Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Distributed computation”

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 1,369 records · Page 76
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

A Privacy-Aware Federated Learning Framework for Distributed Energy Resource Analytics in Constrained Environments

To be resilient against extreme weather events, the rural communities in Puerto Rico are leveraging distributed energy resources (DER). However, computing frameworks sup-porting the grid in critical decision-making are still largely centralized. Sensitive consumer data are transmitted over the Internet or cellular networks to a secondary or tertiary node. It guarantees better situational awareness at the cost of a wider attack surface, jeopardizing user privacy, as more DER come online. Cloud, Edge, and Fog computing all require data aggregation at some level. This paper introduces a privacy-aware federated learning framework that leverages the Fog model by pushing analytics all the way to the DER and load assets. These local models train on individual asset data and transmit only learned parameters (such as weights) over secure communications to a global decision-maker. By abstracting personally identifiable consumer data without impacting decision optimality, this framework better aligns with distributed power generation paradigm.

Sundararajan, Aditya↗
DOE OSTI2019

Learning concave-convex profiles of data transport over dedicated connections

Dedicated data transport infrastructures are increasingly being deployed to support distributed big-data and high-performance computing scenarios. These infrastructures employ data transfer nodes that use sophisticated software stacks to support network transport among sites, which often house distributed file and storage systems. Throughput measurements collected over such infrastructures for a range of round trip times (RTTs) reflect the underlying complex end-to-end connections, and have revealed dichotomous throughput profiles as functions of RTT. In particular, concave regions of throughput profiles at lower RTTs indicate near-optimal performance, and convex regions at higher RTTs indicate bottlenecks due to factors such as buffer or credit limits. We present a machine learning method that explicitly infers these concave and convex regions and transitions between them using sigmoid functions. We also provide distribution-free confidence estimates for the generalization error of these concave-convex profile estimates. Throughput profiles for data transfers over 10 Gbps connections with 0–366 ms RTT provide important performance insights, including the near optimality of transfers performed with the XDD tool between XFS filesystems, and the performance limits of wide-area Lustre extensions using LNet routers. A direct application of generic machine learning packages does not adequately highlight these critical performance regions or provide as precise confidence estimates.

Rao, Nageswara↗
DOE OSTI2012

TRACING OUTFLOWS AND ACCRETION: A BIMODAL AZIMUTHAL DEPENDENCE OF Mg II ABSORPTION

We report a bimodality in the azimuthal angle distribution of gas around galaxies as traced by Mg II absorption: halo gas prefers to exist near the projected galaxy major and minor axes. The bimodality is demonstrated by computing the mean azimuthal angle probability distribution function using 88 spectroscopically confirmed Mg II-absorption-selected galaxies [W{sub r} (2796) {>=} 0.1 A] and 35 spectroscopically confirmed non-absorbing galaxies [W{sub r} (2796) < 0.1 A] imaged with Hubble Space Telescope and Sloan Digital Sky Survey. The azimuthal angle distribution for non-absorbers is flat, indicating no azimuthal preference for gas characterized by W{sub r} (2796) < 0.1 A. We find that blue star-forming galaxies clearly drive the bimodality while red passive galaxies may exhibit an excess along their major axis. These results are consistent with galaxy evolution scenarios where star-forming galaxies accrete new gas, forming new stars and producing winds, while red galaxies exist passively due to reduced gas reservoirs. We further compute an azimuthal angle dependent Mg II absorption covering fraction, which is enhanced by as much as 20%-30% along the major and minor axes. The W{sub r} (2796) distribution for gas along the major axis is likely skewed toward weaker Mg II absorption than for gas along the projected minor axis. These combined results are highly suggestive that the bimodality is driven by gas accreted along the galaxy major axis and outflowing along the galaxy minor axis. Adopting these assumptions, we find that the opening angle of outflows and inflows to be 100 Degree-Sign and 40 Degree-Sign , respectively. We find that the probability of detecting outflows is {approx}60%, implying that winds are more commonly observed.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Perturbative readout-error mitigation for near-term quantum computers

