Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Streaming 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 1,405 records · Page 78

Online randomized interpolative decomposition with a posteriori error estimator for temporal PDE data reduction

Traditional low-rank approximation is a powerful tool for compressing large data matrices that arise in simulations of partial differential equations (PDEs), but suffers from high computational cost and requires several passes over the PDE data. The compressed data may also lack interpretability thus making it difficult to identify feature patterns from the original data. Here, to address these issues, we present an online randomized algorithm to compute the interpolative decomposition (ID) of large-scale data matrices in situ. Compared to previous randomized IDs that used the QR decomposition to determine the column basis, we adopt a streaming ridge leverage score-based column subset selection algorithm that dynamically selects proper basis columns from the data and thus avoids an extra pass over the data to compute the coefficient matrix of the ID. In particular, we adopt a single-pass error estimator based on the non-adaptive Hutch++ algorithm to provide real-time error approximation for determining the best coefficients. As a result, our approach only needs a single pass over the original data and thus is suitable for large and high-dimensional matrices stored outside of core memory or generated in PDE simulations. A strategy to improve the accuracy of the reconstructed data gradient, when desired, within the ID framework is also presented. We provide numerical experiments on turbulent channel flow and ignition simulations, and on the NSTX Gas Puff Image dataset, comparing our algorithm with the offline ID algorithm to demonstrate its utility in real-world applications.

Column subset selection↗

A machine learning based approach to online electron reconstruction at CLAS12

Online reconstruction is key for monitoring purposes and real time analysis in High Energy and Nuclear Physics experiments. A necessary component of reconstruction algorithms is particle identification that combines information left by a particle passing through several detector components to identify the particle’s type. Of particular interest to electro-production Nuclear Physics experiments such as CLAS12 is electron identification which is used to trigger data recording. A machine learning approach was developed for CLAS12 to reconstruct and identify electrons by combining raw signals at the data acquisition level from several detector components. Here, this approach achieves an electron identification purity above 75% whilst retaining an efficiency close to 100%. The machine learning tools are capable of running at high rates exceeding the data acquisition rates and will allow electron reconstruction in real-time. This work enhances online analyses and monitoring and can contribute to improved triggering at CLAS12. This machine learning driven approach will also be crucial for experiments aiming to transition to streaming readout operations where online reconstruction will be a key component of the data taking paradigm.

Artificial intelligence↗

Impact of gel concentration on filter fluxes in microfiltration of Hanford tank wastes and simulants

Abstract Treatment processes have been proposed that will utilize crossflow filtration to concentrate sludge waste streams at the Department of Energy's Hanford Site. Challenges associated with solid–liquid separation of the waste streams drive a necessary evaluation of available Hanford high level waste (HLW) filtration data. Limiting flux conditions during crossflow filtration are elucidated with the formation of a cake layer on the membrane surface. A mass transfer coefficient between the gel and bulk concentrations plays a critical role in determining filter flux. A correlation between the gel concentration and mass transfer coefficient is made to assist in determining filter performance of select HLW streams. As a process alternative to crossflow filtration, gravity settling of waste streams may be deployed as a solid–liquid separation technique. However, this results in a contrasting performance with the centrifuged solids concentration. A method was developed to estimate expected filtration and settling performance based on physical characterization data for Hanford tank waste samples. By assessing the estimated processing performance of HLW, technical support can be provided during flowsheet planning.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

The First Tidally Disrupted Ultra-faint Dwarf Galaxy?: A Spectroscopic Analysis of the Tucana III Stream

In this paper, we present a spectroscopic study of the tidal tails and core of the Milky Way satellite Tucana III, collectively referred to as the Tucana III stream, using the 2dF+AAOmega spectrograph on the Anglo-Australian Telescope and the IMACS spectrograph on the Magellan Baade Telescope. In addition to recovering the brightest nine previously known member stars in the Tucana III core, we identify 22 members in the tidal tails. We observe strong evidence for a velocity gradient of $8.0\pm 0.4\,\mathrm{km}\,{{\rm{s}}}^{-1}\,{\deg }^{-1}$ over at least 3° on the sky. Based on the continuity in velocity, we confirm that the Tucana III tails are real tidal extensions of Tucana III. The large velocity gradient of the stream implies that Tucana III is likely on a radial orbit. We successfully obtain metallicities for four members in the core and 12 members in the tails. We find that members close to the ends of the stream tend to be more metal-poor than members in the core, indicating a possible metallicity gradient between the center of the progenitor halo and its edge. The spread in metallicity suggests that the progenitor of the Tucana III stream is likely a dwarf galaxy rather than a star cluster. Furthermore, we find that with the precise photometry of the Dark Energy Survey data, there is a discernible color offset between metal-rich disk stars and metal-poor stream members. Lastly, this metallicity-dependent color offers a more efficient method to recognize metal-poor targets and will increase the selection efficiency of stream members for future spectroscopic follow-up programs on stellar streams.

