can be used to overcome some of the limitations of single-photon excitation (e.g., scattering and out-of-plane photobleaching) to the imaging of bacterial communities. In this work, we demonstrate in vivo multi-photon microscopy imaging of Streptomyces bacterial communities, based on the excitation of blue endogenous fluorophores, using an ultrafast Yb-fiber laser amplifier. Its parameters, such as the pulse energy, duration, wavelength, and repetition rate, enable in vivo multicolor imaging with a single source through the simultaneous two- and three-photon excitation of different fluorophores. Three-photon excitation at 1040 nm allows fluorophores with blue and green emission spectra to be addressed (and their corresponding ultraviolet and blue single-photon excitation wavelengths, respectively), and two-photon excitation at the same wavelength allows fluorophores with yellow, orange, or red emission spectra to be addressed (and their corresponding green, yellow, and orange single-photon excitation wavelengths). We demonstrate that three-photon excitation allows imaging over a depth range of more than 6 effective attenuation lengths to take place, corresponding to an 800 micrometer depth of imaging, in samples with a high density of fluorescent structures.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

A Unified Mechanism on the Formation of Acenes, Helicenes, and Phenacenes in the Gas Phase

Abstract A unified low‐temperature reaction mechanism on the formation of acenes, phenacenes, and helicenes—polycyclic aromatic hydrocarbons (PAHs) that are distinct via the linear, zigzag, and ortho ‐condensed arrangements of fused benzene rings—is revealed. This mechanism is mediated through a barrierless, vinylacetylene mediated gas‐phase chemistry utilizing tetracene, [4]phenacene, and [4]helicene as benchmarks contesting established ideas that molecular mass growth processes to PAHs transpire at elevated temperatures. This mechanism opens up an isomer‐selective route to aromatic structures involving submerged reaction barriers, resonantly stabilized free‐radical intermediates, and systematic ring annulation potentially yielding molecular wires along with racemic mixtures of helicenes in deep space. Connecting helicene templates to the Origins of Life ultimately changes our hypothesis on interstellar carbon chemistry.

Zhao, Long↗
DOE OSTI2019

A Unified Mechanism on the Formation of Acenes, Helicenes, and Phenacenes in the Gas Phase

A unified low-temperature reaction mechanism on the formation of acenes, phenacenes, and helicenes—polycyclic aromatic hydrocarbons (PAHs) that are distinct via the linear, zigzag, and ortho-condensed arrangements of fused benzene rings—is illustrated. This mechanism is mediated through a barrierless, vinylacetylene mediated gas-phase chemistry utilizing tetracene, [4]phenacene, and [4]helicene as benchmarks contesting established ideas that molecular mass growth processes to PAHs transpire at elevated temperatures. This mechanism opens up an isomer-selective route to aromatic structures involving submerged reaction barriers, resonantly stabilized free-radical intermediates, and systematic ring annulation potentially yielding molecular wires along with racemic mixtures of helicenes in deep space. Thus, connecting helicene templates to the Origins of Life ultimately changes our hypothesis on interstellar carbon chemistry.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2026

Insights from a coupled thermo-hydro-mechanical analysis of a layered high-temperature thermal energy storage reservoir

Coupled thermal-hydraulic-mechanical (THM) modeling is applied to investigate the performance of a seasonal high-temperature aquifer thermal energy storage operation based on data and conditions from current site investigations at the Geostorage Forsthaus pilot project in Bern (Switzerland). The model includes subhorizontal sand lenses of various lengths and dips that are embedded in a low permeability clay matrix. Thermal energy storage is simulated by seasonal injection and withdrawal of hot (up to 90 °C) water from a main well, with reservoir pressure regulated by two auxiliary wells at a distance of about 70 m from the main well. The results show how targeted injection into deeper permeable storage formations, along with active deep well pressure control, can effectively minimize geomechanical impact and the potential risk of damaging subsurface storage and sealing formations, or even surface facilities. With such pressure control, the subsurface mechanical responses are dominated by thermal strain and stress, which can be monitored with subsurface fiber optics. The study demonstrates how coupled THM modeling can be applied for the design of a safe and efficient thermal energy storage operation, and how subsurface fiber optic monitoring can be applied for performance confirmation, allowing for more confident operational forecasting.

