Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Using”

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 1,423 records · Page 79
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

Evaluating Online Global Recovery with Fenix Using Application-Aware In-Memory Checkpointing Techniques

Exascale systems promise the potential for computation at unprecedented scales and resolutions, but achieving exascale by the end of this decade presents significant challenges. A key challenge is due to the very large number of cores and components and the resulting mean time between failures (MTBF) in the order of hours or minutes. Since the typical run times of target scientific applications are longer than this MTBF, fault tolerance techniques will be essential. An important class of failures that must be addressed is process or node failures. While checkpoint/restart (C/R) is currently the most widely accepted technique for addressing processor failures, coordinated, stable-storage-based global C/R might be unfeasible at exascale when the time to checkpoint exceeds the expected MTBF. This paper explores transparent recovery via implicitly coordinated, diskless, application-driven checkpointing as a way to tolerate process failures in MPI applications at exascale. The discussed approach leverages User Level Failure Mitigation (ULFM), which is being proposed as an MPI extension to allow applications to create policies for tolerating process failures. Specifically, this paper demonstrates how different implementations of application-driven in-memory checkpoint storage and recovery compare in terms of performance and scalability. We also experimentally evaluate the effectiveness and scalability of the Fenix online global recovery framework on a production system -the Titan Cray XK7 at ORNL- and demonstrate the ability of Fenix to tolerate dynamically injected failures using the execution of four benchmarks and mini-applications with different behaviors.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

Investigation of Thermal Interface Materials Using Phase-Sensitive Transient Thermoreflectance Technique: Preprint

With increasing power density in electronics packages/modules, thermal resistances at multiple interfaces are a bottleneck to efficient heat removal from the package. In this work, the performance of thermal interface materials such as grease, thermoplastic adhesives and diffusion-bonded interfaces are characterized using the phase-sensitive transient thermoreflectance technique. A multi-layer heat conduction model was constructed and theoretical solutions were derived to obtain the relation between phase lag and the thermal/physical properties. This technique enables simultaneous extraction of the contact resistance and bulk thermal conductivity of the TIMs. With the measurements, the bulk thermal conductivity of Dow TC-5022 thermal grease (70 to 75 um bondline thickness) was 3 to 5 W/(m-K) and the contact resistance was 5 to 10 mm2-K/W. For the Btech thermoplastic material (45 to 80 μm bondline thickness), the bulk thermal conductivity was 20 to 50 W/(m-K) and the contact resistance was 2 to 5 mm2-K/W. Measurements were also conducted to quantify the thermal performance of diffusion-bonded interface for power electronics applications. Results with the diffusion-bonded sample showed that the interfacial thermal resistance is more than one order of magnitude lower than those of traditional TIMs, suggesting potential pathways to efficient thermal management.

30 DIRECT ENERGY CONVERSION↗
DOE OSTI2019

Reaction Rate Enhancement for Cu(In,Ga)Se 2 Absorber Materials Using Ag-Alloying

The addition of Ag to Cu-Ga-In precursors for synthesizing (Ag,Cu)(In,Ga)Se 2 (ACIGS) thin films has shown benefits including improved adhesion, greater process tolerance, and potential for improved device performance. In this study, reaction pathways to form Cu(In,Ga)Se 2 (CIGS) and ACIGS were studied by time-progressive reactions at 450 °C in a 5% Ar/H 2 Se atmosphere followed by ex situ characterization. Results indicated that the addition of 25% Ag/(Ag+Cu) to the CIGS film reduces the reaction time by 50%. X-ray diffraction (XRD) analysis of CIGS films showed that the CuInSe 2 phase initially formed after 3.5 min. The slow reaction of the stable γ-Cu 9 (In,Ga) 4 phase, however, required more than 20 min to complete. Importantly, the addition of Ag to the CIGS film accelerated the reaction. Energy-dispersive X-ray spectroscopy shows that Ga/(Ga+In) grading occurs in the first 10 min of the reaction. XRD analysis showed that the chalcopyrite phase fully forms after 10 min and no significant changes were observed in samples selenized from 10-45 min. Reaction pathways of Ag-alloyed films were further characterized using in situ high temperature XRD analysis. The onset temperature of Se reaction was detected at 230 °C and a AgIn 2 phase transformation to (Ag,Cu)In 2 occurred during the early stage of the reaction.

