Search NASA⌕ Search

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.

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

At least 667 records · Page 37

Model fusion with physics-guided machine learning: Projection-based reduced-order modeling

The unprecedented amount of data generated from experiments, field observations, and large-scale numerical simulations at a wide range of spatiotemporal scales has enabled the rapid advancement of data-driven and especially deep learning models in the field of fluid mechanics. Although these methods are proven successful for many applications, there is a grand challenge of improving their generalizability. This is particularly essential when data-driven models are employed within outer-loop applications like optimization. In this work, we put forth a physics-guided machine learning (PGML) framework that leverages the interpretable physics-based model with a deep learning model. Leveraging a concatenated neural network design from multi-modal data sources, the PGML framework is capable of enhancing the generalizability of data-driven models and effectively protects against or inform about the inaccurate predictions resulting from extrapolation. We apply the PGML framework as a novel model fusion approach combining the physics-based Galerkin projection model and long- to short-term memory (LSTM) network for parametric model order reduction of fluid flows. We demonstrate the improved generalizability of the PGML framework against a purely data-driven approach through the injection of physics features into intermediate LSTM layers. Our quantitative analysis shows that the overall model uncertainty can be reduced through the PGML approach, especially for test data coming from a distribution different than the training data. Moreover, we demonstrate that our approach can be used as an inverse diagnostic tool providing a confidence score associated with models and observations. The proposed framework also allows for multi-fidelity computing by making use of low-fidelity models in the online deployment of quantified data-driven models.

42 ENGINEERING↗

Deep learning of interface structures from simulated 4D STEM data: cation intermixing vs. roughening ∗

Abstract Interface structures in complex oxides remain an active area of condensed matter physics research, largely enabled by recent advances in scanning transmission electron microscopy (STEM). Yet the nature of the STEM contrast in which the structure is projected along the given direction precludes separation of possible structural models. Here, we utilize deep convolutional neural networks (DCNN) trained on simulated 4D STEM datasets to predict structural descriptors of interfaces. We focus on the widely studied interface between LaAlO 3 and SrTiO 3 , using dynamical diffraction theory and leveraging high performance computing to simulate thousands of possible 4D STEM datasets to train the DCNN to learn properties of the underlying structures on which the simulations are based. We test the DCNN on simulated data and show that it is possible (with >95% accuracy) to identify a physically rough from a chemically diffuse interface and create a DCNN regression model to predict step positions. We quantify the applicability of the model to different thicknesses and the transferability of the approach. The method shown here is general and can be applied for any inverse imaging problem where forward models are present.

42 ENGINEERING↗

Revealing Fundamental Physics from the Daya Bay Neutrino Experiment Using Deep Neural Networks

Experiments in particle physics produce enormous quantities of data that must be analyzed and interpreted by teams of physicists. This analysis is often exploratory, where scientists are unable to enumerate the possible types of signal prior to performing the experiment. Thus, tools for summarizing, clustering, visualizing and classifying high-dimensional data are essential. Here in this work, we show that meaningful physical content can be revealed by transforming the raw data into a learned high-level representation using deep neural networks, with measurements taken at the Daya Bay Neutrino Experiment as a case study. We further show how convolutional deep neural networks can provide an effective classification filter with greater than 97% accuracy across different classes of physics events, significantly better than other machine learning approaches.

60 APPLIED LIFE SCIENCES↗

Efficient Data Compression for 3D Sparse TPC via Bicephalous Convolutional Autoencoder

Real-time data collection and analysis in large experimental facilities present a great challenge across multiple domains, including high energy physics, nuclear physics, and cosmology. To address this, machine learning (ML)-based methods for real-time data compression have drawn significant attention. However, unlike natural image data, such as CIFAR and ImageNet that are relatively small-sized and continuous, scientific data often come in as three-dimensional 3D data volumes at high rates with high sparsity (many zeros) and non-Gaussian value distribution. This makes direct application of popular ML compression methods, as well as conventional data compression methods, suboptimal. To address these obstacles, this work introduces a dual-head autoencoder to resolve sparsity and regression simultaneously, called Bicephalous Convolutional AutoEncoder (BCAE). This method shows advantages both in compression fidelity and ratio compared to traditional data compression methods, such as MGARD, SZ, and ZFP. To achieve similar fidelity, the best performer among the traditional methods can reach only half the compression ratio of BCAE. Moreover, a thorough ablation study of the BCAE method shows that a dedicated segmentation decoder improves the reconstruction.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Deep Learning for Unsupervised Insider Threat Detection in Structured Cybersecurity Data Streams

