Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Computational network”

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 1,657 records · Page 92
All sourcesNASA NTRSDOE OSTI
NASA NTRSDate not supplied

Cislunar Autonomous Positioning System Technology Operations and Navigation Experiment (CAPSTONE) Pathfinder for Artemis Gateway

The Cislunar Autonomous Positioning System Technology Operations and Navigation Experiment (CAPSTONE) mission was developed by NASA in collaboration with Advanced Space, LLC of Westminster, Colorado. This technology demonstration mission serves as a pathfinder for near rectilinear halo orbit (NHRO) operations around the Moon. The NHRO, (Perilune = 3,200 km; Apolune = 70,000 km) is the intended orbit for NASA’s Artemis Gateway, a small, human-tended space station planned for lunar orbit. The CAPSTONE mission will validate simulations and confirm operational planning for Gateway while also validating performance of navigation and stationkeeping requirements for Gateway’s Power and Propulsion Element. Therefore, this mission will provide operational experience to NASA, commercial, and international missions for operations in a demanding orbital regime. The CAPSTONE mission consists of a 12-unit (U)+ CubeSat developed, integrated, and tested by the Terran Orbital Corporation that carries a payload communications system capable of crosslink ranging with NASA’s Lunar Reconnaissance Orbiter (LRO). CAPSTONE contains a chip-scale atomic clock (CSAC) for a one-way ranging experiment with NASA’s Deep Space Network, a dedicated payload flight computer for software demonstration, and a camera. The launch, coordinated by NASA’s Launch Services Program, was provided by Rocket Lab on its Electron launch vehicle using their Photon upper stage to deploy the CAPSTONE spacecraft. The mission launched June 28, 2022. The CAPSTONE spacecraft deployed from the Photon stage and traversed an approximately 4-month, highly fuel-efficient transfer phase entering the NRHO November 13, 2022, for a six-month primary mission phase. The mission is currently in a twelve-month technology enhancement operations phase. The CAPSTONE technology demonstration mission is led by Advanced Space, LLC. Spacecraft development and mission operations are conducted by Terran Orbital Corporation of Irvine, California. Noted accomplishments for the CAPSTONE mission include demonstrating the accessibility of NHROs; validating key operational concepts in the NHRO environment; laying the foundation for commercial support of future lunar operations; and accelerating the availability of peer-to-peer navigation capabilities provided by the Cislunar Autonomous Positioning System (CAPS). The CAPSTONE mission is funded through NASA’s Small Spacecraft Technology (SST) program, which is one of several programs within NASA’s Space Technology Mission Directorate. The program is chartered to develop and demonstrate technologies to enhance and expand the capabilities of small spacecraft with a particular focus on enabling new mission architectures through the use of small spacecraft, expanding the reach of small spacecraft to new destinations, and augmenting future missions with supporting small spacecraft. The CAPSTONE mission launch was provided by NASA’s Exploration Systems Development Missions Directorate’s Advanced Exploration Systems Division. Coordination and acquisition of the launch was managed by NASA’s Launch Services Program. The CAPSTONE mission and project status will be presented.

Elwood Agasid↗
NASA NTRSDate not supplied

Implementation of a Hybrid Edge Node-Centroid Node Approach for the Generation of Reduced Thermal Models

Reduced thermal models are often required for delivery to organizations that manage observatory or launch models at the highest levels of assembly. However, the effort to generate reduced models, and verify against their detailed counterparts, is a challenge that has not yet been conclusively solved. Higher level organizations often place a limit on the number of nodes for delivered models with the assumption that smaller models generally result in less computation time. However, the burden of producing and verifying the accuracy of the reduced models is placed primarily on the lower-level organizations, which in turn consumes resources needed to produce these models. Limiting the total number of allowable nodes may also prevent users from taking full advantage of software capabilities that allow for faster generation of models, such as finite elements, which generally require more nodes than older centroid based models. A methodology using Thermal Desktop was described in 2010 which used: (1) finite elements and edge nodes for a conduction matrix, (2) centroid nodes for capacitance and radiative computations, and (3) the super network feature to produce a conduction matrix based only on the centroid nodes. At that time, the methodology was clear, but the implementation would have had to be done manually; however, with the inclusion of the OpenTD API, this methodology can now be implemented programmatically and for the first time, be a viable approach for the generation of reduced models. The approach was implemented and developed at the NASA Goddard Space Flight Center (GSFC) for the Capture, Containment, and Return System (CCRS) payload on the Earth Return Orbiter (ERO) as part of the Mars Sample Return (MSR) mission, resulting in the TCYEE tool. ERO features a spacecraft bus provided by Airbus through the European Space Agency (ESA) with node limitations on the delivered CCRS model provided by GSFC. TCYEE was used to generate the reduced model for this delivery and the predictions compared favorably to the detailed model currently in use for the thermal performance evaluation. Furthermore, TCYEE is being explored for potential use on the Roman Space Telescope (RST) for the generation of reduced models for delivery to the launch provider, which also has node limit requirements on the RST observatory model for use in launch simulation analyses. This paper describes the methodology, its implementation, and compares the performance of reduced models generated by TCYEE to their detailed counterparts.