Readout errors on near-term quantum computers can introduce significant error to the empirical probability distribution sampled from the output of a quantum circuit. These errors can be mitigated by classical postprocessing given the access of an experimental response matrix that describes the error associated with the measurement of each computational basis state. However, the resources required to characterize a complete response matrix and to compute the corrected probability distribution scale exponentially with the number of qubits, n . In this work, we modify standard matrix inversion techniques using perturbative approximations with significantly reduced complexity and bounded error when the likelihood of high-order bit-flip events is strongly suppressed. Given a characteristic error rate q , we discuss a method to recover the probability of the all-zeros bit string p 0 by sampling only a small subspace of the response matrix before inverting readout error, resulting in a relative speedup of poly [ 2 n / ( n w ) ] , which we motivate using a simplified error model for which the approximation incurs only O ( q w ) error for some integer w . We then provide a generalized technique to efficiently recover full output distributions with O ( q w ) error in the perturbative limit. These approximate techniques for readout-error correction may greatly accelerate near-term quantum computing applications.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Transverse momentum distributions at large- x

We investigate the collinear matching of transverse momentum dependent (TMD) distributions at large values of x, computing and resumming the leading large-x asymptotics for matching coefficients. The large-x resummation is done directly within TMD distributions, ensuring the process-independence of the result. The derived resummation formulas are valid for all TMD distributions (except the pretzelosity). Their application improves perturbative convergence, provides practical estimation for unknown higher-order contributions, and sets restrictions for the nonperturbative part of models. Using the known anomalous dimensions, resummation can reach N 3 LL, often exceeding the accuracy of known coefficient functions.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

Machine Learning-Driven Solvent Screening for Biobased 2,3-Butanediol Extraction

Biobased 2,3-butanediol (2,3-BDO) is a valuable biomass-derived chemical due to its versatility in being transformed into a wide variety of products. However, the separation and purification of 2,3-BDO from fermentation broth remain a significant challenge owing to its high boiling point and hydrophilic nature. Herein, we developed a machine learning (ML)-based screening workflow that uses molecular calculations as training data and requires only a small number of experimental measurements for validation to identify alternative solvent candidates for the liquid–liquid extraction (LLE) of 2,3-BDO from aqueous solution. In particular, 130 density functional theory (DFT) calculations with the implicit solvation method not only built a correlation between the computational partition coefficient and the experimental distribution coefficient of 2,3-BDO but also parameterized an Extra-Trees ML model to screen the distribution coefficient for a wider range of 6717 organic solvents. The experimental measurements of only 24 solvents were needed to validate the computational results. A list of 50 prioritized solvents was proposed for 2,3-BDO LLE, and seven additional experimental measurements were conducted to further verify our selected solvents. The impact of the extraction temperature and solvent-to-feed ratio was also investigated for selected solvents in experiments. Furthermore, this work suggested alternative solvents for 2,3-BDO LLE and proposed a versatile workflow that requires fewer experiments and can be applied to a broader range of LLE studies.

Extraction↗
DOE OSTI2017

Parallel Asynchronous Distributed-Memory Maximal Independent Set Algorithm with Work Ordering

The maximal independent set (MIS) graph problem arises in many applications such as computer vision, information theory, molecular biology, and process scheduling. The growing scale of graph data suggests the use of distributed memory hardware as a cost-effective approach to providing necessary compute and memory resources. Existing distributed memory parallel MIS algorithms rely on synchronous communication and use techniques such as subgraph computations. In this paper, we present an asynchronous distributed-memory parallel graph algorithm that relies on a virtual directed acyclic graph (DAG) that is created during the algorithm execution. We introduce two additional algorithms that save computations by ordering generated work. The first algorithm applies ordering globally to reduce computations, and the second algorithm applies ordering locally at the level of threads to minimize the synchronization overhead. We use two different implementations of Luby’s algorithm variants as baseline to compare the performance of the presented algorithms: (1) vertex-centric Luby A and Luby B implementations, and (2) the CombBLAS linear-algebra Luby A implementation. Results show that proposed algorithms outperform both implementations of Luby algorithms, especially in distributed execution. Furthermore, we show that for low-diameter graphs the algorithm that applies global ordering scales better than other algorithms and for high diameter graphs the original asynchronous algorithm and thread-level ordering algorithm show better performance.

maximal independent set (MIS), distributed memory ↗
DOE OSTI2018

Comparative Analysis of the Powers of the Two-Sample Kolmogorov–Smirnov and Anderson–Darling Tests Under Various Alternatives

In this paper we conduct a comparative analysis of the powers of the two-sample Kolmogorov–Smirnov and Anderson–Darling tests under various alternatives using simulation. We consider two examples. In the first example the alternatives to the standard normal distribution are the distributions of the so-called contaminated normal model. We study the influence of a small contamination with a positive shift on the powers of the test. In the second example the alternatives are the logistic and the Laplace distributions, which are symmetric and differ in shape from the normal distribution having a larger kurtosis coefficient and heavier tails.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Topics in Proton Structure: BSM answers to its radius puzzle and latticesubtleties within its momentum distribution