Situational awareness data is a prototypical example of structured heterogeneous streaming data with asynchronous timing that can quickly scale beyond the cognitive power of a human analyst. We present deep learning methods that can train in real time from unlabeled streams and provide probabilistic anomaly scores. We compare our novel deep and recurrent neural network models to three baseline systems using threat detection recall on the CERT Insider Threat Dataset v6.2. In our experiments, our deep neural network and our recurrent neural network systems outperformed all baselines, and we found that using aggregated activity count features yielded the best performance. Our best model produces on average an anomaly score of 95.53 percentile for the day, suggesting it can facilitate a great reduction in analyst workload. Our models are also interpretable, in that they break down the anomaly score into individual features of user behavior, which could further aid analysts in reviewing potential cases of insider threat.

Tuor, Aaron R.↗

Extracting sea quark spin distributions from a global Monte Carlo analysis of semi-inclusive DIS data

Polarized semi-inclusive deep inelastic scattering (SIDIS) plays a crucial role in understanding sea quark contributions to the proton spin through global analyses of spin-dependent parton distribution functions (PDFs). The shape of the strange quark polarization in particular has been shown to be dramatically different between analyses that include or exclude SIDIS data. The inclusion of SIDIS data, however, requires knowledge of fragmentation functions (FFs), which up until now have been fixed in all current analyses of SIDIS data to specific parameterizations determined from single-inclusive annihilation (SIA) and other quark to hadron fragmentation processes. In this talk, I will present results from a first ever simultaneous Monte Carlo analysis of polarized PDFs and FFs using DIS, SIA, and SIDIS data. I will also discuss in detail the extent to which current pion and kaon SIDIS data can distinguish between the valence and sea spin distributions of the proton.

Ethier, Jacob↗

A Statistician’s Overview of Physics-Informed Neural Networks for Spatio-Temporal Data

The recent success of deep neural network models with physical constraints (so-called, Physics-Informed Neural Networks, PINNs) has led to renewed interest in the incorporation of mechanistic information in predictive models. Statisticians and others have long been interested in this problem, which has led to several practical and innovative solutions dating back decades. In this overview, we focus on the problem of data-driven prediction and inference of dynamic spatio-temporal processes that include mechanistic information, such as would be available from partial differential equations, with a strong focus on the quantification of uncertainty associated with data, process, and parameters. Here, we give a brief review of several paradigms and focus our attention on Bayesian implementations given they naturally accommodate uncertainty quantification. We then show that it is straight-forward to include the Bayesian PINN (B-PINN) within the Bayesian hierarchical model (BHM) framework that has long been considered for modeling dynamic spatio-temporal processes. Such a BHM-PINN is illustrated via a simulation study in which a latent nonlinear Burgers’ equation PDE governs the dynamics of Poisson distributed spatio-temporal data. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.

Bayesian↗

Post-hazard condition assessment of nuclear piping-equipment systems: Novel approach to feature extraction and deep learning

Over the past decade, the use of artificial intelligence techniques in the field of health-monitoring has gained significant interest, especially for structures such as building and bridges. However, applications to industrial systems such as equipment-piping systems in nuclear plants have not been explored. Here, in this paper, it is shown that the existing techniques developed for buildings and bridges cannot be extended directly to equipment-piping systems as the response of such systems is governed by multiple localized modes unlike that in buildings and bridges. This paper proposes a new approach that consists of three key aspects: (i) a novel vector of degradation-sensitive features extracted from measured data, (ii) using a deep Artificial Neural Network (ANN) for diagnosis of degradation location and degradation severity, and (iii) consideration of uncertainty in degradation severity when training the ANN. Degradation in piping-equipment systems can occur due to flow-accelerated erosion and corrosion. These locations can potentially exhibit damage such as localized yielding or initiation of cracking due to an external event such as an earthquake. Moreover, such locations can at times go undetected by current inspection techniques. Therefore, a robust framework is needed for detection of degradation after a seismic event. This manuscript proposes a proof-of-concept framework, which utilizes data collected from sensors to generate a deep ANN database for predicting degraded locations and severity in a piping-equipment system. Degradation severity is classified as minor, moderate, and severe. In the suggested methodology, a novel vector of degradation-sensitive features is extracted from the sensor data to train the ANN. A simple piping-equipment system is selected to demonstrate feature extraction as a means to simplify pattern recognition, explore the design and parameters of an ANN, and develop a sensor placement strategy. The effectiveness of the proposed framework is demonstrated on a realistic primary safety system of a two-loop nuclear reactor. It is shown that the proposed post-hazard condition assessment framework is able to detect degraded locations along with the severity levels, including minor degradation, with considerably higher accuracy.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Unified Wind-Wave Growth and Spectrum Functions for All Water Depths: Field Observations and Model Results

