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 577 records · Page 32
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

A comparative analysis of residual stresses from friction stir processing of aluminum cast 380 and wrought 7075 alloy sheets: experimental characterization and modeling

Residual stresses are often overlooked in friction stir processing (FSP), but their significant impact on fatigue performance necessitates their consideration in optimizing processing parameters. The first step in this effort is understanding how process conditions influence residual stress distributions, especially across different alloys. This study focuses on determining and explaining the through-thickness residual stress variations and the effect of process temperature on the residual stress magnitude in wrought AA7075 and cast AA380.0 alloys. Additionally, for AA380.0, the impact of a second FSP pass was investigated. To achieve this, hole-drilling electronic speckle pattern interferometry (ESPI) and the thermal pseudo-mechanical (TPM) model within finite element analysis were employed to study the 3D distributions of in-plane residual stresses in processed samples under various conditions. A key finding was the varying impact of process temperatures on residual stress magnitudes. Higher process temperatures reduced stresses in AA380.0 but increased them in AA7075. Additionally, the through-thickness stress distributions differed between the two alloys. Further analysis revealed that yield stresses are crucial in explaining these phenomena and the effects of additional FSP passes. Further, this fundamental understanding will be vital in guiding the efforts to mitigate residual stresses and assess their impact on the performance of FSP aluminum alloys.

36 MATERIALS SCIENCE↗
DOE OSTI2015

Simulative design and process optimization of the two-stage stretch-blow molding process

The total production costs of PET bottles are significantly affected by the costs of raw material. Approximately 70 % of the total costs are spent for the raw material. Therefore, stretch-blow molding industry intends to reduce the total production costs by an optimized material efficiency. However, there is often a trade-off between an optimized material efficiency and required product properties. Due to a multitude of complex boundary conditions, the design process of new stretch-blow molded products is still a challenging task and is often based on empirical knowledge. Application of current CAE-tools supports the design process by reducing development time and costs. This paper describes an approach to determine optimized preform geometry and corresponding process parameters iteratively. The wall thickness distribution and the local stretch ratios of the blown bottle are calculated in a three-dimensional process simulation. Thereby, the wall thickness distribution is correlated with an objective function and preform geometry as well as process parameters are varied by an optimization algorithm. Taking into account the correlation between material usage, process history and resulting product properties, integrative coupled simulation steps, e.g. structural analyses or barrier simulations, are performed. The approach is applied on a 0.5 liter PET bottle of Krones AG, Neutraubling, Germany. The investigations point out that the design process can be supported by applying this simulative optimization approach. In an optimization study the total bottle weight is reduced from 18.5 g to 15.5 g. The validation of the computed results is in progress.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2016

Frequency domain analysis of errors in cross-correlations of ambient seismic noise

We analyse random errors (variances) in cross-correlations of ambient seismic noise in the frequency domain, which differ from previous time domain methods. Extending previous theoretical results on ensemble averaged cross-spectrum, we estimate confidence interval of stacked cross-spectrum of finite amount of data at each frequency using non-overlapping windows with fixed length. The extended theory also connects amplitude and phase variances with the variance of each complex spectrum value. Analysis of synthetic stationary ambient noise is used to estimate the confidence interval of stacked cross-spectrum obtained with different length of noise data corresponding to different number of evenly spaced windows of the same duration. This method allows estimating Signal/Noise Ratio (SNR) of noise cross-correlation in the frequency domain, without specifying filter bandwidth or signal/noise windows that are needed for time domain SNR estimations. Based on synthetic ambient noise data, we also compare the probability distributions, causal part amplitude and SNR of stacked cross-spectrum function using one-bit normalization or pre-whitening with those obtained without these pre-processing steps. Natural continuous noise records contain both ambient noise and small earthquakes that are inseparable from the noise with the existing pre-processing steps. Using probability distributions of random cross-spectrum values based on the theoretical results provides an effective way to exclude such small earthquakes, and additional data segments (outliers) contaminated by signals of different statistics (e.g. rain, cultural noise), from continuous noise waveforms. This technique is applied to constrain values and uncertainties of amplitude and phase velocity of stacked noise cross-spectrum at different frequencies, using data from southern California at both regional scale (~35 km) and dense linear array (~20 m) across the plate-boundary faults. A block bootstrap resampling method is used to account for temporal correlation of noise cross-spectrum at low frequencies (0.05–0.2 Hz) near the ocean microseismic peaks.

