Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “network acceleration”

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 235 records · Page 13
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

AURORA: Automated Refinement of Coarse-Grained Reconfigurable Accelerators

Coarse-grained reconfigurable arrays (CGRAs), loosely defined as arrays of functional units interconnected through a network-on-chip (NoC), provide higher flexibility than domain-specific ASIC accelerators while offering increased hardware efficiency with respect to fine-grained reconfigurable devices, such as Field Programmable Gate Arrays (FPGAs). Un-fortunately, designing a CGRA for a specific application domain involves enormous software/hardware engineering effort (e.g., designing the CGRA, map operations onto the CGRA, etc) and requires the exploration on a large design space (e.g., applying appropriate loop transformation on each application, specializing the reconfigurable processing elements of the CGRA, refining the network topology, deciding the size of the data memory, etc). Int his paper, we propose AURORA – a software/hardware co-design framework to automatically synthesize optimal CGRA given a set of applications of interest

Tan, Cheng↗
DOE OSTI2021

Machine-learning accelerated geometry optimization in molecular simulation

Geometry optimization is an important part of both computational materials and surface science because it is the path to finding ground state atomic structures and reaction pathways. These properties are used in the estimation of thermodynamic and kinetic properties of molecular and crystal structures. This process is slow at the quantum level of theory because it involves an iterative calculation of forces using quantum chemical codes such as density functional theory (DFT), which are computationally expensive and which limit the speed of the optimization algorithms. It would be highly advantageous to accelerate this process because then one could do either the same amount of work in less time or more work in the same time. Here, we provide a neural network (NN) ensemble based active learning method to accelerate the local geometry optimization for multiple configurations simultaneously. We illustrate the acceleration on several case studies including bare metal surfaces, surfaces with adsorbates, and nudged elastic band for two reactions. In all cases, the accelerated method requires fewer DFT calculations than the standard method. In addition, we provide an Atomic Simulation Environment (ASE)-optimizer Python package to make the usage of the NN ensemble active learning for geometry optimization easier.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Self-Consistent Theory for Structural Relaxation, Dynamic Bond Exchange Times, and the Glass Transition in Polymeric Vitrimers

Here, we formulate a statistical mechanical theory for how dynamic bond exchange influences the activated hopping-driven relaxation of Kuhn segments in dynamically cross-linked networks or vitrimers over a wide range of temperatures and cross-link densities. The key new methodological aspect is to address in a self-consistent manner the dynamic consequences of bond exchange on the Kuhn segmental alpha relaxation, and vice versa. The predicted temperature dependence of the segmental alpha time of vitrimers at high temperatures remains the same as that of permanent networks, but at lower temperatures, a significant acceleration of relaxation occurs due to bond exchanges. From a mechanistic perspective, the vitrimer local cage barrier is very weakly affected by bond exchange, while the collective elastic barrier contribution decreases significantly in the deeply supercooled regime. The vitrimer glass transition temperature is predicted to grow linearly with the square root of the cross-link density, as previously found for permanent networks. Material-specific chemical effects such as cross-linker size relative to that of the normal Kuhn segment or special attraction of a cross-linker with polymers are crudely considered based on model calculations. The theory is quantitatively applied to recent experiments on dry ethylene vitrimers. Good agreements are found including that the bond exchange time follows an Arrhenius law at high enough temperatures but upward non-Arrhenius deviations emerge in the deeply supercooled regime, and a collapsed master curve of the segment alpha time exists over a wide range of cross-link densities and temperatures. Possible extensions to treat dynamic heterogeneity effects and penetrant transport in vitrimers are briefly discussed.

chemical calculations↗
DOE OSTI2022

High-throughput predictions of metal–organic framework electronic properties: theoretical challenges, graph neural networks, and data exploration