Rutqvist, Jonny↗
DOE OSTI2024

Inhibition of MALT1 and BCL2 Induces Synergistic Antitumor Activity in Models of B-Cell Lymphoma

The activated B cell (ABC) subset of diffuse large B-cell lymphoma (DLBCL) is characterized by chronic B-cell receptor signaling and associated with poor outcomes when treated with standard therapy. In ABC-DLBCL, MALT1 is a core enzyme that is constitutively activated by stimulation of the B-cell receptor or gain-of-function mutations in upstream components of the signaling pathway, making it an attractive therapeutic target. We discovered a novel small-molecule inhibitor, ABBV-MALT1, that potently shuts down B-cell signaling selectively in ABC-DLBCL preclinical models leading to potent cell growth and xenograft inhibition. We also identified a rational combination partner for ABBV-MALT1 in the BCL2 inhibitor, venetoclax, which when combined significantly synergizes to elicit deep and durable responses in preclinical models. This work highlights the potential of ABBV-MALT1 monotherapy and combination with venetoclax as effective treatment options for patients with ABC-DLBCL.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

The Large Enriched Germanium Experiment for Neutrinoless ββ Decay (LEGEND-1000 Preconceptual Design Report)

We propose the construction of LEGEND-1000, the ton-scale Large Enriched Germanium Experiment for Neutrinoless ββ Decay. This experiment is an international project designed to answer one of the highest priority questions in fundamental physics. It consists of 1000 kg of Ge detectors enriched to more than 90% in the 76 Ge isotope of interest operated in a liquid argon active shield at a deep underground lab. The experiment is designed to achieve a discovery potential that covers the inverted-ordering neutrino mass scale region. The baseline design assumes installation in the SNOLAB cryopit. A similar experimental setup could also be realized at the LNGS alternative site.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2020

Flexible Reinforcement Learning Framework for Building Control using EnergyPlus-Modelica Energy Models

In recent years, reinforcement learning (RL) methods have been greatly enhanced by leveraging deep learning approaches. RL methods applied to building control have shown potential in many applications due to their ability to complement or replace conventional methods such as model-based or rule-based controls. However, RL-based building control software is likely tailored either to one target building system or to a specific RL method so that significant additional effort would be required to customize the RL-based controller for use in other building systems or with other RL approaches. Also, RL-based building controls usually depend on building energy simulations to train controllers, so emulating building dynamics (i.e., thermal dynamics and control dynamics) and capturing sub-hourly dynamic profiles are crucial to further the development of effective RL-based building control methods. To address these challenges, we present an open source RL-based control software employing a high-fidelity hybrid EnergyPlus-Modelica building energy model which emulates building dynamics at 1-minute resolution. This software consists of decoupled components (environment, building emulator, control agent, and RL algorithm), which allows for quick prototyping and benchmarking of standard RL algorithms in different systems; for example, a single component can be replaced without revising all of the software. To demonstrate this software framework, we conducted a benchmark study using an EnergyPlus-Modelica building energy model for a Chicago office building with an RL-based controller to dynamically control the chilled water temperature setpoint and the air handling unit supply air temperature setpoint on selected floors.

Lee, Joon-Yong↗
DOE OSTI2020

DEMONSTRATION OF A DATA-DRIVEN PHYSICS-BASED APPROACH FOR COMPUTATIONALLY EFFICIENT CFD PREDICTION OF TWO-PHASE BUBBLY FLOW

To realize efficient computational fluid dynamics (CFD) prediction of two-phase flow, a multi-scale physics-guided data-driven approach, Feature Similarity Measurement (FSM) technique was developed for error estimation in two-phase flow simulation using coarse-mesh CFD, to achieve a comparable accuracy as fine-mesh simulations with fast-running feature. By defining physics-guided parameters and variable gradients as physical features, FSM has the capability to capture the underlying local patterns in the coarse-mesh CFD simulation. Massive low-fidelity data and respective high-fidelity data are used to explore the underlying information relevant to the main simulation errors and the effects of phenomenological scaling. By learning from previous simulation data, a surrogate model using deep feedforward neural network (DFNN) can be developed and trained to estimate the simulation error of coarse-mesh CFD. In a demonstration case of two-phase bubbly flow, the DFNN model well captured and corrected the unphysical “peaks” in the velocity and void fraction profiles near the wall in the coarse-mesh configuration, even for extrapolative predictions. The research documented supports the feasibility of the physics-guided deep learning methods for coarse mesh CFD simulations which has a potential for the efficient industrial design.