79 ASTRONOMY AND ASTROPHYSICS↗

NEWTS Coal Mine Drainage Dataset from Cravotta Brady (2015)

Data from Cravotta, Brady, "Priority pollutants and associated constituents in untreated and treated discharges from coal mining or processing facilities in Pennsylvania, USA". Applied Geochemistry, 2015. https://doi.org/10.1016/j.apgeochem.2015.03.001 Dataset includes information on water quality composition including inorganic compounds from untreated and treated streams of coal-mine discharge from coal mining and coal processing locations. Data is provided in the original version as well as in a summarized version for easy input into aqueous chemistry modeling software.

Aqueous Chemistry↗

Parametric model-order reduction for radiation transport using multi-resolution proper orthogonal decomposition

For parametric high-fidelity simulations, it is often desirable to utilize a reduced-order model (ROM) to emulate, at a reduced computational cost, parametric solutions of the governing partial differential equations (PDEs) for unseen parameter values. One commonly employed option is to utilize a data-driven, projection-based ROM supplemented with subspace identification via proper orthogonal decomposition (POD). POD discovers the ROM subspace by computing the singular value decomposition (SVD) of a set of training data from the full-order model (FOM). In streaming-dominated radiation transport simulations with localized sources, solutions often greatly vary over the spatial domain by many orders of magnitude. In such cases, machine-precision arithmetic can be insufficient to obtain an accurate SVD, resulting in a poorly performing ROM. We present a method called multiresolution POD (mrPOD) that mitigates these inaccuracies. The mrPOD method works by decomposing the spatial domain into regions and performing proper orthogonal decomposition on the training dataset separately in each region. In conclusion, mrPOD is tested on single energy group and multigroup atmospheric shielding transport problems and is shown to outperform classic POD.

42 ENGINEERING↗

Mapping longitudinal stream connectivity in the North St. Vrain Creek watershed of Colorado

We use reach-scale stream gradient as an indicator of longitudinal connectivity for water, sediment, and organic matter in a mountainous watershed in Colorado. Stream reaches with the highest gradient tend to have narrow valley bottoms with limited storage space and attenuation of downstream fluxes, whereas stream reaches with progressively lower gradients have progressively more storage and greater attenuation. We compared the distribution of stream gradient to stream-reach connectivity rankings that incorporated multiple potential control variables, including lithology, upland vegetation, hydroclimatology, road crossings, and flow diversions. We then assessed connectivity rankings using different weighting schemes against stream gradient and against field-based understanding of relative connectivity within the watershed. Here, we conclude that stream gradient, which is simple to map using publicly available data and digital elevation models, is the most robust indicator of relative longitudinal connectivity within the river network.

58 GEOSCIENCES↗

Blended polybenzimidazole and melamine-co-formaldehyde thermosets

Polybenzimidazole [PBI; poly-2,2’(m-phenylene)-5,5’-bibenzimidazole] is known to have excellent high temperature stability (up to 450 ºC) and superb H 2 /CO 2 selectivity compared to most high performance (HP) polymers. But, PBI has issues in thin-film formation compared to other HP polymers, due to challenging processing techniques. In this work, new blended thermosets were made with PBI and poly(melamine co-formaldehyde) [PMF] to produce stable thin-films after thermal processing at 220-250 ºC. PBI film formation is difficult, and the film tends to fracture and fissure due to loss of processing aids and stabilizers (salt/acid additives) that are found in PBI solutions above10 wt %. The PBI-PMF blended thermosets we report do not have stabilizers, and can be made into dense thin-films. It is remarkable that these thermally processed PBI-PMF films were stable in deionized water for extended periods, whereas many PBI films that contain additives are not. The PBI-PMF films were analyzed using pure and mixed gas permeability measurement techniques. At 250 °C, the data show H 2 /CO 2 gas selectivities greater than 13. Also, from the gas permeation data, the energy of activation (Ep) of a mixed gas stream for PBI-PMF shows that hydrogen permeates more easily than the other gases, while the permeabilities for the larger kinetic diameter gases are greatly diminished. The FT-IR spectra show that the PBI-PMF films have changed from parent PBI after thermal processing, and PMF dominates the spectra even in minor percent compositions. Altogether, the reported PBI-PMF thermoset films show good stability which can be used for high temperature gas separation.

