Search NASA⌕ Search

SEARCH · Search NASA

Results for “edge computing”

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 1,099 records · Page 61
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Compiling Quantum Circuits for Dynamically Field-Programmable Neutral Atoms Array Processors

Dynamically field-programmable qubit arrays (DPQA) have recently emerged as a promising platform for quantum information processing. In DPQA, atomic qubits are selectively loaded into arrays of optical traps that can be reconfigured during the computation itself. Leveraging qubit transport and parallel, entangling quantum operations, different pairs of qubits, even those initially far away, can be entangled at different stages of the quantum program execution. Such reconfigurability and non-local connectivity present new challenges for compilation, especially in the layout synthesis step which places and routes the qubits and schedules the gates. In this paper, we consider a DPQA architecture that contains multiple arrays and supports 2D array movements, representing cutting-edge experimental platforms. Within this architecture, we discretize the state space and formulate layout synthesis as a satisfiability modulo theories problem, which can be solved by existing solvers optimally in terms of circuit depth. For a set of benchmark circuits generated by random graphs with complex connectivities, our compiler OLSQ-DPQA reduces the number of two-qubit entangling gates on small problem instances by 1.7x compared to optimal compilation results on a fixed planar architecture. To further improve scalability and practicality of the method, we introduce a greedy heuristic inspired by the iterative peeling approach in classical integrated circuit routing. Using a hybrid approach that combined the greedy and optimal methods, we demonstrate that our DPQA-based compiled circuits feature reduced scaling overhead compared to a grid fixed architecture, resulting in 5.1X less two-qubit gates for 90 qubit quantum circuits. These methods enable programmable, complex quantum circuits with neutral atom quantum computers, as well as informing both future compilers and future hardware choices.

Physics↗
DOE OSTI2024

Multi-Level Structural Damage Characterization Using Sparse Acoustic Sensor Networks and Knowledge Transferred Deep Learning

Standard structural health monitoring techniques face well-known difficulties for comprehensive defect diagnosis in real-world structures that have structural, material, or geometric complexity. This motivates the exploration of machine-learning-based structural health monitoring methods in complex structures. However, creating sufficient training data sets with various defects is an ongoing challenge for data-driven machine (deep) learning algorithms. The ability to transfer the knowledge of a trained neural network from one component to another or to other sections of the same component would drastically reduce the required training data set. Also, it would facilitate computationally inexpensive machine learning based inspection systems. In this work, a machine-learning-based multi-level damage characterization is demonstrated with the ability to transfer trained knowledge within the sparse sensor network. A novel network spatial assistance and an adaptive convolution technique are proposed for efficient knowledge transfer within the deep learning algorithm. Proposed structural health monitoring method is experimentally evaluated on an aluminum plate with artificially induced defects. It was observed that the method improves the performance of knowledge transferred damage characterization by 50% during localization and 24% during severity assessment. Further, experiments using time windows with and without multiple edge reflections are studied. Results reveal that multiply scattered waves contain rich and deterministic defect signatures that can be mined using deep learning neural networks, improving the accuracy of both identification and quantification. In the case of a fixed sensor network, using multiply scattered waves shows 100% prediction accuracy at all levels of damage characterization.

36 MATERIALS SCIENCE↗
DOE OSTI2018

Energy release rate approximation for edge cracks using higher-order topological derivatives

Topological derivatives provide the variation of a functional when an infinitesimal hole is introduced into the domain. In Silva et al. (J Mech Phys Solids 59(5):925–939, 2011), the authors developed a first-order approximation of the energy release rate associated with a small crack at any boundary location and at any orientation using a first-order topological derivative. This approach offers significant computational advantages over other methods because (i) it requires only a single analysis while other methods require an analysis for each crack size-location-direction combination, and (ii) it is performed on the non-cracked domain, removing the need to create very fine meshes in the vicinity of the crack. In the present study, higher-precision approximations of the energy release rate are developed using higher-order topological derivatives which allow the analyst to accurately treat longer cracks and determine the crack lengths for which the first-order approximation is accurate.