36 MATERIALS SCIENCE↗
DOE OSTI2012

PV Ramping in a Distributed Generation Environment: A Study Using Solar Measurements; Preprint

Variability in Photovoltaic (PV) generation resulting from variability in the solar radiation over the PV arrays is a topic of continuing concern for those involved with integrating renewables onto existing electrical grids. The island of Lanai, Hawaii is an extreme example of the challenges that integrators will face due to the fact that it is a small standalone grid. One way to study thisproblem is to take high-resolution solar measurements in multiple locations and model simultaneous PV production for various sizes at those locations. The National Renewable Energy Laboratory (NREL) collected high-resolution solar data at four locations on the island where proposed PV plants will be deployed in the near future. This data set provides unique insight into how the solar radiationmay vary between points that are proximal in distance, but diverse in weather, due to the formation of orographic clouds in the center of the island. Using information about each proposed PV plant size, power output was created at high resolution. The team analyzed this output to understand power production ramps at individual locations and the effects of aggregating the production from all fourlocations. Hawaii is a unique environment, with extremely variable events occurring on a daily basis. This study provided an excellent opportunity for understanding potential worst-case scenarios for PV ramping. This paper provides an introduction to the datasets that NREL collected over a year and a comprehensive analysis of PV variability in a distributed generation scenario.

14 SOLAR ENERGY↗
DOE OSTI2012

Examination of 1D Solar Cell Model Limitations Using 3D SPICE Modeling: Preprint

To examine the limitations of one-dimensional (1D) solar cell modeling, 3D SPICE-based modeling is used to examine in detail the validity of the 1D assumptions as a function of sheet resistance for a model cell. The internal voltages and current densities produced by this modeling give additional insight into the differences between the 1D and 3D models.

14 SOLAR ENERGY↗
DOE OSTI2015

Hydro-frac monitoring using ground time-domain electromagnetics: Hydro-frac monitoring using ground time-domain EM

& Engineers. As motivation for considering new electromagnetic techniques for hydraulic fracture monitoring, we develop a simple financial model for the net present value offered by geophysical characterization to reduce the error in stimulated reservoir volume calculations. Additionally, this model shows that even a 5% improvement in stimulated reservoir volume for a 1 billion barrel (bbl) field results in over 1 billion U.S. dollars (US$) in net present value over 24 years for US$100/bbl. oil and US$0.5 billion for US$50/bbl. oil. The application of conductivity upscaling, often used in electromagnetic modeling to reduce mesh size and thus simulation runtimes, is shown to be inaccurate for the high electrical contrasts needed to represent steel-cased wells in the earth. Fine-scale finite-difference modeling with 12.22-mm cells to capture the steel casing and fractures shows that the steel casing provides a direct current pathway to a created fracture that significantly enhances the response compared with neglecting the steel casing. We consider conductively enhanced proppant, such as coke-breeze-coated sand, and a highly saline brine solution to produce electrically conductive fractures. For a relatively small frac job at a depth of 3 km, involving 5,000 bbl. of slurry and a source midpoint to receiver separation of 50 m, the models show that the conductively enhanced proppant produces a 15% increase in the electric field strength (in-line with the transmitter) in a 10-Ωm background. In a 100-Ωm background, the response due to the proppant increases to 213%. Replacing the conductive proppant by brine with a concentration of 100,000-ppm NaCl, the field strength is increased by 23% in the 100-Ωm background and by 2.3% in the 10-Ωm background. Finally, all but the 100,000-ppm NaCl brine in a 10-Ωm background produce calculated fracture-induced electric field increases that are significantly above 2%, a value that has been demonstrated to be observable in field measurements. © 2015 European Association of Geoscientists

58 GEOSCIENCES↗
DOE OSTI2022

Deterministic Linear Time for Maximal Poisson‐Disk Sampling using Chocks without Rejection or Approximation

Abstract We show how to sample uniformly within the three‐sided region bounded by a circle, a radial ray, and a tangent, called a “chock.” By dividing a 2D planar rectangle into a background grid, and subtracting Poisson disks from grid squares, we are able to represent the available region for samples exactly using triangles and chocks. Uniform random samples are generated from chock areas precisely without rejection sampling. This provides the first implemented algorithm for precise maximal Poisson‐disk sampling in deterministic linear time. We prove O(n · M(b) log b), where n is the number of samples, b is the bits of numerical precision and M is the cost of multiplication. Prior methods have higher time complexity, take expected time, are non‐maximal, and/or are not Poisson‐disk distributions in the most precise mathematical sense. We fill this theoretical lacuna.

Mitchell, Scott A.↗