36 MATERIALS SCIENCE↗

Physical insight into switchgrass dissolution in the ionic liquid 1-ethyl-3-methylimidazolium acetate

Small-angle neutron scattering was used to characterize solutions of switchgrass and the constituent biopolymers cellulose, hemicellulose, and lignin, as well as a physical mixture of them mimicking the composition of switchgrass, dissolved in the ionic liquid (IL) 1-ethyl-3-methylimidazolium acetate. The results demonstrate that the IL dissolves the cellulose fibrils of switchgrass, although a supramolecular biopolymer network remains that is not present in solutions of the individual biopolymers and that does not self-assemble in a solution containing the physical mixture of the individual biopolymers. The persistence of a network-like structure indicates that dissolving switchgrass in the IL does not disrupt all of the physical entanglements and covalent linkages between the biopolymers created during plant growth. Reconstitution of the IL-dissolved switchgrass yields carbohydrate-rich material containing cellulose with a low degree of crystallinity, as determined by powder X-ray diffraction, which impacts potential down-stream uses of the biopolymers produced by the process. The data suggests that the use of chemical additives which would break bonds that exist between the lignin and hemicellulose might improve the purity of the resulting product, but may not be able to disrupt the highly physically-entangled biopolymer network sufficiently to facilitate their separation.

Wang, Hui↗

MAGNETIC HELICITY SPECTRUM OF SOLAR WIND FLUCTUATIONS AS A FUNCTION OF THE ANGLE WITH RESPECT TO THE LOCAL MEAN MAGNETIC FIELD

Magnetic field data acquired by the Ulysses spacecraft in high-speed streams over the poles of the Sun are used to investigate the normalized magnetic helicity spectrum {sigma}{sub m} as a function of the angle {theta} between the local mean magnetic field and the flow direction of the solar wind. This spectrum provides important information about the constituent modes at the transition to kinetic scales that occurs near the spectral break separating the inertial range from the dissipation range. The energetically dominant signal at scales near the thermal proton gyroradius k{sub perpendicular{rho}i} {approx} 1 often covers a wide band of propagation angles centered about the perpendicular direction, {theta} {approx_equal} 90{sup 0} {+-} 30{sup 0}. This signal is consistent with a spectrum of obliquely propagating kinetic Alfven waves with k{sub perpendicular} >> k{sub ||} in which there is more energy in waves propagating away from the Sun and along the direction of the local mean magnetic field than toward the Sun. Moreover, this signal is principally responsible for the reduced magnetic helicity spectrum measured using Fourier transform techniques. The observations also reveal a subdominant population of nearly parallel propagating electromagnetic waves near the proton inertial scale k{sub ||} c/{omega}{sub pi} {approx} 1 that often exhibit high magnetic helicity |{sigma}{sub m}| {approx_equal} 1. These waves are believed to be caused by proton pressure anisotropy instabilities that regulate distribution functions in the collisionless solar wind. Because of the existence of a drift of alpha particles with respect to the protons, the proton temperature anisotropy instability that operates when T{sub pperpendicular}/T{sub p||} > 1 preferentially generates outward propagating ion-cyclotron waves and the fire-hose instability that operates when T{sub pperpendicular}/T{sub p||} < 1 preferentially generates inward propagating whistler waves. These kinetic processes provide a natural explanation for the magnetic field observations.

79 ASTRONOMY AND ASTROPHYSICS↗

DEEP OPTICAL PHOTOMETRY OF SIX FIELDS IN THE ANDROMEDA GALAXY

Using the Advanced Camera for Surveys on the Hubble Space Telescope, we have obtained deep optical images reaching well below the oldest main-sequence turnoff in six fields of the Andromeda Galaxy. The fields fall at four positions on the southeast minor axis, one position in the giant stellar stream, and one position on the northeast major axis. These data were obtained as part of three large observing programs designed to probe the star formation history of the stellar population in various structures of the galaxy. In this paper, we present the images, catalogs, and artificial star tests for these observing programs as a supplement to the analyses published previously. These high-level science products are also archived at the Multimission Archive at the Space Telescope Science Institute.

79 ASTRONOMY AND ASTROPHYSICS↗

Robustness of cosmic neutrino background detection in the cosmic microwave background

The existence of a cosmic neutrino background can be probed indirectly by CMB experiments, not only by measuring the background density of radiation in the universe, but also by searching for the typical signatures of the fluctuations of free-streaming species in the temperature and polarisation power spectrum. Previous studies have already proposed a rather generic parametrisation of these fluctuations, that could help to discriminate between the signature of ordinary free-streaming neutrinos, or of more exotic dark radiation models. Current data are compatible with standard values of these parameters, which seems to bring further evidence for the existence of a cosmic neutrino background. In this work, we investigate the robustness of this conclusion under various assumptions. We generalise the definition of an effective sound speed and viscosity speed to the case of massive neutrinos or other dark radiation components experiencing a non-relativistic transition. We show that current bounds on these effective parameters do not vary significantly when considering an arbitrary value of the particle mass, or extended cosmological models with a free effective neutrino number, dynamical dark energy or a running of the primordial spectrum tilt. We conclude that it is possible to make a robust statement about the detection of the cosmic neutrino background by CMB experiments.