42 ENGINEERING↗
DOE OSTI2021

Equation-of-Motion Coupled-Cluster Cumulant Green’s Function for Excited States and X-ray Spectra

Green’s function methods within many-body perturbation theory provide a general framework for treating electronic correlations in excited states and spectra. Conventional methods using the Dyson equation or the cumulant expansion are typically based on the GW self-energy approximation. In order to extend this approximation in molecular systems, a non-perturbative real-time coupled-cluster cumulant Green’s function approach has been introduced, where the cumulant is obtained as the solution to a set of coupled first order, non-linear differential equations. This approach naturally includes non-linear corrections to conventional cumulant Green’s function techniques where the cumulant is linear in the GW self-energy. The method yields the spectral function associated with the core Green’s function, which is directly related to the x-ray photoemission spectra (XPS) of molecular systems. The approach also yields very good results for binding energies and satellite excitations. The x-ray absorption spectrum (XAS) is then obtained as a convolution of the core spectral function and an effective one-body absorption spectrum. Here this approach is extended to include the full coupled-cluster-singles (CCS) core Green’s function by including the complete form of the non-linear contributions to the cumulant as well as all single, double, and triple cluster excitations in the CC amplitude equations. This approach naturally builds in orthogonality and shake-up effects analogous to those in the Mahan-Noizeres-de Dominicis edge singularity corrections that enhance the XAS near the edge. Themethod is illustrated for the XPS and XAS of NH 3 .

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2019

A Numerically Exact Approach to Quantum Impurity Problems in Realistic Lattice Geometries

We review a controlled numerical approach to quantum impurity problems in realistic geometries, consisting of exactly mapping the complete lattice Hamiltonian onto an equivalent one dimensional system through a unitary transformation. The resulting dimensional and entanglement reduction allows one to study the quantum many-body problem on arbitrary d-dimensional lattices using the density matrix renormalization group (DMRG) method. The real-space resolution allows one to position the impurities at the boundary or bulk of the sample and to study screening effects due to edge or surface modes. We describe how to generalize this approach to multi-impurity problems, discuss applications and possible extensions.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Physics-informed heterogeneous graph neural networks for DC blocker placement

The threat of geomagnetic disturbances (GMDs) to the reliable operation of the bulk energy system has spurred the development of effective strategies for mitigating their impacts. One such approach involves placing transformer neutral blocking devices, which interrupt the path of geomagnetically induced currents (GICs) to limit their impact. The high cost of these devices and the sparsity of transformers that experience high GICs during GMD events, however, calls for a sparse placement strategy that involves high computational cost. To address this challenge, we developed a physics-informed heterogeneous graph neural network (PIHGNN) for solving the graph-based dc-blocker placement problem. Our approach combines a heterogeneous graph neural network (HGNN) with a physics-informed neural network (PINN) to capture the diverse types of nodes and edges in ac/dc networks and incorporates the physical laws of the power grid. We train the PIHGNN model using a surrogate power flow model and validate it using case studies. Results demonstrate that PIHGNN can effectively and efficiently support the deployment of GIC dc-current blockers, ensuring the continued supply of electricity to meet societal demands. Furthermore, our approach has the potential to contribute to the development of more reliable and resilient power grids capable of withstanding the growing threat that GMDs pose.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2020

Charge carrier dynamics in two-dimensional hybrid perovskites: Dion–Jacobson vs. Ruddlesden–Popper phases

Two-dimensional (2D) halide perovskites are promising materials for environmentally stable next-generation optoelectronic device applications. Besides the widely investigated Ruddlesden–Popper (RP) phases, Dion–Jacobson (DJ) phases are attracting considerable attention due to their rapid emergence as efficient solar cell materials. However, there is very little atomistic understanding of the charge carrier dynamics under ambient conditions for these DJ-phases, limiting the possibilities to tune their optoelectronic performances through compositional engineering routes. In this study, by combining nonadiabatic molecular dynamics with time-domain density functional theory methods at room temperature, we compare the dominant non-radiative carrier recombination and dephasing processes in RP and DJ monolayered lead halide perovskites. Our systematic study demonstrates that performance-limiting nonradiative carrier recombination processes greatly depends on the electron–phonon interactions induced by structural fluctuations and instantaneous charge localization in these materials. The stiffer interlayer packing due to the presence of single spacer dications, which separates the lead iodide slabs, reduces the thermal fluctuations in the DJ-phase to a greater extent than that in the RP-phase 2D-perovskites. Specific electronic coupling between the closely spaced lead iodide layers enhances the delocalization of band-edge charge densities in DJ-phase systems. Compared to the RP-phase, reduced inelastic electron–phonon scattering in DJ-phase perovskites significantly limits intrinsic non-radiative recombination processes. The consequent enhancement in the photogenerated charge carrier lifetime makes DJ-phase perovskites potentially suitable for various optoelectronic devices. The computational insights gained from this study allow us to outline a set of robust design principles for DJ-phase perovskites to strategically tune their optoelectronic properties.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Development of five-moment two-fluid modeling for Z-pinch physics