Abstract Wind-wave development is governed by the fetch- or duration-limited growth principle that is expressed as a pair of similarity functions relating the dimensionless elevation variance (wave energy) and spectral peak frequency to fetch or duration. Combining the pair of similarity functions, the fetch or duration variable can be removed to form a dimensionless function of elevation variance and spectral peak frequency, which is interpreted as the wave energy evolution with wave age. The relationship is initially developed for quasi-neural stability and quasi-steady wind forcing conditions. Further analyses show that the same fetch, duration, and wave-age similarity functions are applicable to unsteady wind forcing conditions, including rapidly accelerating and decelerating mountain gap wind episodes and tropical cyclone (TC) wind fields. Here it is shown that with the dimensionless frequency converted to dimensionless wavenumber using the surface wave dispersion relationship, the same similarity function is applicable in all water depths. Field data collected in shallow to deep waters and mild to TC wind conditions and synthetic data generated by spectrum model computations are assembled to illustrate the applicability. For the simulation work, the finite-depth wind-wave spectrum model and its shoaling function are formulated for variable spectral slopes. Given wind speed, wave age, and water depth, the measured and spectrum-computed significant wave heights and the associated growth parameters are in good agreement in forcing conditions from mild to TC winds and in all depths from deep ocean to shallow lake. Significance Statement This paper presents a growth function and spectrum model to describe wind-wave development in all water depths. Their applicability covers a wide range of wind forcing conditions including steady, accelerating, decelerating, and tropical cyclone events. Support for the unified spectrum model and growth function is presented with field observations and numerical computations.

Hwang, Paul A.↗

DEEP NEAR-INFRARED SURVEY OF THE PIPE NEBULA. II. DATA, METHODS, AND DUST EXTINCTION MAPS

We present a new set of high-resolution dust extinction maps of the nearby and essentially starless Pipe Nebula molecular cloud. The maps were constructed from a concerted deep near-infrared imaging survey with the ESO-VLT, ESO-NTT, CAHA 3.5 m telescopes, and 2MASS data. The new maps have a resolution three times higher than the previous extinction map of this cloud by Lombardi et al. and are able to resolve structures down to 2600 AU. We detect 244 significant extinction peaks across the cloud. These peaks have masses between 0.1 and 18.4 M{sub sun}, diameters between 1.2 and 5.7 x 10{sup 4} AU (0.06 and 0.28 pc), and mean densities of about 10{sup 4} cm{sup -3}, all in good agreement with previous results. From the analysis of the mean surface density of companions we find a well-defined scale near 1.4 x 10{sup 4} AU below which we detect a significant decrease in structure of the cloud. This scale is smaller than the Jeans length calculated from the mean density of the peaks. The surface density of peaks is not uniform but instead it displays clustering. Extinction peaks in the Pipe Nebula appear to have a spatial distribution similar to the stars in Taurus, suggesting that the spatial distribution of stars evolves directly from the primordial spatial distribution of high-density material.

79 ASTRONOMY AND ASTROPHYSICS↗

Transfer-Learnt Energy Models for Predicting Electricity Consumption in Buildings with Limited and Sparse Field Data

Modeling energy consumption is critical for energy-efficient utilization of the electric appliances in a building, smart grid programs (like demand-response), and many other smart home applications. State-of-the-art energy modeling techniques either rely on theoretical models, or extensive instrumentation of the building envelope to gather ``big" data to train a deep neural network. While theoretical models are often limited by their estimation accuracy, it is not always feasible to gather a significant amount of field data. In this paper, we explore transfer learning-based strategies to train much more accurate model for energy estimation when using a sparse field data. We transferred knowledge, in the form of data and parameters, from the simulation framework to the field data. We evaluated the efficacy of our approach on field data collected from six commercial buildings and our results indicate that transfer learning-based models trained over one month data can perform comparative (and in some cases better) than the state-of-the-art machine learning and deep learning solutions.