79 ASTRONOMY AND ASTROPHYSICS↗

Machine Optimized Analysis Tools

This software package provides a common "front end" for various math operations that are relevant to astrophysical data processing. This includes special function evaluation, fast Fourier transforms, streamed random number generation and spherical geometry projections. Computer-vendor specific libraries and internal algorithms can be selected on different systems in order to achieve the best possible performance on all systems. This software is still in an early state of development.

Kisner, Theodore↗

TOMOCUPY

ANL REFERENCE SF-22-102 DESCRIPTION: Tomocupy is a Python package and a command-line interface for GPU reconstruction of tomographic/laminographic data in 16-bit and 32-bit precision. It implements an efficient data processing conveyor allowing to overlap all data transfers with computations. First, independent Python threads are started for reading data chunks from the hard disk into a Python data queue and for writing reconstructed chunks from the Python queue to the hard disk. Second, CPU-GPU data transfers are overlapped with GPU computations by using CUDA streams.

NIKITIN, VIKTOR↗

NOODLES Grid [SWR-25-93]

NOODLES Grid is a real-time visualization server for power system simulation data. It loads precomputed datasets, builds optimized instance renderings, and streams live interactive scenes to connected clients. It is built for high scalability, flexible visualization, and fast interaction. NOODLES is a cross-platform/device/tool protocol for collaborative visualization. NOODLES was Developed at the National Renewable Energy Laboratory (NREL) as a capability of the Insight Center https://www.nrel.gov/computational-science/insight-center.html

Brunhart-Lupo, Nicholas [National Renewable Energy↗

Method and apparatus of parallel computing with simultaneously operating stream prefetching and list prefetching engines

A prefetch system improves a performance of a parallel computing system. The parallel computing system includes a plurality of computing nodes. A computing node includes at least one processor and at least one memory device. The prefetch system includes at least one stream prefetch engine and at least one list prefetch engine. The prefetch system operates those engines simultaneously. After the at least one processor issues a command, the prefetch system passes the command to a stream prefetch engine and a list prefetch engine. The prefetch system operates the stream prefetch engine and the list prefetch engine to prefetch data to be needed in subsequent clock cycles in the processor in response to the passed command.

97 MATHEMATICS AND COMPUTING↗

High-Performance data flows using analytical models and measurements

The combination of analytical models and measurements provide practical configurations and parameters to achieve high data transport rates: (a) buffer sizes and number of parallel streams for improved memory and file transfer rates, (b) Hamilton and Scalable TCP congestion control modules for memory transfers in place of default CUBIC, and (c) direct IO mode for Lustre file systems for wide-area transfers. Conventional parameter selection using full sweeps is impractical in many cases since it takes months. By exploiting the unimodality of throughput profiles, we developed the d-w method that significantly reduces the number of measurements needed for parameter identification. This heuristic method was effective in practice in reducing the measurements by about 90% for Lustre and XFS file transfers.

Rao, Nageswara S↗

Models of TCP in high-BDP environments and their experimental validation

In recent years, the computer networking community has seen a steady growth in bandwidth-delay products (BDPs). Several TCP variants were created to combat the shortcomings of legacy TCP when it comes to operation in high-BDP environments. These variants, among which are CUBIC, STCP, and H-TCP, have been extensively studied in some empirical contexts, and some analytical models exist for CUBIC and STCP. However, since these studies have been conducted, BDPs have risen even more, and new bulk data transfer tools have emerged that utilize multiple parallel TCP streams. In view of these new developments, it is imperative to revisit the question: Which congestion control algorithms are best adapted to current networking environments? In order to help resolve this question, we contribute the following: (i) using first principles, we develop a general throughput-prediction framework that takes into account buffer sizes and maximum window constraints; (ii) we validate the models using measurements and achieve low prediction errors; (iii) we note differences in TCP dynamics between two experimental configurations and find one of them to be significantly more deterministic than the other; we also find that CUBIC and H-TCP outperform STCP, especially when multiple streams are used; and (iv) we present preliminary results for modelling multiple TCP streams for CUBIC and STCP.

Vardoyan, G.↗