The Z-pinch m = 0 instability as well as its stabilization by radially sheared axial flow is studied using the nonlinear ideal five-moment twofluid (5M2F) model with an extension of that model to include Braginskii heat and momentum transport. Using the ideal 5M2F model, linear growth rate results are compared with prior work using MHD and Hall MHD. At small normalized wavenumber, 1 < k a < 4 , where a is the effective pinch radius, 5M2F results agree with Hall MHD within 20% in scenarios without radially sheared axial flow. With the sheared flow and focusing on ka = 10/3, agreement with Hall MHD is excellent. In the limit of small ion inertial length, results also match with MHD. A comparison with PIC modeling of shear-free m = 0 stability focuses on a plasma scenario based on recent experimental results. In a scan of mode wavenumber, ideal 5M2F results are qualitatively similar to PIC: the growth rate rises to a peak at a moderate wavenumber and declines at a large wavenumber in contrast to MHD results, which show the saturation of the growth rate with the increasing wavenumber rather than a decline. The peak normalized 5M2F growth rate is &#x3B3; &#x3C4; A = 1.5 , where sA is the Alfven transit time across the pinch. The peak occurs at normalized wavenumber ka = 10. For comparison, PIC results have a peak growth of &#x3B3; &#x3C4; A = 0.77 at ka = 5. Including Braginskiibased closure of the 5M2F model does not qualitatively change the ideal results in this particular case. Nonlinear saturation is studied using the 5M2F model with the dissipative Braginskii-based closure in cases with pinch-edge sheared-flow speed equal to half the Alfven speed. Nonlinear mixing due to the sheared flow yields a quasi-steady state after modest losses of pinch ion inventory and pinch thermal energy, approximately 30% and 10%, respectively. 5M2F modeling captures the essential physics of m = 0 instability and offers a computationally tractable route to high-fidelity modeling of 3D Z-pinch behavior, including m = 1 instability.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2018

A practical computational method for the anisotropic redshift-space three-point correlation function

Here, we present an algorithm enabling computation of the anisotropic redshift-space galaxy threepoint correlation function ( 3 PCF) scaling as N 2 , with N the number of galaxies. Our previous work showed how to compute the isotropic 3 PCF with this scaling by expanding the radially binned density field around each galaxy in the survey into spherical harmonics and combining these coefficients to form multipole moments. The N 2 scaling occurred because this approach never explicitly required the relative angle between a galaxy pair about the primary galaxy. Here, we generalize this work, demonstrating that in the presence of azimuthally symmetric anisotropy produced by redshift-space distortions (RSD), the 3 PCF can be described by two triangle side lengths, two independent total angular momenta, and a spin. This basis for the anisotropic 3 PCF allows its computation with negligible additional work over the isotropic 3 PCF. We also present the covariance matrix of the anisotropic 3 PCF measured in this basis. Our algorithm tracks the full 5 D redshift-space 3 PCF, uses an accurate line of sight to each triplet, is exact in angle, and easily handles edge correction. It will enable use of the anisotropic large-scale 3 PCF as a probe of RSD in current and upcoming large-scale redshift surveys.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2018

High-Entropy Alloys (HEA): Fundamentals and Applications

This book provides a systematic and comprehensive description of high-entropy alloys (HEAs). The authors summarize key properties of HEAs from the perspective of both fundamental understanding and applications, which are supported by in-depth analyses. The book also contains computational modeling in tackling HEAs, which help elucidate the formation mechanisms and properties of HEAs from various length and time scales.<br>High-entropy alloys (HEAs) [1,2] loosely refer to multi-principal-element solid solution alloys due to their high configurational entropy, in contrast to traditional alloys, which focus on the edge or corner of phase diagrams with one principal component. The unique compositions and the resulting attractive properties of HEAs have stimulated growing research interest due to scientific curiosity and potential industrial applications [3,4]. In order to balance the properties for targeted applications, the microstructure of HEAs can be a single phase or multiphases, and traditional physical metallurgy principles have been applied to develop a variety of HEAs, including high-entropy stainless steels, high-entropy superalloys, high-entropy refractory alloys, high-entropy light-weight alloys, high-entropy oxides, high-entropy metallic compounds, etc. However, great challenges still remain in the fundamental understanding of HEAs formation and their properties, and potential high-performance HEAs are yet to be explored. This special issue is meant to collect contributions from authors working in the various fields of HEAs to timely disseminate the rapid progress in this fascinating and expanding class of advanced materials.