Jain, Milan↗

Combining Multicomponent Seismic Attributes, New Rock Physics Models, and In Situ Data to Estimate Gas-Hydrate Concentrations in Deep-Water, Near-Seafloor Strata of the Gulf of Mexico

The Bureau of Economic Geology was contracted to develop technologies that demonstrate the value of multicomponent seismic technology for evaluating deep-water hydrates across the Green Canyon area of the Gulf of Mexico. This report describes the methodologies that were developed to create compressional (P-P) and converted-shear (P-SV) images of near-seafloor geology from four-component ocean-bottom-cable (4C OBC) seismic data and the procedures used to integrate P-P and P-SV seismic attributes with borehole calibration data to estimate hydrate concentration across two study areas spanning 16 and 25 lease blocks (or 144 and 225 square miles), respectively. Approximately 200 km of two-dimensional 4C OBC profiles were processed and analyzed over the course of the 3-year project. The strategies we developed to image near-seafloor geology with 4C OBC data are unique, and the paper describing our methodology was peer-recognized with a Best Paper Award by the Society of Exploration Geophysicists in the first year of the project (2006). Among the valuable research findings demonstrated in this report, the demonstrated ability to image deep-water near-seafloor geology with sub-meter resolution using a standard-frequency (10-200 Hz) air gun array on the sea surface and 4C sensors on the seafloor has been the accomplishment that has received the most accolades from professional peers. Our study found that hydrate is pervasive across the two study areas that were analyzed but exists at low concentrations. Although our joint inversion technique showed that in some limited areas, and in some geologic units across those small areas, hydrates occupied up to 40-percent of the sediment pore space, we found that when hydrate was present, hydrate concentration tended to occupy only 10-percent to 20-percent of the pore volume. We also found that hydrate concentration tended to be greater near the base of the hydrate stability zone than it was within the central part of the stability zone.

03 NATURAL GAS↗

Probing Large-scale Coherence between Spitzer IR and Chandra X-Ray Source-subtracted Cosmic Backgrounds

We present new measurements of the large-scale clustering component of the cross-power spectra of the source-subtracted Spitzer -IRAC cosmic infrared background and Chandra -ACIS cosmic X-ray background surface brightness fluctuations Our investigation uses data from the Chandra Deep Field South, Hubble Deep Field North, Extended Groth Strip/AEGIS field, and UDS/SXDF surveys, comprising 1160 Spitzer hours and ∼12 Ms of Chandra data collected over a total area of 0.3 deg{sup 2}. We report the first (>5 σ ) detection of a cross-power signal on large angular scales >20″ between [0.5–2] keV and the 3.6 and 4.5 μ m bands, at ∼5 σ and 6.3 σ significance, respectively. The correlation with harder X-ray bands is marginally significant. Comparing the new observations with existing models for the contribution of the known unmasked source population at z < 7, we find an excess of about an order of magnitude at 5 σ confidence. We discuss possible interpretations for the origin of this excess in terms of the contribution from accreting early black holes (BHs), including both direct collapse BHs and primordial BHs, as well as from scattering in the interstellar medium and intra-halo light.

79 ASTRONOMY AND ASTROPHYSICS↗

Deep learning tackles cloud detection

To better understand Earth systems, atmospheric scientists use cloud data collected by a remote sensing tool called lidar. These data are an important part of global forecasting, but analyzing lidar images by hand is time-consuming. Researchers at Pacific Northwest National Laboratory teamed up to see if a deep learning model could process lidar data better than current automatic detection techniques.

