Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “distributed processing”

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 613 records · Page 34
All sourcesNASA NTRSDOE OSTI
DOE OSTI2015

A generalized parallel replica dynamics

Metastability is a common obstacle to performing long molecular dynamics simulations. Many numerical methods have been proposed to overcome it. One method is parallel replica dynamics, which relies on the rapid convergence of the underlying stochastic process to a quasi-stationary distribution. Two requirements for applying parallel replica dynamics are knowledge of the time scale on which the process converges to the quasi-stationary distribution and a mechanism for generating samples from this distribution. By combining a Fleming–Viot particle system with convergence diagnostics to simultaneously identify when the process converges while also generating samples, we can address both points. This variation on the algorithm is illustrated with various numerical examples, including those with entropic barriers and the 2D Lennard-Jones cluster of seven atoms.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2012

Interpreting Temperature Strain Data from Meso-Scale Clathrate Experiments

Gas hydrates are important in global climate change, carbon sequestra- tion, and seafloor stability. Currently, formation and dissociation pathways are poorly defined. We present a new approach for processing large amounts of data from meso-scale experiments, such as the LUNA distributed sensing system (DSS) in the seafloor process simulator (SPS) at Oak Ridge National Laboratory. The DSS provides a proxy for temperature measurement with a high spatial resolution allowing the heat of reaction during gas hydrate formation/dissociation to aid in locating clathrates in the vessel. The DSS fibers are placed in the sediment following an Archimedean spiral design and then the position of each sensor is solved by iterating over the arc length formula with Newtons method. The data is then gridded with 1 a natural neighbor interpolation algorithm to allow contouring of the data. The solution of the sensor locations is verified with hot and cold stimulus in known locations. An experiment was preformed with a vertically split column of sand and silt. The DSS system clearly showed hydrate forming in the sand first, then slowly creeping into the silt. Similar systems and data processing techniques could be used for monitoring of hydrates in natural environments or in any situation where a hybrid temperature/strain index is useful. Further ad- vances in fiber technology allow the fiber to be applied in any configuration and the position of each sensor to be precisely determined making practical applications easier.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2012

Simulating diffusion processes in discontinuous media: A numerical scheme with constant time steps

In this article, we propose new Monte Carlo techniques for moving a diffusive particle in a discontinuous media. In this framework, we characterize the stochastic process that governs the positions of the particle. The key tool is the reduction of the process to a Skew Brownian motion (SBM). In a zone where the coefficients are locally constant on each side of the discontinuity, the new position of the particle after a constant time step is sampled from the exact distribution of the SBM process at the considered time. To do so, we propose two different but equivalent algorithms: a two-steps simulation with a stop at the discontinuity and a one-step direct simulation of the SBM dynamic. Some benchmark tests illustrate their effectiveness.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2013

2-dimensional ion velocity distributions measured by laser-induced fluorescence above a radio-frequency biased silicon wafer

The dynamics of ions traversing sheaths in low temperature plasmas are important to the formation of the ion energy distribution incident onto surfaces during microelectronics fabrication. Ion dynamics have been measured using laser-induced fluorescence (LIF) in the sheath above a 30 cm diameter, 2.2 MHz-biased silicon wafer in a commercial inductively coupled plasma processing reactor. The velocity distribution of argon ions was measured at thousands of positions above and radially along the surface of the wafer by utilizing a planar laser sheet from a pulsed, tunable dye laser. Velocities were measured both parallel and perpendicular to the wafer over an energy range of 0.4–600 eV. The resulting fluorescence was recorded using a fast CCD camera, which provided resolution of 0.4 mm in space and 30 ns in time. Data were taken at eight different phases during the 2.2 MHz cycle. The ion velocity distributions (IVDs) in the sheath were found to be spatially non-uniform near the edge of the wafer and phase-dependent as a function of height. Several cm above the wafer the IVD is Maxwellian and independent of phase. Experimental results were compared with simulations. The experimental time-averaged ion energy distribution function as a function of height compare favorably with results from the computer model.