Abstract With the goal of accelerating the design and discovery of metal–organic frameworks (MOFs) for electronic, optoelectronic, and energy storage applications, we present a dataset of predicted electronic structure properties for thousands of MOFs carried out using multiple density functional approximations. Compared to more accurate hybrid functionals, we find that the widely used PBE generalized gradient approximation (GGA) functional severely underpredicts MOF band gaps in a largely systematic manner for semi-conductors and insulators without magnetic character. However, an even larger and less predictable disparity in the band gap prediction is present for MOFs with open-shell 3 d transition metal cations. With regards to partial atomic charges, we find that different density functional approximations predict similar charges overall, although hybrid functionals tend to shift electron density away from the metal centers and onto the ligand environments compared to the GGA point of reference. Much more significant differences in partial atomic charges are observed when comparing different charge partitioning schemes. We conclude by using the dataset of computed MOF properties to train machine-learning models that can rapidly predict MOF band gaps for all four density functional approximations considered in this work, paving the way for future high-throughput screening studies. To encourage exploration and reuse of the theoretical calculations presented in this work, the curated data is made publicly available via an interactive and user-friendly web application on the Materials Project.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Accelerating the Inference of the Exa.TrkX Pipeline

Recently, graph neural networks (GNNs) have been successfully used for a variety of particle reconstruction problems in high energy physics, including particle tracking. The Exa.TrkX pipeline based on GNNs demonstrated promising performance in reconstructing particle tracks in dense environments. It includes five discrete steps: data encoding, graph building, edge filtering, GNN, and track labeling. All steps were written in Python and run on both GPUs and CPUs. In this work, we accelerate the Python implementation of the pipeline through customized and commercial GPU-enabled software libraries, and develop a C++ implementation for inferencing the pipeline. The implementation features an improved, CUDA-enabled fixed-radius nearest neighbor search for graph building and a weakly connected component graph algorithm for track labeling. GNNs and other trained deep learning models are converted to ONNX and inferenced via the ONNX Runtime C++ API. The complete C++ implementation of the pipeline allows integration with existing tracking software. We report the memory usage and average event latency tracking performance of our implementation applied to the TrackML benchmark dataset.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2024

Inverse Design of Photonic Surfaces via High throughput Femtosecond Laser Processing and Tandem Neural Networks

Abstract This work demonstrates a method to design photonic surfaces by combining femtosecond laser processing with the inverse design capabilities of tandem neural networks that directly link laser fabrication parameters to their resulting textured substrate optical properties. High throughput fabrication and characterization platforms are developed that generate a dataset comprising 35280 unique microtextured surfaces on stainless steel with corresponding measured spectral emissivities. The trained model utilizes the nonlinear one‐to‐many mapping between spectral emissivity and laser parameters. Consequently, it generates predominantly novel designs, which reproduce the full range of spectral emissivities (average root‐mean‐squared‐error < 2.5%) using only a compact region of laser parameter space 25 times smaller than what is represented in the training data. Finally, the inverse design model is experimentally validated on a thermophotovoltaic emitter design application. By synergizing laser‐matter interactions with neural network capabilities, the approach offers insights into accelerating the discovery of photonic surfaces, advancing energy harvesting technologies.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Modeling laser-driven ion acceleration with deep learning

Developments in machine learning promise to ameliorate some of the challenges of modeling complex physical systems through neural-network-based surrogate models. High-intensity, short-pulse lasers can be used to accelerate ions to mega-electronvolt energies, but to model such interactions requires computationally expensive techniques such as particle-in-cell simulations. Multilayer neural networks allow one to take a relatively sparse ensemble of simulations and generate a surrogate model that can be used to rapidly search the parameter space of interest. In this study, we created an ensemble of over 1,000 simulations modeling laser-driven ion acceleration and developed a surrogate to study the resulting parameter space. A neural-network-based approach allows for rapid feature discovery not possible for traditional parameter scans given the computational cost. A notable observation made during this study was the dependence of ion energy on the pre-plasma gradient length scale. While this methodology harbors great promise for ion acceleration, it has ready application to all topics in which large-scale parameter scans are restricted by significant computational cost or relatively large, but sparse, domains.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