58 GEOSCIENCES↗
DOE OSTI2019

Thermomechanical Analysis of A-TIG and MP-TIG Welding of 2.25Cr-1Mo Steel Considering Phase Transformation

In the present study, a finite element-based 3D mesh model is used to simulate the effect of welding processes on the temperature distribution, residual stress and distortion in 11-mm-thick 2.25Cr-1Mo steel weld joints. Tungsten inert gas (TIG) and activated-TIG (A-TIG) welding processes were employed for the fabrication of weld joints. Austenite-to-bainite phase transformation is taken into account in the numerical analysis. The heat input fitting tool has been used to calibrate the heat source by matching the simulated weld bead with that of the actual weld bead. Peak temperature was higher at the weld center for A-TIG weld joint, and the weld joint exhibited a sharp temperature gradient at the weld interface. There was good agreement between the predicted and measured thermal cycles. A-TIG and multipass-TIG (MP-TIG) weld joints exhibited similar longitudinal residual stresses in heat affected zone with a difference of 50 MPa. Moreover, A-TIG weld joint fabricated by double-side welding technique employing two passes exhibited reduced distortion than MP-TIG weld joints. A higher magnitude of distortion was in MP-TIG weld joint which was attributed to more weld metal volume caused by more number of weld passes required to fill the V-groove. There was good agreement between the predicted and measured residual stress and distortion values.

36 MATERIALS SCIENCE↗
DOE OSTI2011

Experimental and Numerical Analysis on the Distortion of Parts Made of 20MnCr5 by Hot Metal Forming

For high performance applications, shafts and gears made of 20MnCr5 (AISI 5120) are manufactured in large numbers every year. Inhomogeneities in the material properties, process perturbations and asymmetries in shape and operation setups provide a potential for the distortion of parts, often released by heat treatment. In this contribution experimental results on the distortion of shafts and the dishing of disk-like gear wheel blanks are presented. The numerical analysis of the hot-rolling process allowed to trace a peculiar segregation distribution at the cross-section of the bars back to the casting process, and to identify an asymmetric strain distribution which may be the main cause for shaft distortion. For the dishing of the disks a correlation to the resulting distribution of the material flow was found and, a process perturbation parameter identified which is assumed to be responsible for the observed material flow variation.

36 MATERIALS SCIENCE↗
DOE OSTI2016

ENERGY DISSIPATION AND LANDAU DAMPING IN TWO- AND THREE-DIMENSIONAL PLASMA TURBULENCE

Plasma turbulence is ubiquitous in space and astrophysical plasmas, playing an important role in plasma energization, but the physical mechanisms leading to dissipation of the turbulent energy remain to be definitively identified. Kinetic simulations in two dimensions (2D) have been extensively used to study the dissipation process. How the limitation to 2D affects energy dissipation remains unclear. This work provides a model of comparison between two- and three-dimensional (3D) plasma turbulence using gyrokinetic simulations; it also explores the dynamics of distribution functions during the dissipation process. It is found that both 2D and 3D nonlinear gyrokinetic simulations of a low-beta plasma generate electron velocity-space structures with the same characteristics as that of the linear Landau damping of Alfvén waves in a 3D linear simulation. The continual occurrence of the velocity-space structures throughout the turbulence simulations suggests that the action of Landau damping may be responsible for the turbulent energy transfer to electrons in both 2D and 3D, and makes possible the subsequent irreversible heating of the plasma through collisional smoothing of the velocity-space fluctuations. Although, in the 2D case where variation along the equilibrium magnetic field is absent, it may be expected that Landau damping is not possible, a common trigonometric factor appears in the 2D resonant denominator, leaving the resonance condition unchanged from the 3D case. The evolution of the 2D and 3D cases is qualitatively similar. However, quantitatively, the nonlinear energy cascade and subsequent dissipation is significantly slower in the 2D case.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2011