This thesis covers two subjects, brie y outlined below, of hadronic physics. The proton radius puzzle and the muon anomalous magnetic moment discrepancy point to possible signs of lepton-universality violation. We introduce the context and background necessary to understand these two problems. And we analyze two suitable beyond-the-standard-model solutions, one vector-based and one scalar-based, which simultaneously solves both issues. Furthermore, we demonstrate that the parameter space of the respective solutions can be chosen so as to not be completely forbidden by considered experimental constraints. Specifically, we show that certain violations, analyzed by other authors in the context of similar solutions to those proposed, are ameliorated if one adds to the proposed solutions particular criteria required for renormalizability. We also explore the field of lattice perturbation theory in the context of deep inelastic scattering. In particular, we study the connection between lattice-based Euclidean parton distributions and their experiment-based Minkowskian equivalents. These parton distributions are extracted from deep inelastic scattering processes in which leptons are scattered of hadronic targets at energy levels high enough to resolve the internal hadronic structure. To synchronize experiment with theory, one must find the proper matching criteria to transform approximations of parton distributions, known as quasidistributions, as they are resolved on a computer (i.e. the lattice), back to the full parton distributions extracted from experiment. We provide the necessary background to understand context of the above and show the procedures necessary to match experiment with the lattice. We demonstrate how one must tread carefully to properly perform such a matching.

Freid, Michael↗
DOE OSTI2026

Multidimensional Distributional Neural Network Output Demonstrated in Super‐Resolution of Surface Wind Speed

Accurate quantification of uncertainty in neural network predictions remains a central challenge for scientific applications involving high-dimensional, correlated data. While existing methods capture either aleatoric or epistemic uncertainty, few offer closed-form, multidimensional distributions that preserve spatial correlation while remaining computationally tractable. In this work, we present a framework for training neural networks with a multidimensional Gaussian loss, generating a closed-form predictive distribution over outputs informed by non-identically distributed training data. Our approach captures aleatoric uncertainty by iteratively estimating the means and covariance matrices, and is demonstrated on a super-resolution example out-of-training-sample. We leverage a Fourier representation of the covariance matrix to stabilize network training and preserve spatial correlation. We introduce a novel regularization strategy—referred to as information sharing—that interpolates between image-specific and global covariance estimates, enabling convergence of the super-resolution downscaling network trained on image-specific distributional loss functions. This framework allows for efficient sampling, explicit correlation modeling, and extensions to more complex distribution families all without disrupting prediction performance. We demonstrate the method on a surface wind speed downscaling task and discuss its broader applicability to uncertainty-aware prediction in scientific models.

17 WIND ENERGY↗
DOE OSTI2025

Modeling and Characterizing the electron backscatter in a cylindrical anode-based distributed X-ray source

Upcoming advancements in computed tomography architectures warrants the investigation of new X-ray source designs and the impacts that electron backscatter can have on these designs. One such design being investigated is a distributed, cylindrical anode-based X-ray source. For such a distributed X-ray source, we developed a modeling pipeline for simulating electron optics and transport to characterize the quality of the primary X-ray beam and the electron backscatter behavior. We report our results on the energy distributions of the bremsstrahlung spectra; electron backscatter ratio; and spatial, temporal, and energy distributions of backscattered electrons that return to the anode.

46 - INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AN↗
DOE OSTI2022

Stable parallel training of Wasserstein conditional generative adversarial neural networks

In this work, we propose a stable, parallel approach to train Wasserstein conditional generative adversarial neural networks (W-CGANs) under the constraint of a fixed computational budget. Differently from previous distributed GANs training techniques, our approach avoids inter-process communications, reduces the risk of mode collapse and enhances scalability by using multiple generators, each one of them concurrently trained on a single data label. The use of the Wasserstein metric also reduces the risk of cycling by stabilizing the training of each generator. We illustrate the approach on the CIFAR10, CIFAR100, and ImageNet1k datasets, three standard benchmark image datasets, maintaining the original resolution of the images for each dataset. Performance is assessed in terms of scalability and final accuracy within a limited fixed computational time and computational resources. To measure accuracy, we use the inception score, the Fréchet inception distance, and image quality. An improvement in inception score and Fréchet inception distance is shown in comparison to previous results obtained by performing the parallel approach on deep convolutional conditional generative adversarial neural networks as well as an improvement of image quality of the new images created by the GANs approach. Weak scaling is attained on both datasets using up to 2000 NVIDIA V100 GPUs on the OLCF supercomputer Summit.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Effects of anisotropy and disorder on the superconducting properties of niobium