36 MATERIALS SCIENCE↗
DOE OSTI2011

Observation of pressure gradient and related flow rate effect on the plasma parameters in plasma processing reactor

In industrial plasma processes, flow rate has been known to a key to control plasma processing results and has been discussed with reactive radical density, gas residence time, and surface reaction. In this study, it was observed that the increase in the flow rate can also change plasma parameters (electron temperature and plasma density) and electron energy distribution function in plasma processing reactor. Based on the measurement of gas pressure between the discharge region and the pumping port region, the considerable differences in the gas pressure between the two regions were found with increasing flow rate. It was also observed that even in the discharge region, the pressure gradient occurs at the high gas flow rate. This result shows that increasing the flow rate results in the pressure gradient and causes the changes in the plasma parameters.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2023

FPGA Acceleration of GCN in Light of the Symmetry of Graph Adjacency Matrix

Graph Convolutional Neural Networks (GCNs) are widely used to process large-scale graph data. Different from deep neural networks (DNNs), GCNs are sparse, irregular, and unstructured, posing unique challenges to hardware acceleration with regular processing elements (PEs). In particular, the adjacency matrix of a GCN is extremely sparse, leading to frequent but irregular memory access, low spatial/temporal data locality and poor data reuse. Furthermore, a realistic graph usually consists of unstructured data (e.g., unbalanced distributions), creating significantly different processing times and imbalanced workload for each node in GCN acceleration. To overcome these challenges, we propose an end-to-end hardware-software co-design to accelerate GCNs on resource-constrained FPGAs with the features including: (1) A custom dataflow that leverages symmetry along the diagonal of the adjacency matrix to accelerate feature aggregation for undirected graphs. We utilize either the upper or the lower triangular matrix of the adjacency matrix to perform aggregation in GCN to improve data reuse. (2) Unified compute cores for both aggregation and transform phases, with full support to the symmetry-based dataflow. These cores can be dynamically reconfigured to the systolic mode for transformation or as individual accumulators for aggregation in GCN processing. (3) Preprocessing of the graph in software to rearrange the edges and features to match the custom dataflow. This step improves the regularity in memory access and data reuse in the aggregation phase. Moreover, we quantize the GCN precision from FP32 to INT8 to reduce the memory footprint without losing the inference accuracy. We implement our accelerator design in Intel Stratix10 MX FPGA board with HBM2, and demonstrate 1.3x-110.5x improvement in end-to-end GCN latency as compared to the state-of the-art FPGA implementations, on the graph datasets of Cora, Pubmed, Citeseer and Reddit.

Nair, Gopikrishnan R.↗
DOE OSTI2016

Optical arc sensor using energy harvesting power source

Wireless sensors without external power supply gained considerable attention due to convenience both in installation and operation. Optical arc detecting sensor equipping with self sustaining power supply using energy harvesting method was investigated. Continuous energy harvesting method was attempted using thermoelectric generator to supply standby power in micro ampere scale and operating power in mA scale. Peltier module with heat-sink was used for high efficiency electricity generator. Optical arc detecting sensor with hybrid filter showed insensitivity to fluorescent and incandescent lamps under simulated distribution panel condition. Signal processing using integrating function showed selective arc discharge detection capability to different arc energy levels, with a resolution below 17 J energy difference, unaffected by bursting arc waveform. The sensor showed possibility for application to arc discharge detecting sensor in power distribution panel. Also experiment with proposed continuous energy harvesting method using thermoelectric power showed possibility as a self sustainable power source of remote sensor.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2016

Nautical Dawn - Wide-area Motion Imagery and RF Compressive Sensing Applications: CRADA Final Report

The global information grid (GIG) provides access to massive amounts of data from a vast number of intelligence, surveillance, and reconnaissance (ISR) sensors and sources. A significant challenge exists in the actual exploitation of the sensor data. Sensor data is often large, and in the present paradigm analysts extract information from it in raw form. This approach taxes the limited network-bandwidth of the GIG, since the raw data must be transmitted to analysts prior to processing. The parties proposed a new approach to advance the rate of production of actionable information from ISR data, one that extracts information from raw data closer to the sensors. This research developed an information-centric capability that uses a distributed compressive sensing and processing (DCSP) framework.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2022

Population balance models for polymer upcycling: signatures of the mechanism in the molecular weight evolution

Chemical and catalytic upcycling processes could help realize a circular plastics economy, but current models for testing mechanistic hypotheses and designing catalysts remain primitive. This work shows how proposed catalytic mechanisms can be incorporated into population balance models to predict the time evolution of molecular weight distributions. We develop models for homogeneous and heterogeneous catalysts, including catalysts that cut at chain ends and catalysts that cut at random locations. For heterogeneous catalysts, we illustrate the effect of adsorption constants that depend on polymer chain length. Here, we discuss ongoing efforts and challenges in measuring and modeling the time evolving molecular weight distributions in polymer upcycling processes.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2018