54 ENVIRONMENTAL SCIENCES↗

Hawthorne Nevada Deep Direct-Use Feasibility Study - Navy Geothermal Program Office General Data

The objective of this project is to use a multi-disciplinary, three-tiered approach to assess the geothermal resource and determine the feasibility of implementing a large-scale, direct-use facility for the Hawthorne Army Depot (HAD) and the various town and county facilities in Hawthorne, Nevada. This assessment directly targets a geothermal resource recently characterized by the Navy Geothermal Program Office (GPO) as part of a focused exploration and development campaign. This data set contains: Excel file of the temperature at depth logs; Zip file containing shapefiles of gravity magnitude and base data for the Hawthorne Geothermal area; Word doc files containing HWAD geothermal report, figures and local stratigraphy; Zip folder containing PDF files of relevant HWAD document; Zip folder containing JPEGs of the John Oldow maps; Zip folder containing MXDs, xls, and dbf files of the Hawthorne Geothermal area and figures; Zip folder containing Hawthorne Geothermal data from Epsilon

15 GEOTHERMAL ENERGY↗

Super-Resolution Approaches in Three-Dimensions for Classification and Screening of Commercial-Off-The-Shelf Components

X-ray computed tomography is generally a primary step in characterization of defective electronic components, but is generally too slow to screen large lots of components. Super-resolution imaging approaches, in which higher-resolution data is inferred from lower-resolution images, have the potential to substantially reduce collection times for data volumes accessible via x-ray computed tomography. Here we seek to advance existing two-dimensional super-resolution approaches directly to three-dimensional computed tomography data. Multiple scan resolutions over a half order of magnitude of resolution were collected for four classes of commercial electronic components to serve as training data for a deep-learning, super-resolution network. A modular python framework for three-dimensional super-resolution of computed tomography data has been developed and trained over multiple classes of electronic components. Initial training and testing demonstrate the vast promise for these approaches, which have the potential for more than an order of magnitude reduction in collection time for electronic component screening.

47 OTHER INSTRUMENTATION↗

Robust Vector BOTDA Signal Processing with Probabilistic Machine Learning

This paper presents a novel probabilistic machine learning (PML) framework to estimate the Brillouin frequency shift (BFS) from both Brillouin gain and phase spectra of a vector Brillouin optical time-domain analysis (VBOTDA). The PML framework is used to predict the Brillouin frequency shift (BFS) along the fiber and to assess its predictive uncertainty. We compare the predictions obtained from the proposed PML model with a conventional curve fitting method and evaluate the BFS uncertainty and data processing time for both methods. The proposed method is demonstrated using two BOTDA systems: (i) a BOTDA system with a 10 km sensing fiber and (ii) a vector BOTDA with a 25 km sensing fiber. The PML framework provides a pathway to enhance the VBOTDA system performance.

47 OTHER INSTRUMENTATION↗

PHOTOMETRIC CONSTRAINTS ON THE REDSHIFT OF z {approx} 10 CANDIDATE UDFj-39546284 FROM DEEPER WFC3/IR+ACS+IRAC OBSERVATIONS OVER THE HUDF

Ultra-deep WFC3/IR observations on the HUDF from the HUDF09 program revealed just one plausible z {approx} 10 candidate, UDFj-39546284. UDFj-39546284 had all the properties expected of a galaxy at z {approx} 10 showing (1) no detection in the deep ACS+WFC3 imaging data blueward of the F160W band, exhibiting (2) a blue spectral slope redward of the break, and showing (3) no prominent detection in deep IRAC observations. The new, similarly deep WFC3/IR HUDF12 F160W observations over the HUDF09/XDF allow us to further assess this candidate. These observations show that this candidate, previously only detected at {approx}5.9{sigma} in a single band, clearly corresponds to a real source. It is detected at {approx}5.3{sigma} in the new H{sub 160}-band data and at {approx}7.8{sigma} in the full 85-orbit H{sub 160}-band stack. Interestingly, the non-detection of the source (<1{sigma}) in the new F140W observations suggests a higher redshift. Formally, the best-fit redshift of the source utilizing all the WFC3+ACS (and IRAC+K{sub s} -band) observations is 11.8 {+-} 0.3. However, we consider the z {approx} 12 interpretation somewhat unlikely, since the source would either need to be {approx}20 Multiplication-Sign more luminous than expected or show very high-EW Ly{alpha} emission (which seems improbable given the extensive neutral gas prevalent early in the reionization epoch). Lower-redshift solutions fail if only continuum models are allowed. Plausible lower-redshift solutions require that the H{sub 160}-band flux be dominated by line emission such as H{alpha} or [O III] with extreme EWs. The tentative detection of line emission at 1.6 {mu}m in UDFj-39546284 in a companion paper suggests that such emission may have already been found.

79 ASTRONOMY AND ASTROPHYSICS↗