Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Large scale”

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 1,747 records · Page 97
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Dynamics-based halo model for large scale structure

Accurate modeling of the one-to-two halo transition has long been difficult to achieve. Here, we demonstrate that physically motivated halo definitions that respect the bimodal phase-space distribution of dark matter particles near halos resolves this difficulty. Specifically, the two phase-space components are overlapping and correspond to (1) particles orbiting the halo and (2) particles infalling into the halo for the first time. Motivated by this decomposition, García et al. [Mon. Not. R. Astron. Soc. 521, 2464 (2023)] advocated for defining halos as the collection of particles orbiting their self-generated potential. This definition identifies the traditional one-halo term of the halo-mass correlation function with the distribution of orbiting particles around a halo, while the two-halo term governs the distribution of infalling particles. We use dark matter simulations to demonstrate that the distribution of orbiting particles is finite and can be characterized by a single physical scale 𝑟 h , which we refer to as the halo radius. The two-halo term is described using a simple yet accurate empirical model based on the Zel’dovich correlation function. We further demonstrate that the halo radius imprints itself on the distribution of infalling particles at small scales. Our final model for the halo-mass correlation function is accurate at the ≈ 2% level for 𝑟∈ [0.1, 50] ℎ −1 Mpc. The Fourier transform of our best-fit model describes the halo-mass power spectrum with comparable accuracy for 𝑘 ∈ [0.06, 6.0] ℎ Mpc −1 .

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2026

Large-scale calculations of 𝛽-decay rates and implications for 𝑟-process nucleosynthesis

Nuclear 𝛽 decay is a key element of the astrophysical rapid neutron capture process (𝑟 process). In this work, we present state-of-the-art global 𝛽-decay calculations based on the quantified relativistic nuclear energy density functional theory and the deformed proton-neutron quasiparticle random-phase approximation. Our analysis considers contributions from allowed and first-forbidden transitions. We used two point-coupling functionals with carefully calibrated time-odd terms and isoscalar pairing strength. The new calculations display consistent results for both employed functionals, especially near the neutron drip line, suggesting slower 𝛽 decays past the 𝑁=126 neutron shell closure than in commonly used 𝛽-decay models. The new rates, along with the existing rates based on the recent nonrelativistic global calculations, are found to slow down the synthesis of heavy elements in the 𝑟 process and significantly reduce the contribution of neutron-induced fission.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2026

Understanding Large-Scale Dynamos in Unstratified Rotating Shear Flows

For this work, we combine simulations with new analyses that overcome previous pitfalls to explicate how nonhelical mean-field dynamos grow and saturate in unstratified, magnetorotationally driven turbulence. Shear of the mean radial magnetic field amplifies the azimuthal component. Radial fields are regenerated by velocity fluctuations that induce shear of radial magnetic fluctuations, followed by Lorentz and Coriolis forces that source a negative off-diagonal component in the turbulent diffusivity tensor. We present a simple schematic to illustrate this dynamo growth. A different part of the Lorentz force forms a third-order correlator in the mean electromotive force that saturates the dynamo.

accretion disk & black-hole plasma↗
DOE OSTI2020

Modular Isolated Soft-Switching Medium Voltage String Inverter for Large-Scale PV Farm

This paper presents a single-stage modular isolated soft-switching medium voltage string inverter (MVSI) with tri-port configuration to interconnect PV and storage to medium voltage (MV) grid. The modular MVSI provides advanced functionalities such as energy storage elements integration without additional converters, connection to MV grid resulting in lower distribution losses in the PV farm, high-frequency galvanic isolation, independent control of PV, storage and grid power, high modularity and scalability. The topology of the MVSI, operation principle and model-predictive priority-switching (MPPS) control to achieve fast dynamic balancing across stacked modules are discussed. The independent power control is validated in PLECS simulation and the operation of stacked MVSI with MPPS control is verified through Hardware-in-the-loop (HIL) simulation. Here, the hardware prototype of 25 kVA MVSI building block is developed. 18 kW experimental results of single building block verifying tri-ports operation and 20 kW experimental results of two modules stacked operation are presented. A clean sinusoidal waveform with negligibly low THD in both tests is obtained. A low dv/dt of <; 1kV/us and soft-switching are achieved.