Gao, Michael↗
DOE OSTI2021

ANTS

The ANTS code (Alternate Non-Linear Two-phase Solver) is based on a novel non-linear solution algorithm for the solution of the two-phase, subchannel fluid equations. It achieves its performance through decoupling of the two-phase momentum equations (axial and transverse) from the axial phasic mass and energy equations which allows for a nested non-linear iteration scheme. This enables a plane-by-plane solution which the inner iteration focuses on a reduced non-linear equation set for the primitives in phasic mass flow rate, enthalpy and void for each node edge. Single node edges are coupled as part of the outer iteration via surface mass fluxes which appear as source terms in the inner iteration scheme. The outer iteration readily accommodates two-phase flow phenomena closure relationships for subchannel mixing and void drift. A primary feature is the use of a non-staggered mesh computational mesh and steady-state iterative solver in contrast to all existing subchannel codes.

Kropaczek, David J↗
DOE OSTI2016

Strengthening LLNL Missions through Laboratory Directed Research and Development in High Performance Computing

High performance computing (HPC) has been a defining strength of Lawrence Livermore National Laboratory (LLNL) since its founding. Livermore scientists have designed and used some of the world’s most powerful computers to drive breakthroughs in nearly every mission area. Today, the Laboratory is recognized as a world leader in the application of HPC to complex science, technology, and engineering challenges. Most importantly, HPC has been integral to the National Nuclear Security Administration’s (NNSA’s) Stockpile Stewardship Program—designed to ensure the safety, security, and reliability of our nuclear deterrent without nuclear testing. A critical factor behind Lawrence Livermore’s preeminence in HPC is the ongoing investments made by the Laboratory Directed Research and Development (LDRD) Program in cutting-edge concepts to enable efficient utilization of these powerful machines. Congress established the LDRD Program in 1991 to maintain the technical vitality of the Department of Energy (DOE) national laboratories. Since then, LDRD has been, and continues to be, an essential tool for exploring anticipated needs that lie beyond the planning horizon of our programs and for attracting the next generation of talented visionaries. Through LDRD, Livermore researchers can examine future challenges, propose and explore innovative solutions, and deliver creative approaches to support our missions. The present scientific and technical strengths of the Laboratory are, in large part, a product of past LDRD investments in HPC. Here, we provide seven examples of LDRD projects from the past decade that have played a critical role in building LLNL’s HPC, computer science, mathematics, and data science research capabilities, and describe how they have impacted LLNL’s mission.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

A High-Throughput Solver for Marginalized Graph Kernels on GPU

Here, we present the design and optimization of a solver for efficient and high-throughput computation of the marginalized graph kernel on General Purpose GPUs. The graph kernel is computed using the conjugate gradient method to solve a generalized Laplacian of the tensor product between a pair of graphs. To cope with the large gap between the instruction throughput and the memory bandwidth of the GPUs, our solver forms the graph tensor product on-the-fly without storing it in memory. This is achieved by using threads in a warp cooperatively to stream the adjacency and edge label matrices of individual graphs by small square matrix blocks called tiles, which are then staged in registers and the shared memory for later reuse. Warps across a thread block can further share tiles via the shared memory to increase data reuse. We exploit the sparsity of the graphs hierarchically by storing only non-empty tiles using a coordinate format and nonzero elements within each tile using bitmaps. We propose a new partition-based reordering algorithm for aggregating nonzero elements of the graphs into fewer but denser tiles to further exploit sparsity. We carry out extensive theoretical analyses on the graph tensor product primitives for tiles of various density and evaluate their performance on synthetic and real-world datasets. Our solver delivers three to four orders of magnitude speedup over existing CPU-based solvers such as GraKeL and GraphKernels. The capability of the solver enables kernel-based learning tasks at unprecedented scales.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Spin Solar Cell Phenomenon on a Single-Molecule Magnet (SMM) Impacted CoFeB-Based Magnetic Tunnel Junctions