Thermal desktop↗
DOE OSTI2020

Wide neural networks with bottlenecks are deep Gaussian processes

There has recently been much work on the "wide limit" of neural networks, where Bayesian neural networks (BNNs) are shown to converge to a Gaussian process (GP) as all hidden layers are sent to infinite width. However, these results do not apply to architectures that require one or more of the hidden layers to remain narrow. In this paper, we consider the wide limit of BNNs where some hidden layers, called "bottlenecks", are held at finite width. The result is a composition of GPs that we term a "bottleneck neural network Gaussian process" (bottleneck NNGP). Although intuitive, the subtlety of the proof is in showing that the wide limit of a composition of networks is in fact the composition of the limiting GPs. We also analyze theoretically a single-bottleneck NNGP, finding that the bottleneck induces dependence between the outputs of a multi-output network that persists through extreme post-bottleneck depths, and prevents the kernel of the network from losing discriminative power at extreme post-bottleneck depths.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Solving inverse problems in stochastic models using deep neural networks and adversarial training

Inverse problems associated with stochastic models constitute a significant portion of scientific and engineering applications. In such cases the unknown quantities are distributions. The applicability of traditional methods is limited because of their demanding assumptions or prohibitive computational consumption; for example, maximum likelihood methods require closed-form density functions, and Markov Chain Monte Carlo needs a large number of simulations. We propose a new method that estimates the unknown distribution by matching the statistical properties between observed and simulated random processes. We leverage the expressive power of neural networks to approximate the unknown distribution and use a discriminative neural network for computing the statistical discrepancies between the observed and simulated random processes. Here we demonstrated numerically that the proposed methods can estimate both the model parameters and learn complicated unknown distributions.

42 ENGINEERING↗
DOE OSTI2023

Encoding integers and rationals on neuromorphic computers using virtual neuron

Abstract Neuromorphic computers emulate the human brain while being extremely power efficient for computing tasks. In fact, they are poised to be critical for energy-efficient computing in the future. Neuromorphic computers are primarily used in spiking neural network–based machine learning applications. However, they are known to be Turing-complete, and in theory can perform all general-purpose computation. One of the biggest bottlenecks in realizing general-purpose computations on neuromorphic computers today is the inability to efficiently encode data on the neuromorphic computers. To fully realize the potential of neuromorphic computers for energy-efficient general-purpose computing, efficient mechanisms must be devised for encoding numbers. Current encoding mechanisms (e.g., binning, rate-based encoding, and time-based encoding) have limited applicability and are not suited for general-purpose computation. In this paper, we present the virtual neuron abstraction as a mechanism for encoding and adding integers and rational numbers by using spiking neural network primitives. We evaluate the performance of the virtual neuron on physical and simulated neuromorphic hardware. We estimate that the virtual neuron could perform an addition operation using just 23 nJ of energy on average with a mixed-signal, memristor-based neuromorphic processor. We also demonstrate the utility of the virtual neuron by using it in some of the μ -recursive functions, which are the building blocks of general-purpose computation.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Analysis and mitigation of parasitic resistance effects for analog in-memory neural network acceleration

To support the increasing demands for efficient deep neural network processing, accelerators based on analog in-memory computation of matrix multiplication have recently gained significant attention for reducing the energy of neural network inference. However, analog processing within memory arrays must contend with the issue of parasitic voltage drops across the metal interconnects, which distort the results of the computation and limit the array size. This work analyzes how parasitic resistance affects the end-to-end inference accuracy of state-of-the-art convolutional neural networks, and comprehensively studies how various design decisions at the device, circuit, architecture, and algorithm levels affect the system's sensitivity to parasitic resistance effects. Here, a set of guidelines are provided for how to design analog accelerator hardware that is intrinsically robust to parasitic resistance, without any explicit compensation or re-training of the network parameters.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