30 DIRECT ENERGY CONVERSION↗
DOE OSTI2023

A Deep Learning Pipeline for Optimizing Large-scale Phase Field Simulations

Phase field (PF) simulations are computationally expensive but remain a key analysis tool to understand the complex mechanisms of additive manufacturing (AM) processes. Each PF simulation-aided analysis requires thousands of node hours on leadership-class supercomputers. One of the main goals of these analyses is the study of microstructure evolution during the build process which begins with the onset of nucleation. Nucleation occurs under certain thermomechanical conditions which are not known a priori and many PF simulations are required to identify ranges of input thermo-mechanical parameters that can result in the onset of nucleation. Since many of the simulations do not result in nucleation, an analysis campaign often ends up wasting tremendous amounts of precious computing resources executing nucleation-absent simulations. The goal of this work is to design and train deep learning models to inform a PF simulation about the likelihood of the occurrence of nucleation in a future simulation time-step based on the state summary over a finite number of past time-steps of a running simulation. If the prediction determines that the running simulation is unlikely to reach nucleation in the allotted time, then its execution is stopped immediately ultimately resulting in vast reduction in wasted computations when accrued over all the PF simulations typically performed in a single or multiple analysis campaign(s). The paper presents the performance of a machine learning pipeline that uses a convolutional neural network (CNN) model to learn an embedding which is then used with a self-attention network to build a multi-task deep learning model to predict the likelihood of nucleation. The model also predicts the input parameters used in a simulation. Performance is compared with a baseline pipeline that uses an off-the-shelf LeNet-5 model to learn the initial embedding. Despite their smaller size, performance results indicate significant improvement in accuracy of the proposed models compared to the larger baseline models.

Kannan, Ramakrishnan {ramki}↗
DOE OSTI2023

Fidelity Analysis of Complex Large-Scale Simulation Models of PEs (MARS) in Future Power Grids

With the proliferation of power electronic (PE)-based resources in the electric grid, there have been numerous different kinds of simulation models developed and studied to understand their dynamic behavior. Study of PE-based resources using each simulation model has its own distinct advantages and disadvantages. Therefore it is very important to carefully select the appropriate model for the evaluation of the PE-based resources. In this paper, one such effort is made to analyze the different types of models in the case of a multi-port autonomous reconfigurable solar power plant (MARS). The MARS integrates a hybrid photovoltaic (PV)-energy storage system (ESS) plant to high-voltage direct current (HVdc) system and transmission alternating current (ac) grid. The dynamic models are evaluated under different test cases varying the short-circuit ratio (SCR) and control algorithms. Recommendations are provided on the equipment study needs based on the analysis completed.

Marthi, Phani Ratna Vanamali↗
DOE OSTI2023

Building the I (Interoperability) of FAIR for performance reproducibility of large-scale composable workflows in RECUP

Abstract-Scientific computing communities increasingly run their experiments using complex data- and compute-intensive workflows that utilize distributed and heterogeneous architectures targeting numerical simulations and machine learning, often executed on the Department of Energy Leadership Computing Facilities (LCFs). We argue that a principled, systematic approach to implementing FAIR principles at scale, including fine-grained metadata extraction and organization, can help with the numerous challenges to performance reproducibility posed by such workflows. We extract workflow patterns, propose a set of tools to manage the entire life cycle of performance metadata, and aggregate them in an HPC-ready framework for reproducibility (RECUP). We describe the challenges in making these tools interoperable, preliminary work, and lessons learned from this experiment.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Charaterization of Emerging Computing Architectures for Dynamic Simulation of Future Power Grids with Large-Scale Power Electronics

The increasing penetration of power electronics in power grids significantly raises the computing requirements in a real-time (and/or fast) simulation of the power grid. The real-time simulation is an enabler for evaluating controllers, protection systems, new equipment, and twinning. In this paper, emerging computing architectures such as tensor processing units (TPU), neural/neuromorphic processing units (NPU), and quantum processing units (QPU) are introduced and characterized for the real-time (and/or fast) simulation of power electronics-dominated power grids. The metrics and the process to characterize emerging computing architectures to perform real-time (and/or fast) simulations of future power grids with power electronics are discussed. Three of the emerging computing units are characterized based on these metrics and the process developed. This characterization will enable identification and comparison of emerging computing architectures that can perform real-time (and/or fast) simulation of future power grids.

Choi, Jongchan↗