PRODUCTION OF THE p-PROCESS NUCLEI IN THE CARBON-DEFLAGRATION MODEL FOR TYPE Ia SUPERNOVAE

We calculate the nucleosynthesis of proton-rich isotopes in the carbon-deflagration model for Type Ia supernovae (SNe Ia). The seed abundances are obtained by calculating the s-process nucleosynthesis that is expected to occur in the repeating helium shell flashes on the carbon-oxygen (CO) white dwarf (WD) during mass accretion from a binary companion. When the deflagration wave passes through the outer layer of the CO WD, p-nuclei are produced by photodisintegration reactions on s-nuclei in a region where the peak temperature ranges from 1.9 to 3.6 x 10{sup 9} K. We confirm the sensitivity of the p-process on the initial distribution of s-nuclei. We show that the initial C/O ratio in the WD does not affect much the yield of p-nuclei. On the other hand, the abundance of {sup 22}Ne left after s-processing has a large influence on the p-process via the {sup 22}Ne({alpha},n) reaction. We find that about 50% of p-nuclides are co-produced when normalized to their solar abundances in all adopted cases of seed distribution. Mo and Ru, which are largely underproduced in Type II supernovae (SNe II), are produced more than in SNe II although they are underproduced with respect to the yield levels of other p-nuclides. The ratios between p-nuclei and iron in the ejecta are larger than the solar ratios by a factor of 1.2. We also compare the yields of oxygen, iron, and p-nuclides in SNe Ia and SNe II and suggest that SNe Ia could make a larger contribution than SNe II to the solar system content of p-nuclei.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Design of Resilient Electric Distribution Systems for Remote Communities: Surgical Load Management Using Smart Meters: Preprint

This paper describes a systematic process of designing resilient electric distribution systems and microgrids using smart meters for surgical load management (SLM) as part of Advanced Metering Infrastructure (AMI). The work focuses on selection approach, integration, and interoperability aspects for AMI in microgrids. SLM is proposed as a granular control methodology for serving selective critical loads across different distribution feeders in the system during extreme events. The surgical load shedding as well as load pick-up provides a robust approach for maximizing critical load served in a resource-constrained electric distribution system or a microgrid. We present the case of a 20 MW islanded microgrid in Cordova, AK, USA, which is the demonstration site for field validation of resilience enhancement technologies for the DOE-funded Grid Modernization project RADIANCE. Cordova microgrid is an islanded distribution grid that provides an environment to prove the approach, and the techniques may also be applicable to other regional distribution systems.

microgrids↗
DOE OSTI2012

Asymptotic equilibrium between Langmuir turbulence and suprathermal electrons in three dimensions

The suprathermal electron distributions with non-Maxwellian tail are commonly detected in space environment as well as in the laboratory. In a previous article [P. H. Yoon, Phys. Plasmas 18, 122303 (2011)], a self-consistent theory of asymptotically steady-state electron distribution function and Langmuir turbulence intensity was formulated in which the steady-state electron distribution is obtained by balancing the spontaneous (or collisional) drag term and velocity space diffusion term, while the steady-state turbulence intensity was obtained by balancing spontaneous and induced processes. The resulting electron distribution function was a type of kappa distribution that features a non-Maxwellian energetic tail component. However, such a discussion was based upon the assumption that both electrons and Langmuir waves are confined to one-dimensional space along the direction of the implicit ambient magnetic field. However, such an assumption may be invalidated when the implicit ambient magnetic field is sufficiently weak or absent. Consequently, in this paper the assumption of one-dimensionality is relaxed to three-dimensional situation.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2021

Inverse Biot–Savart Optimization for Superconducting Accelerator Magnets

Superconducting (SC) magnets for accelerator concepts are often synthesized by numerically optimizing magnetic field waveforms, a process that requires a subsequent solution of a constrained inverse problem to identify suitable SC magnet windings. When the desired field distribution is intuitive, the inverse process is facilitated by seeding preconceived coil distributions into design optimization methods for refinement. With more complex magnetic field distributions, an initial design may be unknown, and topology optimization tools are required to synthesize current distributions without a priori guidance from a subject matter expert. In this work, we develop a constrained inverse Biot-Savart topology optimization methodology that synthesizes optimal distributions of current density in racetrack-like SC coils. The problem structure is exploited through a computationally efficient quadratic programming formulation, and the method is applied to recently published magnetic field waveforms for a recirculating proton phase shifter, a proton therapy gantry, and dipole magnets with sharp field transitions. The method and results herein identify novel winding configurations that can help magnet designers bring accelerator concepts to fruition.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2019