A survey of techniques for optimizing transformer inference

Recent years have seen a phenomenal rise in the performance and applications of transformer neural networks. The family of transformer networks, including Bidirectional Encoder Representations from Transformer (BERT), Generative Pretrained Transformer (GPT) and Vision Transformer (ViT), have shown their effectiveness across Natural Language Processing (NLP) and Computer Vision (CV) domains. Transformer-based networks such as ChatGPT have impacted the lives of common men. However, the quest for high predictive performance has led to an exponential increase in transformers' memory and compute footprint. Researchers have proposed techniques to optimize transformer inference at all levels of abstraction. Further, this paper presents a comprehensive survey of techniques for optimizing the inference phase of transformer networks. We survey techniques such as knowledge distillation, pruning, quantization, neural architecture search and lightweight network design at the algorithmic level. We further review hardware-level optimization techniques and the design of novel hardware accelerators for transformers. We summarize the quantitative results on the number of parameters/FLOPs and the accuracy of several models/techniques to showcase the tradeoff exercised by them. We also outline future directions in this rapidly evolving field of research. We believe that this survey will educate both novice and seasoned researchers and also spark a plethora of research efforts in this field.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Detecting hidden layers from spreading dynamics on complex networks

When dealing with spreading processes on networks it can be of the utmost importance to test the reliability of data and identify potential unobserved spreading paths. In this paper we address these problems and propose methods for hidden layer identification and reconstruction. We also explore the interplay between difficulty of the task and the structure of the multilayer network describing the whole system where the spreading process occurs. Our methods stem from an exact expression for the likelihood of a cascade in the susceptible-infected model on an arbitrary graph. We then show that by imploring statistical properties of unimodal distributions and simple heuristics describing joint likelihood of a series of cascades one can obtain an estimate of both existence of a hidden layer and its content with success rates far exceeding those of a null model. Furthermore, we conduct our analyses on both synthetic and real-world networks providing evidence for the viability of the approach presented.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Deep Nonparametric Estimation of Operators between Infinite Dimensional Spaces

Learning operators between infinitely dimensional spaces is an important learning task arising in machine learning, imaging science, mathematical modeling and simulations, etc. This paper studies the nonparametric estimation of Lipschitz operators using deep neural networks. Non-asymptotic upper bounds are derived for the generalization error of the empirical risk minimizer over a properly chosen network class. Under the assumption that the target operator exhibits a low dimensional structure, our error bounds decay as the training sample size increases, with an attractive fast rate depending on the intrinsic dimension in our estimation. Our assumptions cover most scenarios in real applications and our results give rise to fast rates by exploiting low dimensional structures of data in operator estimation. We also investigate the influence of network structures (e.g., network width, depth, and sparsity) on the generalization error of the neural network estimator and propose a general suggestion on the choice of network structures to maximize the learning efficiency quantitatively.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1995

Parallel processing for scientific computations

The scope of this project dealt with the investigation of the requirements to support distributed computing of scientific computations over a cluster of cooperative workstations. Various experiments on computations for the solution of simultaneous linear equations were performed in the early phase of the project to gain experience in the general nature and requirements of scientific applications. A specification of a distributed integrated computing environment, DICE, based on a distributed shared memory communication paradigm has been developed and evaluated. The distributed shared memory model facilitates porting existing parallel algorithms that have been designed for shared memory multiprocessor systems to the new environment. The potential of this new environment is to provide supercomputing capability through the utilization of the aggregate power of workstations cooperating in a cluster interconnected via a local area network. Workstations, generally, do not have the computing power to tackle complex scientific applications, making them primarily useful for visualization, data reduction, and filtering as far as complex scientific applications are concerned. There is a tremendous amount of computing power that is left unused in a network of workstations. Very often a workstation is simply sitting idle on a desk. A set of tools can be developed to take advantage of this potential computing power to create a platform suitable for large scientific computations. The integration of several workstations into a logical cluster of distributed, cooperative, computing stations presents an alternative to shared memory multiprocessor systems. In this project we designed and evaluated such a system.

Alkhatib, Hasan S.↗
DOE OSTI2024

GraphCH: A Deep Framework for Assessing Cyber-Human Aspects in Insider Threat Detection

Insider threat is one of the most damaging cyber attacks that could cause the loss of intellectual property and enterprise data security breaches. Action sequence data such as host logs are used to investigate such threats and develop anomaly-based AI detectors. However, insider threat actions are similar to legitimate user activities, causing AI detectors to fail and suffer from high false alarm rates. Therefore, user cyber activity logs are inadequate to fully unfold insider threats. In this study, we adopt human psychological principles of risk-taking and impulsiveness along with host data to assess the influence and usefulness of human behavioral aspects in insider threat detection. Here, we hypothesize that individuals' impulsive and risk-taking behavior correlates with cyberspace activities. To validate our hypothesis, we conducted an IRB-approved study recruiting 35 participants who work in a large U.S. university and collected their cyber and psychological data for 90 days. Host and human-behavioral data analysis and mapping indicate that impulsive and risk-taking users trigger more system errors causing (un)intentional insider threats and are susceptible to attackers' social engineering and cognitive hacking. Utilizing cyber-human aspects, we introduce a Cyber-Human Graph Neural Network (GNN) based framework GraphCH to identify abnormal user behaviors and detect insider threats.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Machine learning-based analysis of COVID-19 pandemic impact on US research networks