The single-molecule magnet (SMM) is demonstrated here to transform conventional magnetic tunnel junctions (MTJs), a memory device used in present-day computers, into solar cells. For the first time, we demonstrated an electronic spin-dependent solar cell effect on an SMM-transformed MTJ under illumination from unpolarized white light. We patterned cross-junction-shaped devices to form a CoFeB/MgO/CoFeB-based MTJ. The MgO barrier thickness at the intersection between the two exposed junction edges was less than the SMM extent, which enabled the SMM molecules to serve as channels to conduct spin-dependent transport. The SMM channels yielded a region of long-range magnetic ordering around these engineered molecular junctions. Our SMM possessed a hexanuclear [Mn 6 (μ 3 -O) 2 (H 2 N-sao) 6 (6-atha) 2 (EtOH) 6 ] [H 2 N-saoH = salicylamidoxime, 6-atha = 6-acetylthiohexanoate] complex and thioesters end groups to form bonds with metal films. SMM-doped MTJs were shown to exhibit a solar cell effect and yielded ≈80 mV open-circuit voltage and ≈10 mA/cm 2 saturation current density under illumination from one sun equivalent radiation dose. A room-temperature Kelvin Probe AFM (KPAFM) study provided direct evidence that the SMM transformed the electronic properties of the MTJ’s electrodes over a lateral area in excess of several thousand times larger in extent than the area spanned by the molecular junctions themselves. As a result, the decisive factor in observing this spin-photovoltaic effect is the formation of SMM spin channels between the two different ferromagnetic electrodes, which in turn is able to catalyze the long-range transformation in each electrode around the junction area.

14 SOLAR ENERGY↗
DOE OSTI2018

An electron Bernstein wave heating scheme for the Proto-MPEX linear device

Linear plasma devices, such as the Prototype Material Plasma Exposure eXperiment (Proto-MPEX), are an economic method to study plasma-material interactions under high heat and particle fluxes. The Proto-MPEX device at Oak Ridge National Laboratory is a high-density helicon plasma generator with additional resonant electron heating to study plasma-material interactions in ITER-like conditions. Here, we present modeling results of an electron Bernstein wave (EBW) heating scheme for this device using a microwave-based system. A 28 GHz injection system is available to provide power levels up to 100 kW injected into low electron temperature and high electron density plasmas. A modified version of the GENRAY ray-tracing code, GENRAY-C, has been used to determine the EBW and electron cyclotron heating wave accessibility for these overdense plasmas. Modeling has shown that greater than 80% of mode conversion from the ordinary mode to the EBW is possible for 28 GHz injection in Proto-MPEX. Calculations show that by utilizing the effects of Doppler-shifted resonance absorption, power deposition near the core is possible. Finally, significant collisional damping, leading to edge absorption, is expected to occur for neutral pressures >1.0 mTorr; however, experiments can be designed to minimize these effects.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2024

Development of a neural network model for peeling–ballooning stability analysis in the KSTAR tokamak pedestals

The neural network model, MISHKA-NN is developed to mitigate the computational burden associated with the linear ideal magnetohydrodynamic (MHD) stability analysis of the pedestal based on the peeling–ballooning (P–B) model. By utilizing both 1D plasma profiles (current density, pressure gradient, and safety factor) and 0D parameters (plasma geometry, total current, and toroidal mode number), the model predicts linear growth rate of edge-localized ideal MHD instability in a given equilibrium state. By enabling the prediction of each instability within a second, the model reduces the time required for plotting a pedestal P–B stability diagram (j - α diagram) from approximately 100 CPU hours to a few CPU minutes. Notably, even with the utilization of parametric pressure and current profiles and plasma boundary shapes for the training dataset, the model shows a satisfactory level of performance in benchmarking the j - α diagram for the reconstructed equilibrium from a KSTAR tokamak experiment. We anticipate the model to serve as a versatile alternative to 2D linear MHD stability codes, alleviating numerical costs.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2019

Evolving Energy Efficient Convolutional Neural Networks

As deep neural networks have been deployed in more and more applications over the past half decade and are finding their way into an ever increasing number of operational systems, their energy consumption becomes a concern whether running in the datacenter or on edge devices. Hyperparameter optimization and automated network design for deep learning is a quickly growing field, but much of the focus has remained only on optimizing for the performance of the machine learning task. In this work, we demonstrate that the best performing networks created through this automated network design process have radically different computational characteristics (e.g. energy usage, model size, inference time), presenting the opportunity to utilize this optimization process to make deep learning networks more energy efficient and deployable to smaller devices. Optimizing for these computational characteristics is critical as the number of applications of deep learning continues to expand.

Young, Steven↗