Effect of α/β Forging on Microstructure and Texture Inhomogeneity in a Ti-1023 Forged Disk

The microstructure and texture distribution of a Ti-1023 forged disk were investigated by scanning electron microscopy and synchrotron-based high-energy X-ray diffraction. The finite element method was used to simulate temperature and strain distribution in order to investigate the relationship of the α/β forging process with microstructure and texture distribution. A bimodal microstructure and rolling textures with large inhomogeneity were observed in the disk. A plate-like and a necklace-like morphology and volume fraction variations of the primary α phase were observed in different regions with different forging conditions, such as temperature, deformation, position, and high density of flow lines. Texture sharpness distribution of the β phase was in good agreement with the strain distribution, which suggests deformation may play the most important part in the texture inhomogeneity. A weak cube texture was obtained near the center of the disk, and fiber textures were found near the rim of the disk. The primary a phase also exhibited a transverse texture, which is favored by large deformation.

36 MATERIALS SCIENCE↗
DOE OSTI2015

Recent Results From MINERvA

The MINERvA detector is situated in Fermilab's NuMI beam, which provides neutrinos and antineutrinos in the 1-20 GeV range. It is designed to make precision cross-section measurements for scattering processes on various nuclei. These proceedings summarize the differential cross-section distributions measured for several different processes. Comparison of these with various models hints at additional nuclear effects not included in common simulations. These results will help constrain generators' nuclear models and reduce systematic uncertainties on their predictions. An accurate cross-section model, with minimal uncertainties, is vital to oscillation experiments.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2011

On the Use of the Beta Distribution in Probabilistic Resource Assessments

The triangular distribution is a popular choice when it comes to modeling bounded continuous random variables. Its wide acceptance derives mostly from its simple analytic properties and the ease with which modelers can specify its three parameters through the extremes and the mode. On the negative side, hardly any real process follows a triangular distribution, which from the outset puts at a disadvantage any model employing triangular distributions. At a time when numerical techniques such as the Monte Carlo method are displacing analytic approaches in stochastic resource assessments, easy specification remains the most attractive characteristic of the triangular distribution. The beta distribution is another continuous distribution defined within a finite interval offering wider flexibility in style of variation, thus allowing consideration of models in which the random variables closely follow the observed or expected styles of variation. Despite its more complex definition, generation of values following a beta distribution is as straightforward as generating values following a triangular distribution, leaving the selection of parameters as the main impediment to practically considering beta distributions. This contribution intends to promote the acceptance of the beta distribution by explaining its properties and offering several suggestions to facilitate the specification of its two shape parameters. In general, given the same distributional parameters, use of the beta distributions in stochastic modeling may yield significantly different results, yet better estimates, than the triangular distribution.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2018

Integrated Transmission and Distribution Systems Restoration with Distributed Generation Scheduling

The self-healing smart grids enable system operators to quickly and efficiently recover transmission and distribution systems from outages and blackouts. The current restoration procedure solves the transmission and distribution system restoration problems separately without considering their impacts on each other, which may prolong the restoration process. However, with the emerging active distribution networks, distributed generators can contribute into the bottom-up restoration strategy. The interaction between the restoration of two interconnected systems should be taken into account. In this paper, a hybrid model of transmission and distribution systems restoration (HTDSR) is proposed by exchanging data of power and voltage between two separate optimization problems. The effectiveness of the HTDSR algorithm is demonstrated through case studies of the integrated IEEE 13-bus distribution network and IEEE 39-bus transmission grid. Simulation results demonstrate that the integrated restoration strategy with optimal scheduling of distributed generation sources can expedite the restoration process.