Here in this study we explore how fallout from the changing public health policy around COVID-19 has changed how researchers access and process their science experiments. Using a combination of techniques from statistical analysis and machine learning, we conduct a retrospective analysis of historical network data for a period around the stay-at-home orders that took place in March 2020. Our analysis takes data from the entire ESnet infrastructure to explore DOE high-performance computing (HPC) resources at OLCF, ALCF, and NERSC, as well as User sites such as PNNL and JLAB. We look at detecting and quantifying changes in site activity using a combination of t-Distributed Stochastic Neighbor Embedding (t-SNE) and decision tree analysis. Our findings bring insights into the working patterns and impact on data volume movements, particularly during late-night hours and weekends.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1995

A versatile system for processing geostationary satellite data with run-time visualization capability

To better predict global climate change, scientists are developing climate models that require interdisciplinary and collaborative efforts in their building. We are currently involved in several such projects but will briefly discuss activities in support of two such complementary projects: the Atmospheric Radiation Measurement (ARM) program of the Department of Energy and Sequoia 2000, a joint venture of the University of California, the private sector, and government agencies. Our contribution to the ARM program is to investigate the role of clouds on the top of the atmosphere and on surface radiance fields through the data analysis of surface and satellite observations and complex modeling of the interaction of radiation with clouds. One of our first ARM research activities involves the computation of the broadband shortwave surface irradiance from satellite observations. Geostationary satellite images centered over the first ARM observation site are received hourly over the Internet network and processed in real time to compute hourly and daily composite shortwave irradiance fields. The images and the results are transferred via a high-speed network to the Sequoia 2000 storage facility in Berkeley, where they are archived These satellite-derived results are compared with the surface observations to evaluate the accuracy of the satellite estimate and the spatial representation of the surface observations. In developing the software involved in calculating the surface shortwave irradiance, we have produced an environment whereby we can easily modify and monitor the data processing as required. Through the principles of modular programming, we have developed software that is easily modified as new algorithms for computation are developed or input data availability changes. In addition, the software was designed so that it could be run from an interactive, icon-driven, graphical interface, TCL-TK, developed by Sequoia 2000 participants. In this way, the data flow can be interactively assessed and altered as needed. In this environment, the intermediate data processing 'images' can be viewed, enabling the investigator to easily monitor the various data processing steps as they progress. Additionally, this environment allows the rapid testing of new processing modules and allows their effects to be visually compared with previous results.

Landsfeld, M.↗
DOE OSTI2021

Energy Material Network Data Hubs

In early 2015 the United States Department of Energy conceived of a consortium of collaborative bodies based on shared expertise, data, and resources that could be targeted towards the more difficult problems in energy materials research. The concept of virtual laboratories had been envisioned and discussed earlier in the decade in response to the advent of the Materials Genome Initiative and similar scientific thrusts. To be effective, any virtual laboratory needed a robust method for data management, communication, security, data sharing, dissemination, and demonstration to work efficiently and effectively for groups of remote researchers. With the accessibility of new, easily deployed cloud technology and software frameworks, such individual elements could be integrated, and the required collaboration architecture is now possible. The developers have leveraged open-source software frameworks, customized them, and merged them into a platform to enable collaborative energy materials science, regardless of the geographic dispersal of the people and resources. After five years in operations, the systems are demonstratively an effective platform for enabling research within the Energy Material Networks (EMN). This paper will show the design and development of a secured scientific data sharing platform, the ability to customize the system to support diverse workflows, and examples of the enabled research and results connected with some of the Energy Material Networks.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

How Many Trip Requests Could We Support? An Activity-Travel Based Vehicle Scheduling Approach