On Maximal Hard-Core Thinnings of Stationary Particle Processes

The present paper studies existence and distributional uniqueness of subclasses of stationary hard-core particle systems arising as thinnings of stationary particle processes. These subclasses are defined by natural maximality criteria. We investigate two specific criteria, one related to the intensity of the hard-core particle process, the other one being a local optimality criterion on the level of realizations. In fact, the criteria are equivalent under suitable moment conditions. We show that stationary hard-core thinnings satisfying such criteria exist and are frequently distributionally unique. More precisely, distributional uniqueness holds in subcritical and barely supercritical regimes of continuum percolation. Additionally, based on the analysis of a specific example, we argue that fluctuations in grain sizes can play an important role for establishing distributional uniqueness at high intensities. Finally, we provide a family of algorithmically constructible approximations whose volume fractions are arbitrarily close to the maximum.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2016

Work stealing for GPU-accelerated parallel programs in a global address space framework

Task parallelism is an attractive approach to automatically load balance the computation in a parallel system and adapt to dynamism exhibited by parallel systems. Exploiting task parallelism through work stealing has been extensively studied in shared and distributed-memory contexts. In this paper, we study the design of a system that uses work stealing for dynamic load balancing of task-parallel programs executed on hybrid distributed-memory CPU-graphics processing unit (GPU) systems in a global-address space framework. We take into account the unique nature of the accelerator model employed by GPUs, the significant performance difference between GPU and CPU execution as a function of problem size, and the distinct CPU and GPU memory domains. We consider various alternatives in designing a distributed work stealing algorithm for CPU-GPU systems, while taking into account the impact of task distribution and data movement overheads. These strategies are evaluated using microbenchmarks that capture various execution configurations as well as the state-of-the-art CCSD(T) application module from the computational chemistry domain

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Organismal and spatial partitioning of energy and macronutrient transformations within a hypersaline mat

Phototrophic mat communities are model ecosystems for studying energy cycling and elemental transformations because complete biogeochemical cycles occur over millimeter-to-centimeter scales. Characterization of energy and nutrient capture within hypersaline phototrophic mats has focused on specific processes and organisms, however little is known about community-wide distribution of and linkages between these processes. To investigate energy and macronutrient capture and flow through a structured community, the spatial and organismal distribution of metabolic functions within a compact hypersaline mat community from Hot Lake have been broadly elucidated through species-resolved metagenomics and geochemical, microbial diversity, and metabolic gradient measurements. Draft reconstructed genomes of abundant organisms revealed three dominant cyanobacterial populations differentially distributed across the top layers of the mat suggesting niche separation along light and oxygen gradients. Many organisms contained diverse functional profiles, allowing for metabolic response to changing conditions within the mat. Organisms with partial nitrogen and sulfur metabolisms were widespread indicating dependence upon metabolite exchange. In addition, changes in community spatial structure were observed over the diel. These results indicate that organisms within the mat community have adapted to the temporally dynamic environmental gradients in this hypersaline mat through metabolic flexibility and fluid syntrophic interactions, including shifts in spatial arrangements.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

Distributed Acoustic Sensing Based on Coherent Microwave Photonics Interferometry

A microwave photonics method has been developed for measuring distributed acoustic signals. This method uses microwave-modulated low coherence light as a probe to interrogate distributed in-fiber interferometers, which are used to measure acoustic-induced strain. By sweeping the microwave frequency at a constant rate, the acoustic signals are encoded into the complex microwave spectrum. The microwave spectrum is transformed into the joint time–frequency domain and further processed to obtain the distributed acoustic signals. The method is first evaluated using an intrinsic Fabry Perot interferometer (IFPI). Acoustic signals of frequency up to 15.6 kHz were detected. The method was further demonstrated using an array of in-fiber weak reflectors and an external Michelson interferometer. Two piezoceramic cylinders (PCCs) driven at frequencies of 1700 Hz and 3430 Hz were used as acoustic sources. The experiment results show that the sensing system can locate multiple acoustic sources. The system resolves 20 nε when the spatial resolution is 5 cm. The recovered acoustic signals match the excitation signals in frequency, amplitude, and phase, indicating an excellent potential for distributed acoustic sensing (DAS).