We report results for the superconducting transition temperature and anisotropic energy gap for pure niobium based on Eliashberg’s equations and electron and phonon band structures computed from density functional theory. The electronic band structure is used to construct the Fermi surface and calculate the Fermi velocity at each point on the Fermi surface. The phonon bands are in excellent agreement with inelastic neutron scattering data. The corresponding phonon density of states and electron–phonon coupling define the electron–phonon spectral function, α 2 F(p, p'; ω), and the corresponding electron–phonon pairing interaction, which is the basis for computing the superconducting properties. The electron–phonon spectral function is in good agreement with existing tunneling spectroscopy data except for the spectral weight of the longitudinal phonon peak at ℏω LO = 23 meV. We obtain an electron–phonon coupling constant of λ = 1.057, renormalized Coulomb interaction μ* = 0.218, and transition temperature T c = 9.33 K. The corresponding strong-coupling gap at T = 0 is modestly enhanced, Δ 0 = 1.55 meV, compared to the weak-coupling BCS value $Δ^{\text {wc}}_{0}$ = 1.78 k B T c = 1.43 meV. The superconducting gap function exhibits substantial anisotropy on the Fermi surface. We analyze the distribution of gap anisotropy and compute the suppression of the superconducting transition temperature using a self-consistent T-matrix theory for quasiparticle-impurity scattering to describe niobium doped with non-magnetic impurities. We compare these results with experimental results on niobium SRF cavities doped with nitrogen impurities.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2019

A Game-Theoretic Data-Driven Approach for Pseudo-Measurement Generation in Distribution System State Estimation

In this paper, we present an efficient computational framework with the purpose of generating weighted pseudo-measurements to improve the quality of distribution system state estimation (DSSE) and provide observability with advanced metering infrastructure (AMI) against unobservable customers and missing data. The proposed technique is based on a game-theoretic expansion of relevance vector machines (RVMs). This platform is able to estimate the nodal power consumption and quantify its uncertainty while reducing the prohibitive computational burden of model training for large AMI datasets. To achieve this objective, the large training set is decomposed and distributed among multiple parallel learning entities. The resulting estimations from the parallel RVMs are then combined using a game-theoretic model based on the idea of repeated games with vector payoff. It is observed that through this approach and by exploiting the seasonal changes in customers’ behavior the accuracy of pseudo-measurements can be considerably improved, while introducing robustness against bad training data samples. The proposed pseudo-measurement generation model is integrated into a DSSE using a closed-loop information system, which takes advantage of a branch current state estimator (BCSE) to further improve the performance of the designed machine learning framework. This method has been tested on a practical distribution feeder model with smart meter data for verification.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2023

Temporally continuous thermofluidic–thermomechanical modeling framework for metal additive manufacturing

Additive manufacturing (AM) is known to generate large magnitudes of residual stresses (RS) within builds due to steep and localized thermal gradients. In the current state of commercial AM technology, manufacturers generally perform heat treatments in effort to reduce the generated RS and its detrimental effects on part distortion and in-service failure. Computational models that effectively simulate the deposition process can provide valuable insights to improve RS distributions. Accordingly, it is common to employ Computational fluid dynamics (CFD) models or finite element (FE) models. While CFD can predict geometric and thermal-fluid behavior, it cannot predict the structural response (e.g., stress–strain) behavior. On the other hand, an FE model can predict mechanical behavior, but it lacks the ability to predict geometric and fluid behavior. Thus, an effectively integrated thermofluidic–thermomechanical modeling framework that exploits the benefits of both techniques while avoiding their respective limitations can offer valuable predictive capability for AM processes. In contrast to previously published efforts, the work herein describes a one-way coupled CFD-FEA framework that abandons major simplifying assumptions, such as geometric steady-state conditions, the absence of material plasticity, and the lack of detailed RS evolution/accumulation during deposition, as well as insufficient validation of results. Here, the presented framework is demonstrated for a directed energy deposition (DED) process, and experiments are performed to validate the predicted geometry and RS profile. Both single- and double-layer stainless steel 316L builds are considered. Geometric data is acquired via 3D optical surface scans and X-ray micro-computed tomography, and residual stress is measured using neutron diffraction (ND). Comparisons between the simulations and measurements reveal that the described CFD-FEA framework is effective in capturing the coupled thermomechanical and thermofluidic behaviors of the DED process. The methodology presented is extensible to other metal AM processes, including power bed fusion and wire-feed-based AM.