In a world of ever-changing travel behavior and ever-increasing modal options, is vital to have integrated models that could capture the interactions between supply and demand layers of travel. Addressing this need, we propose three different versions of network representation and mathematical models for the activity-based vehicle routing problem to connect activity-travel graphs of passengers (demand layer) to spatio-temporal networks of vehicles (supply layer). Versions I and II are arc-based, while version III is path-based. In version I, we introduce the concept of activity-travel graphs for passengers. For vehicles, we construct space–time networks and add a new dimension, called “under-service state”, to track the execution status of trip requests at any location and time. In version II, we reduce the complexity of the network structure by eliminating the state dimension and some other modifications in the structure of the passengers’ and vehicles’ network. Although both versions can capture various behavioral constraints of the activity-based vehicle routing problem (e.g., mandatory and optimal activities, duration of activities, chain of activities, preferred starting and ending times of activities), due to the high level of complexity of the network structure, both versions can only solve small-sized problems. To tackle the computational complexity, we propose a path-based network representation in version III, and to make a balance between the disutility of passengers and vehicles, we present a tolled user equilibrium problem. Mathematical models are coded in C and GAMS and implemented on real-world Phoenix regional transportation network with more than 39 million trip requests, which demonstrate the effectiveness of the proposed solution for the original and restricted master problems.

33 ADVANCED PROPULSION SYSTEMS↗
NASA NTRS1998

Tools for Atmospheric Radiative Transfer: Streamer and FluxNet

Two tools for the solution of radiative transfer problems are presented. Streamer is a highly flexible medium spectral resolution radiative transfer model based on the plane-parallel theory of radiative transfer. Capable of computing either fluxes or radiances, it is suitable for studying radiative processes at the surface or within the atmosphere and for the development of remote-sensing algorithms. FluxNet is a fast neural network-based implementation of Streamer for computing surface fluxes. It allows for a sophisticated treatment of radiative processes in the analysis of large data sets and potential integration into geophysical models where computational efficiency is an issue. Documentation and tools for the development of alternative versions of Fluxnet are available. Collectively, Streamer and FluxNet solve a wide variety of problems related to radiative transfer: Streamer provides the detail and sophistication needed to perform basic research on most aspects of complex radiative processes while the efficiency and simplicity of FluxNet make it ideal for operational use.

Key, Jeffrey R.↗
DOE OSTI2024

Neural network based emulation of galaxy power spectrum covariances: A reanalysis of BOSS DR12 data

We train neural networks to quickly generate redshift-space galaxy power spectrum covariances from a given parameter set (cosmology and galaxy bias). This covariance emulator utilizes a combination of traditional fully connected network layers and transformer architecture to accurately predict covariance matrices for the high redshift, north galactic cap sample of the BOSS DR12 galaxy catalog. We run simulated likelihood analyses with emulated and brute-force computed covariances, and we quantify the network’s performance via two different metrics: (1) difference in Χ 2 and (2) likelihood contours for simulated BOSS DR 12 analyses. We find that the emulator returns excellent results over a large parameter range. We then use our emulator to perform a reanalysis of the BOSS HighZ NGC galaxy power spectrum, and find that varying covariance with cosmology along with the model vector produces Ω m = $0.27⁢6$$^{+0.013}_{–0.015}$, H 0 = 70.2 ± 1.9 km/s/Mpc, and σ 8 = $0.67⁢4$$^{+0.058}_{–0.077}$. These constraints represent an average 0.46⁢σ shift in best-fit values and a 5% increase in constraining power compared to fixing the covariance matrix (Ω m = 0.293 ± 0.017, H 0 = 70.3 ± 2.0 km/s/Mpc, σ 8 = $0.70⁢2$$^{+0.063}_{–0.075}$). As a result, this work demonstrates that emulators for more complex cosmological quantities than second-order statistics can be trained over a wide parameter range at sufficiently high accuracy to be implemented in realistic likelihood analyses.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Training Spiking Neural Networks with Synaptic Plasticity under Integer Representation

Neuromorphic computing is emerging as a promising Beyond Moore computing paradigm that employs event-triggered computation and non-von Neumann hardware. Spike Timing Dependent Plasticity (STDP) is a well-known bio-inspired learning rule that relies on activities of locally connected neurons to adjust the weights of their respective synapses. In this work, we analyze a basic STDP rule and its sensitivity on the different hyperparameters for training spiking neural networks (SNNs) with supervision, customized for a neuromorphic hardware implementation with integer weights. We compare the classification performance on four UCI datasets (iris, wine, breast cancer and digits) that depict varying levels of complexity. We perform a search for optimal set of hyperparameters using both grid search and Bayesian optimization. Through the use of Bayesian optimization, we show the general trends in hyperparameter sensitivity in SNN classification problem. With the best sets of hyperparameters, we achieve accuracies comparable to some of the best performing SNNs on these four datasets. With a highly optimized supervised STDP rule we show that these accuracies can be achieved with just 20 epochs of training.

Kulkarni, Shruti↗