42 ENGINEERING↗
DOE OSTI2020

Ecological Assembly Processes Are Coordinated between Bacterial and Viral Communities in Fractured Shale Ecosystems

The ecological drivers that concurrently act upon both a virus and its host and that drive community assembly are poorly understood despite known interactions between viral populations and their microbial hosts. Hydraulically fractured shale environments provide access to a closed ecosystem in the deep subsurface where constrained microbial and viral community assembly processes can be examined. Here, we used metagenomic analyses of time-resolved-produced fluid samples from two wells in the Appalachian Basin to track viral and host dynamics and to investigate community assembly processes. Hypersaline conditions within these ecosystems should drive microbial community structure to a similar configuration through time in response to common osmotic stress. However, viral predation appears to counterbalance this potentially strong homogeneous selection and pushes the microbial community toward undominated assembly. In comparison, while the viral community was also influenced by substantial undominated processes, it assembled, in part, due to homogeneous selection. When the overall assembly processes acting upon both these communities were directly compared with each other, a significant relationship was revealed, suggesting an association between microbial and viral community development despite differing selective pressures. These results reveal a potentially important balance of ecological dynamics that must be in maintained within this deep subsurface ecosystem in order for the microbial community to persist over extended time periods. More broadly, this relationship begins to provide knowledge underlying metacommunity development across trophic levels.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Learning genetic perturbation effects with variational causal inference

Advances in sequencing technologies have enhanced the understanding of gene regulation in cells. In particular, Perturb-seq has enabled high-resolution profiling of the transcriptomic response to genetic perturbations at the single-cell level. This understanding has implications in functional genomics and potentially for identifying therapeutic targets. Various computational models have been developed to predict perturbational effects. While deep learning models excel at interpolating observed perturbational data, they tend to overfit in the lack of enough data and may not generalize well to unseen perturbations. In contrast, mechanistic models, such as linear causal models based on gene regulatory networks, hold greater potential for extrapolation, as they encapsulate regulatory information that can predict responses to unseen perturbations. However, their application has been limited to small studies due to overly simplistic assumptions, making them less effective in handling noisy, large-scale single-cell data. We propose a hybrid approach that combines a mechanistic causal model with variational deep learning, termed Single Cell Causal Variational Autoencoder (SCCVAE). The mechanistic model employs a learned regulatory network to represent perturbational changes as shift interventions that propagate through the learned network. SCCVAE integrates this mechanistic causal model into a variational autoencoder, generating rich, comprehensive transcriptomic responses. Our results indicate that SCCVAE exhibits superior performance over current state-of-the-art baselines for extrapolating to predict unseen perturbational responses. Additionally, for the observed perturbations, the latent space learned by SCCVAE allows for the identification of functional perturbation modules and simulation of single-gene knockdown experiments of varying penetrance, presenting a robust tool for interpreting and interpolating perturbational responses at the single-cell level.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

Identification and widespread environmental distribution of a gene cassette implicated in anaerobic dichloromethane degradation

Anthropogenic activities and natural processes release dichloromethane (DCM, methylene chloride), a toxic chemical with substantial ozone-depleting capacity. Specialized anaerobic bacteria metabolize DCM; however, the genetic basis for this process has remained elusive. Comparative genomics of the three known anaerobic DCM-degrading bacterial species revealed a homologous gene cluster, designated the methylene chloride catabolism (mec) gene cassette, comprising 8–10 genes encoding proteins with 79.6%–99.7% amino acid identities. Functional annotation identified genes encoding a corrinoid-dependent methyltransferase system, and shotgun proteomics applied to two DCM-catabolizing cultures revealed high expression of proteins encoded on the mec gene cluster during anaerobic growth with DCM. In a DCM-contaminated groundwater plume, the abundance of mec genes strongly correlated with DCM concentrations (R 2 = 0.71–0.85) indicating their potential value as process-specific bioremediation biomarkers. mec gene clusters were identified in metagenomes representing peat bogs, the deep subsurface, and marine ecosystems including oxygen minimum zones (OMZs), suggesting a capacity for DCM degradation in diverse habitats. Here, the broad distribution of anaerobic DCM catabolic potential infers a role for DCM as an energy source in various environmental systems, and implies that the global DCM flux (i.e., the rate of formation minus the rate of consumption) might be greater than emission measurements suggest.

59 BASIC BIOLOGICAL SCIENCES↗