42 ENGINEERING↗
DOE OSTI2023

Double-sided electron energy analyzer for measurement of non-Maxwellian electron energy distributions

Here, a double-sided electron energy analyzer is developed for studies of magnetic reconnection. It can measure electron energy distribution functions along two directions opposite to each other at the same time. Each side is composed of a floating reference grid, an energy selector grid, and a collector plate. The voltage of the selector grid is swept from –40 to 0 V with respect to the reference grid with a frequency of 1 MHz. This fast sweeping is required to resolve sub-Alfvénic changes in plasma quantities of the Magnetic Reconnection Experiment, where the typical Alfvénic time is a few microseconds. The reliability of the energy analyzer is checked in Maxwellian plasmas away from the reconnection region. In this case, the electron temperature computed from the electron energy distribution function agrees with measurements of a reference triple Langmuir probe. When it is located near the reconnection region, the temperatures of the tail electron population from both sides, facing into the electron flow and facing away from it, exceed the bulk electron temperature measured by the Langmuir probe by a factor of about 2.

47 OTHER INSTRUMENTATION↗
DOE OSTI2020

Radiation Mapping for an Unmanned Aerial Vehicle: Development and Simulated Testing of Algorithms for Source Mapping and Navigation Path Generation

Image reconstruction algorithms were developed for radiation source mapping and used for generating the search path of a moving radiation detector, such as one onboard an unmanned aerial vehicle. Simulations consisted of first assuming radioactive sources of varying complexity and estimating the radiation fields that would then be produced by that source distribution. Next, the "measurements" that would result from a pair of adjacent spatial locations were computed. A crude estimate of the source distribution likely to have produced such "measurements" was reconstructed based upon the limited measurements. Location of the next "measurement" was then determined as halfway between the location of the estimated source and the current "measurement." With each additional sample, improved source distribution reconstructions were made and used to inform the immediate direction of detector motion. Source reconstruction or mapping was formulated as an inverse problem solved with either maximum a posteriori or least squares (LS) regression deconvolution methods. Different amounts of noise were added to the simulated "measurements," allowing evaluation of the methods' performances as functions of signal-to-noise ratio of the measured map. As expected, methods that promote sparsity were better suited in reconstructing point sources. Reliable prior information of the source distribution also improved the reconstruction results, especially with distributed sources. With a non-negative least square algorithm and the suggested paths it generated, location of sources was successfully estimated to an accuracy of 0.014 m within nine iterations in a single-source scenario and 12 iterations in a two-source scenario, given a 10% error on the integrated counts and a Poisson distribution of the noise associated with the measured counts.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗
DOE OSTI2017

Optimizing Microgrid Energy Delivery Under High Uncertainty

One of the largest transitions in the power system today is the shift to a more sustainable and resilient power system. This is being driven by public opinion, changes in regulatory policies, and advancements in smart grid technologies. The most noticeable changes taking place is the integration of distributed energy sources (DERs); this study uses the term DER in the most general way as a resource that can be manipulated to alter energy delivery and flow in the transmission and distribution networks. Also, here it is preferred to focus on energy as the true need while power is a function of the equipment rating. As such, wind and solar, demand that can be manipulated, electric vehicles, electric energy storage, thermal storage, and storage in water system are all considered DERs. These additions to the distribution system are evolving the operation of distribution feeders into microgrids- communication, computing, and control-enabled resources that produce, transport, and utilize energy in a manner that provides cost, reliability, and resilience benefits. As this evolution progresses, the planning and operational management (scheduling and control) must explicitly include the consideration of risk. The management of system risk is currently in the purview of the utility and will likely remain so in the future. However, as each microgrid, as well as federation of microgrids, sees autonomy in order to provide maximum benefits to their constituents, they must assume responsibility to manage their internal risk. The primary scope of this study is the scheduling of resources in a distribution feeder(s) operating as microgrids. The study explores a distribution algorithm to develop the transactive schedule for the DERs, to minimize cost and risk over a time horizon, and an initial laboratory-scale to conduct implementation on distributed hardware. Results from case studies are presented that show that solutions derived by the distributed algorithm are valid. This study also discusses the continuing work on the expansion of: 1) the distributed algorithm from a deterministic to stochastic optimization formulation, and 2) implementation of the distributed algorithm into real-time simulation within the Power System laboratory at New Mexico State University (NMSU) and expanding to the Southwest Technology Development Institute located at NMSU where actual solar, energy storage, and demand response resources are installed.

24 POWER TRANSMISSION AND DISTRIBUTION↗