GenomeFace v1.0

GenomeFace is meta-genome binning software. Metagenomic binning, the process of grouping DNA sequences into taxonomic units, is critical for understanding the functions, interactions, and evolutionary dynamics of microbial communities. We propose a deep learning approach to binning using two neural networks, one based on composition and another on environmental abundance, dynamically weighting the contribution of each based on characteristics of the input data. Trained on over 43,000 prokaryotic genomes, our network for composition-based binning is inspired by metric learning techniques used for facial recognition. Using a task-specific, multi-GPU accelerated algorithm to cluster the embeddings produced by our network, our binner leverages marker genes observed to be universally present in nearly all taxa to grade and select optimal clusters of sequences from a hierarchy of candidates. We evaluate our approach on four simulated datasets with known ground truth. Our linear time integration of marker genes recovers more near complete genomes than state of the art but computationally infeasible solutions using them, while being over an order of magnitude faster. Finally, we demonstrate the scalability and acuity of our approach by testing it on three of the largest metagenome assemblies ever performed. Compared to other binners, we produced 47%-183% more near complete genomes. From these datasets, we find over the genomes of over 3000 new candidate species which have never been previously cataloged, representing a potential 4% expansion of the known bacterial tree of life.

Lettich, Richard [Lawrence Berkeley National Labor↗
DOE OSTI2021

Cyber-Physical Events Emulation Based Transmission and Distribution Co-Simulation for Situation Awareness and Grid Anomaly (SAGA) Detection: Preprint

Energy management of transmission and distribution networks is becoming more challenged with the accelerated increasing of distributed energy resources (DERs) such as distributed photovoltaic (PV) generation and distributed energy storage. To better analyze the impacts of DERs on both transmission and distribution systems, a comprehensive transmission and distribution co-simulation platform should be developed. Furthermore, with DERs more actively participated in system operation such as providing real time grid services, their cyber vulnerability should be better understood to maintain system reliability. This paper discussed a cyber-physical events emulation based transmission & distribution co-simulation platform to perform different cyber events emulation and analyze the impacts of cyber physical events happened in distribution system on the T&D system operation. The case studies with both a transmission network and a synthetic distribution network data validate that the proposed T&D co-simulation platform can perform comprehensive cyber physical events emulation. Therefore, with extensive simulation using the proposed model, the system operator can accumulate adequate training data for the system situation awareness and grid anomaly detection purpose.

31 CESER - Office of Cybersecurity, Energy Securit↗
NASA NTRS2005

Adaptive Control for Microgravity Vibration Isolation System

Most active vibration isolation systems that try to a provide quiescent acceleration environment for space science experiments have utilized linear design methods. In this paper, we address adaptive control augmentation of an existing classical controller that employs a high-gain acceleration feedback together with a low-gain position feedback to center the isolated platform. The control design feature includes parametric and dynamic uncertainties because the hardware of the isolation system is built as a payload-level isolator, and the acceleration Sensor exhibits a significant bias. A neural network is incorporated to adaptively compensate for the system uncertainties, and a high-pass filter is introduced to mitigate the effect of the measurement bias. Simulations show that the adaptive control improves the performance of the existing acceleration controller and keep the level of the isolated platform deviation to that of the existing control system.

Yang, Bong-Jun↗
DOE OSTI2024

IRIS-GNN: Leveraging Graph Neural Networks for Scheduling on Truly Heterogeneous Runtime Systems

The diversity of accelerators in computer systems poses significant challenges for software developers, such as managing vendor-specific compiler toolchains, code fragmentation requiring different kernel implementations, and performance portability issues. To address these, the Intelligent Runtime System (IRIS) was developed. IRIS works across various systems, from smartphones to supercomputers, enabling automatic performance scaling based on available accelerators. It introduces abstract tasks for seamless execution transitions between accelerators while ensuring memory consistency and task dependencies. Although IRIS simplifies system details, optimal dynamic scheduling still requires user input to understand workload structures. To address this, we introduce a new scheduling policy for IRIS, termed IRIS-GNN, which is the first IRIS hybrid policy that operates in conjunction with the dynamic policies. This policy employs a Graph-Neural Network (GNN) to conduct Graph Classification of any task graphs submitted to IRIS. This GNN analyzes the structure and attributes of the task graph, categorizing it as either locality, concurrency, or mixed. This classification subsequently guides the selection of the dynamic policy used by IRIS. We provide a comparison of the performance of IRIS-GNN against the complete spectrum of IRIS’s dynamic policies, assess the overhead introduced by the GNN within this scheduling framework, and ultimately explore its practical application in real-world scenarios.