47 OTHER INSTRUMENTATION↗
DOE OSTI2020

Realistic Synthetic Distribution Grids: Summary of Validation Results

The presentation slides will walk us through the validation process and results of the synthetic distribution grid data. The synthetic grids for the distribution systems were developed as part of the ARPA-E funded GRIDDATA program project, SMART-DS. Three large scale datasets for US style distribution system were developed: 1) SAF: Santa Fe urban region, 2) GSO: Greensboro urban, semi-urban and industrial region, and 3) SFO: San Francisco Bay Area urban, semi-urban, industrial and rural areas. These datasets were validated using a three-pronged validation process, including statistical, operational and expert validation phases; the results of which have been summarized in this deck.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2014

Predictions of Water-Column Properties under Widespread Artificial Upwelling Scenarios in the North Pacific Subtropical Gyre using an Ocean General Circulation Model

The physical evolution of water-column stratification following the implementation of widespread artificial upwelling across the North Pacific Subtropical Gyre (NPSG) is examined in a numerical experiment using the ocean general circulation model MITgcm. The upwelling process is represented by distributed pairs of fluid sinks and sources located at different depths to mimic the withdrawal of cool nutrient-rich water at about 400 m below the surface and its discharge in the upper water column. The overarching conclusion from the simulations is that within the NPSG, the upwelling process as modeled appears to enhance relatively strong stratification and to weaken relatively weak stratification. This generally holds for the short term as well as when a new steady-state is reached after a few decades. In the latter case, it applies temporally, with deeper mixed layers in the winter and shallower ones in the summer, as well as geographically with a broad separation between less stratified northern regions and more stratified southern regions within the NPSG. Here, the global model allows an estimation of possible unintended consequences well outside the upwelling implementation area. It is predicted, for example, that surface cooling in the NPSG allows heat to penetrate the ocean interior until the overall oceanic surface heat flux is restored to zero (in a yearly averaged sense). As a result, changes in the thermal structure of the water column occur. Regions with a net surface warming develop outside the NPSG, most notably along the East coast of Japan and the West coasts of the Americas.

13 HYDRO ENERGY↗
DOE OSTI2018

Rare behavior of growth processes via umbrella sampling of trajectories

We compute probability distributions of trajectory observables for reversible and irreversible growth processes. These results reveal a correspondence between reversible and irreversible processes, at particular points in parameter space, in terms of their typical and atypical trajectories. Thus key features of growth processes can be insensitive to the precise form of the rate constants used to generate them, recalling the insensitivity to microscopic details of certain equilibrium behavior. We obtained these results using a sampling method, inspired by the "s-ensemble" large-deviation formalism, that amounts to umbrella sampling in trajectory space. The method is a simple variant of existing approaches, and applies to ensembles of trajectories controlled by the total number of events. It can be used to determine large-deviation rate functions for trajectory observables in or out of equilibrium.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Neural network modeling of in situ fluid-filled pore size distributions in subsurface shale reservoirs under data constraints

Subsurface nuclear magnetic resonance (NMR) logs acquired in the wellbore environment are sensitive to fluid-filled pore size distribution, fluid mobility, permeability, and porosity in the near-wellbore reservoir volume. NMR response of a formation layer is processed to extract the T2 distribution, which approximates the fluid-filled pore size distribution. NMR logs are acquired in limited number of wells due to financial and operational challenges, which adversely affects reservoir characterization. We developed here two neural-network-based machine learning techniques, long short-term memory (LSTM) network and variational autoencoder with a convolutional layer (VAEc) network, to process the ‘easy-to-acquire’ formation mineral and fluid saturation logs to generate synthetic NMR T2 distributions in the absence of ‘hard-to-acquire’ NMR T2 distribution log. Both the predictive models are trained and tested on limited wireline log measurements randomly selected from a 300-ft depth interval of the Bakken shale formation. Synthesis performances of LSTM and VAEc models in terms of R 2 are 0.78 and 0.75, respectively. Noise is inevitable in logging data due to the complex wellbore and formation conditions. Notably, both the predictive models robustly synthesize the fluid-filled pore size distributions in the presence of 50% noise in input logs and 30% noise in training T2 data. The performance of the proposed methodology improves with access to larger volume of training data from other formation types. The proposed method is critical to the synthesis of in situ fluid-filled pore size distributions in shale formations under data constraints due to financial and operational challenges.

58 GEOSCIENCES↗