Distributed generation↗
DOE OSTI2023

Dynamics of streamflow permanence in a headwater network: Insights from catchment-scale model simulations

The hillslope and channel dynamics that govern streamflow permanence in headwater systems have important implications for ecosystem functioning and downstream water quality. Recent advancements in process-based, semi-distributed hydrologic models that build upon empirical studies of streamflow permanence in well-monitored headwater catchments show promise for characterizing the dynamics of streamflow permanence in headwater systems. However, few process-based models consider the continuum of hillslope-stream network connectivity as a control on streamflow permanence in headwater systems. The objective of this study was to expand a process-based, catchment-scale hydrologic model to better understand the spatiotemporal dynamics of headwater streamflow permanence and to identify controls of streamflow expansion and contraction in a headwater network. Further, we aimed to develop an approach that enhanced the fidelity of model simulations, yet required little additional data, with the intent that the model might be later transferred to catchments with limited long-term and spatially explicit measurements. This approach facilitated network-scale estimates of the controls of streamflow expansion and contraction, albeit with higher degrees of uncertainty in individual reaches due to data constraints. Our model simulated that streamflow permanence was highly dynamic in first-order reaches with steep slopes and variable contributing areas. The simulated stream network length ranged from nearly 98±2% of the geomorphic channel extent during wet periods to nearly 50±10% during dry periods. The model identified a discharge threshold of approximately 1 mm d –1 , above which the rate of streamflow expansion decreases by nearly an order of magnitude, indicating a lack of sensitivity of streamflow expansion to hydrologic forcing during high-flow periods. Overall, we demonstrate that process-based, catchment-scale models offer important insights on the controls of streamflow permanence, despite uncertainties and limitations of the model. We encourage researchers to increase data collection efforts and develop benchmarks to better evaluate such models.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2015

Confidence intervals with a priori parameter bounds

We review the methods of constructing confidence intervals that account for a priori information about one-sided constraints on the parameter being estimated. We show that the so-called method of sensitivity limit yields a correct solution of the problem. Derived are the solutions for the cases of a continuous distribution with non-negative estimated parameter and a discrete distribution, specifically a Poisson process with background. For both cases, the best upper limit is constructed that accounts for the a priori information. A table is provided with the confidence intervals for the parameter of Poisson distribution that correctly accounts for the information on the known value of the background along with the software for calculating the confidence intervals for any confidence levels and magnitudes of the background (the software is freely available for download via Internet).

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2015

Advanced Precursor Reaction Processing for Cu(InGa)(SeS)2 Solar Cells

This project “Advanced Precursor Reaction Processing for Cu(InGa)(SeS)2 Solar Cells”, completed by the Institute of Energy Conversion (IEC) at the University of Delaware in collaboration with the Department of Chemical Engineering at the University of Florida, developed the fundamental understanding and technology to increase module efficiency and improve the manufacturability of Cu(InGa)(SeS)2 films using the precursor reaction approach currently being developed by a number of companies. Key results included: (1) development of a three-step H2Se/Ar/H2S reaction process to control Ga distribution through the film and minimizes back contact MoSe2 formation; (2) Ag-alloying to improve precursor homogeneity by avoiding In phase agglomeration, faster reaction and improved adhesion to allow wider reaction process window; (3) addition of Sb, Bi, and Te interlayers at the Mo/precursor junction to produce more uniform precursor morphology and improve adhesion with reduced void formation in reacted films; (4) a precursor structure containing Se and a reaction process to reduce processing time to 5 minutes and eliminate H2Se usage, thereby increasing throughput and reducing costs. All these results were supported by detailed characterization of the film growth, reaction pathways, thermodynamic assessment and device behavior.

25 ENERGY STORAGE↗
DOE OSTI2024

MDLoader: A Hybrid Model-Driven Data Loader for Distributed Graph Neural Network Training

Scalable data management is essential for processing large scientific dataset on HPC platforms for distributed deep learning. In-memory distributed storage is preferred for its speed, enabling rapid, random, and frequent data access required by stochastic optimizers. Processes use one-sided or collective communication to fetch remote data, with optimal performance depending on (i) dataset characteristics, (ii) training scale, and (iii) interconnection network. Empirical analysis shows collective communication excels with larger mini-batch sizes and/or fewer processes, whereas one-sided communication outperforms at larger scales. We propose MDLoader, a hybrid in-memory data loader for distributed graph neural network training. MDLoader features a model-driven performance estimator that dynamically selects between one-sided and collective communication at the beginning of training using Tree of Parzen Estimators (TPE). Evaluations on NERSC Perlmutter and OLCF Summit show MDLoader outperforms single-backend loaders by up to 2.83 × and predicts the suitable communication method with 96.3% (Perlmutter) and 94.3% (Summit) success rate.

Bae, Jonghyun↗