Johnston, Beau↗
DOE OSTI2024

Using the Metropolis algorithm to explore the loss surface of a recurrent neural network

In the limit of small trial moves the Metropolis Monte Carlo algorithm is equivalent to gradient descent on the energy function in the presence of Gaussian white noise. This observation was originally used to demonstrate a correspondence between Metropolis Monte Carlo moves of model molecules and overdamped Langevin dynamics, but it also applies in the context of training a neural network: making small random changes to the weights of a neural network, accepted with the Metropolis probability, with the loss function playing the role of energy, has the same effect as training by explicit gradient descent in the presence of Gaussian white noise. We explore this correspondence in the context of a simple recurrent neural network. We also explore regimes in which this correspondence breaks down, where the gradient of the loss function becomes very large or small. In these regimes the Metropolis algorithm can still effect training, and so can be used as a probe of the loss function of a neural network in regimes in which gradient descent struggles. We also show that training can be accelerated by making purposely-designed Monte Carlo trial moves of neural-network weights.

Casert, Corneel↗
NASA NTRS1998

GPS Monitoring of Surface Change During and Following the Fortuitous Occurrence of the M(sub w) = 7.3 Landers Earthquake in our Network

Accomplishments: (1) Continues GPS monitoring of surface change during and following the fortuitous occurrence of the M(sub w) = 7.3 Landers earthquake in our network, in order to characterize earthquake dynamics and accelerated activity of related faults as far as 100's of kilometers along strike. (2) Integrates the geodetic constraints into consistent kinematic descriptions of the deformation field that can in turn be used to characterize the processes that drive geodynamics, including seismic cycle dynamics. In 1991, we installed and occupied a high precision GPS geodetic network to measure transform-related deformation that is partitioned from the Pacific - North America plate boundary northeastward through the Mojave Desert, via the Eastern California shear zone to the Walker Lane. The onset of the M(sub w) = 7.3 June 28, 1992, Landers, California, earthquake sequence within this network poses unique opportunities for continued monitoring of regional surface deformation related to the culmination of a major seismic cycle, characterization of the dynamic behavior of continental lithosphere during the seismic sequence, and post-seismic transient deformation. During the last year, we have reprocessed all three previous epochs for which JPL fiducial free point positioning products available and are queued for the remaining needed products, completed two field campaigns monitoring approx. 20 sites (October 1995 and September 1996), begun modeling by development of a finite element mesh based on network station locations, and developed manuscripts dealing with both the Landers-related transient deformation at the latitude of Lone Pine and the velocity field of the whole experiment. We are currently deploying a 1997 observation campaign (June 1997). We use GPS geodetic studies to characterize deformation in the Mojave Desert region and related structural domains to the north, and geophysical modeling of lithospheric behavior. The modeling is constrained by our existing and continued GPS measurements, which will provide much needed data on far-field strain accumulation across the region and on the deformational response of continental lithosphere during and following a large earthquake, forming the basis for kinematic and dynamic modeling of secular and seismic-cycle deformation. GPS geodesy affords both regional coverage and high precision that uniquely bear on these problems.

Miller, M. Meghan↗
DOE OSTI2026

RU-net for automatic characterization of TRISO fuel cross sections

During irradiation, phenomena such as kernel swelling and buffer densification may impact the performance of tristructural isotropic (TRISO) particle fuel. Post-irradiation microscopy is often used to identify these irradiation-induced morphologic changes. However, each fuel compact generally contains thousands of TRISO particles. Manually performing the work to get statistical information on these phenomena is cumbersome and subjective. Here, to reduce the subjectivity inherent in that process and to accelerate data analysis, we used convolutional neural networks (CNNs) to automatically segment cross-sectional images of microscopic TRISO layers. CNNs are a class of machine-learning algorithms specifically designed for processing structured grid data. They have gained popularity in recent years due to their remarkable performance in various computer vision tasks, including image classification, object detection, and image segmentation. In this research, we generated a large irradiated TRISO layer dataset with more than 2,000 microscopic images of cross-sectional TRISO particles and the corresponding annotated images. Based on these annotated images, we used different CNNs to automatically segment different TRISO layers. These CNNs include RU-Net (developed in this study), as well as three existing architectures: U-Net, Residual Network (ResNet), and Attention U-Net. The preliminary results show that the model based on RU-Net performs best in terms of Intersection over Union (IoU). Using CNN models, we can expedite the analysis of TRISO particle cross sections, significantly reducing the manual labor involved and improving the objectivity of the segmentation results.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

Multi-modality deep learning for pulse prediction in homogeneous nonlinear systems via parametric conversion

In this Letter, we introduce FusionNet, a multi-modality deep learning framework designed to predict and analyze output pulses in high-power rare-earth-doped laser systems driving parametric conversion in homogeneous guided nonlinear media. FusionNet integrates temporal, spectral, and physical experimental conditions to model ultrafast nonlinear phenomena, including parametric nonlinear frequency conversion, self-phase modulation, and cross-phase modulation in homogeneous guided systems such as gas-filled hollow-core fibers. These systems bridge physical models with experimental data, advancing our understanding of light-guiding principles and nonlinear interactions while expediting the design and optimization of on-demand high-power, high-brightness systems. Our results demonstrate a 73% reduction in prediction error and an 83% improvement in computational efficiency compared to conventional neural networks. This work establishes a new paradigm for accelerating parametric simulations and optimizing experimental designs in high-power laser systems, with further implications for high-precision spectroscopy, quantum information science, and distributed entangled interconnects.

47 OTHER INSTRUMENTATION↗
DOE OSTI2023

Multifold Insights for Power System Dynamics From Data Assimilation: Meeting Current Challenges

The world’s electric power grids have evolved over the last 120 years from a single power line to today’s large networks. The evolution will continue at an accelerated rate with extensive smart grid development worldwide. Here for example, the U.S. government has set a goal of reaching 100 percent carbon pollution-free electricity by 2035, while the Department of Energy (DOE) shared a goal to deploy 30 GW and 110 GW of offshore wind by 2030 and 2050, respectively. To meet such ambitious goals, in the years to come, a significant percentage of electricity will come from intermittent renewable sources, electric vehicles (EVs), and be supplied to a vast number of loads that will actively respond to grid conditions and incentive signals. This development is largely driven by environmental and economic factors, such as reducing carbon emissions and saving electricity cost for consumers. These energy resources lead to new uncertain behaviors and dynamics which the grid has never seen and were not considered in its design. Operating such a dynamic grid with sufficient reliability and efficiency is a monumental challenge.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

Neural Phase Simulation

The Neural Phase Simulation (NPS) is a package of codes for simulating microstructure evolution and accelerated molecular dynamics with deep neural-networks based surrogate models. NPS is designed to offer quantitatively accurate and computationally efficient simulation capabilities by leveraging modern machine-learning techniques. The primary intended use cases of NPS are training neural network surrogate models, though performing simulations on a single node is also supported. The NPS surrogate models can be trained from ground truth simulation methods, which are supposed to be accurate but expensive, such as molecular dynamics, phase field methods, kinetic Monte Carlo and discrete dislocation dynamics.

Zhou, Fei↗
DOE OSTI2021

Enhanced Signal Processing of Distributed Brillouin Fiber Sensors using a Decoupled Radial Basis Function Network

A novel decoupled radial basis function network (D-RBFN) is proposed to accelerate signal processing and address the big data challenges associated with ultra-long distance Brillouin optical time-domain analysis (BOTDA) systems. The proposed frame- work is demonstrated on a dataset measured over a 100 km distance using a bi-directional Raman assisted BOTDA system.

Venketeswaran, Abhishek↗