Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “network solutions”

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 541 records · Page 30
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

CRCNS22 Learning Rules in the Hippocampus and their Mapping to Neuromorphic Systems (Final Technical Report)

Large scale biologically-realistic computational models are key to investigating the interplay between structure and function in nervous systems, thus paving the way to new clinical methods and neuro-inspired computing solutions. This project focuses on the hippocampus, in particular the CA3-CA1 regions, due to their role in associative learning and memory, pattern separation and completion, and spatial navigation. Investigations into the neuronal organization and learning rule(s) of this circuit can shed light into how declarative memories are formed, stored, recalled and forgotten and inform computational, experimental and clinical neuroscience work. Our project aims at developing a novel data-driven methodology supported by a broad heterogeneous base of neuroscience experimental knowledge and inspired from advances in computer science and engineering. Specifically, this work will benchmark existing and new learning rules within a full-scale spiking neural network simulation of the CA3-CA1 region. The model will be based on an open-source repository, called the Hippocampome, which contains neuronal morphologies, firing patterns, synapse probabilities, and most other required parameters for all known neuron types in the rodent hippocampal formation. The model will be first trained in a supervised fashion for associative memory tasks using backpropagation through time traditionally used in computer science, enhanced with a new technique called the surrogate gradient method. This optimization method will be used to obtain a global loss minimization, but it is not biologically inspired as it assumes the use of data not locally available to the synapses. However, we propose its use as a benchmarking tool, to compare the training performance of local biologically plausible and hardware-mappable learning rules at scale. New rules or combinations will be proposed and tested as needed, based on the obtained results. Progress in this area will also drive the development of novel hardware-mappable algorithms for continual lifelong learning and categorization of new events from few presented examples. This project goes beyond the existing state-of-the-art by looking at large scale realistic neuronal circuits as networks trainable via global optimization methods such as surrogate gradient descent. The objective function of the brain that supports learning is largely unknown, but it is likely that it operates through local learning rules. Studying network trajectories around local minima as proposed in this work represents a useful strategy for understanding whether a network is training by using a specific (set of) learning rule(s). Starting from a completely untrained network is a challenging test since it is difficult to determine how the learning rule affects the trajectory of the network. This interdisciplinary project will help understand what rule governs learning in these regions or if multiple learning rules are involved. The work will develop a robust methodology to measure if the network is converging to the target solution, oscillating around it, or diverging away.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Emissions minimization on road networks via Generic Second Order Models

In this paper we consider the problem of estimating emissions due to vehicular traffic on complex networks, and minimizing their effect by regulating traffic at junctions. For the traffic evolution, we consider a Generic Second Order Model, which encompasses the majority of two-equations (i.e., second-order) models available in the literature, and extend it to road networks with merge and diverge junctions. The dynamics on the whole network is determined by selecting a solution to the Riemann Problems at junctions, i.e., the Cauchy problems with constant initial data on each incident road. The latter are solved by assuming the maximization of the flow and assigning a traffic distribution coefficient for outgoing roads of diverges, and a priority rule for incoming roads of merges. A general emission model is considered and its parameters are tuned to the $ {\mathrm{NO_{x}}} $ emission rate. The minimization of emissions is then formulated in terms of the traffic distribution and priority parameters, taking into account travel times. A comparison is provided between roundabouts with optimized parameters and traffic lights, which correspond to time-varying traffic priorities. Our approach can be adapted to manage traffic in complex networks in order to reduce emissions while keeping travel time at acceptable levels.

33 ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2021

Hybrid Machine Learning for Scanning Near-Field Optical Spectroscopy

The underlying physics behind an experimental observation often lacks a simple analytical description. This is especially the case for scanning probe microscopy techniques, where the interaction between the probe and the sample is nontrivial. Realistic modeling to include the details of the probe is always exponentially more difficult than its "spherical cow" counterparts. On the other hand, a well-trained artificial neural network based on real data can grasp the hidden correlation between the signal and sample properties. In this work, we show that, via a combination of model calculation and experimental data acquisition, a physics-infused hybrid neural network can predict the tip-sample interaction in the widely used scattering-type scanning near-field optical microscope. This hybrid network provides a long-sought solution for accurate extraction of material properties from tip-specific raw data. The methodology can be extended to other scanning probe microscopy techniques as well as other data-oriented physical problems in general.

36 MATERIALS SCIENCE↗
DOE OSTI2022

SMALE: Enhancing Scalability of Machine Learning Algorithms on Extreme-Scale Computing Platforms

Deployment and execution of machine learning tasks on extreme-scale computing platforms face several significant technical challenges: 1) High computing cost incurred by dense networks – The computing workload of deep networks with densely-connected topology increases rapidly with the network size, imposing a non-scalable computing model of extreme-scale computing platforms; 2) Non-optimized workload distribution – Many advanced deep learning algorithms, e.g., sparsification and irregular net-work topology, produce very unbalanced workload distribution on extreme-scale computing platforms. The computation efficiency is greatly hindered by the incurred data and computation redundancies as well as long tails of the node with extensive workload; 3) Constraints in data movement and I/O bottle-neck – Inter-node data movement in extreme-scale computing platforms are associated with high energy and latency costs, and subject to the constraints of I/O bandwidth; and 4) Generalization of algorithm realization and acceleration on computing platforms – The large varieties of machine learning algorithms and structures of extreme-scale computing platforms make the derivation of a generalized algorithm realization and acceleration method very challenging, which, however, is the requirement by domain scientists and interested users. We call the above challenges Smale’s Problems in Machine Learning and Understanding for High-Performance Computing Scientific Discovery. The objective of our three-year research project is to develop a holistic innovation set at structure, assembly, and acceleration layers of machine learning algorithms to address the above challenges in algorithm deployment and execution. Three tasks are particularly performed, including: At the algorithm structure level, we investigate the techniques that can structurally sparsify on the topology of deep networks for computing workload reduction. We also study clustering and pruning techniques that can optimize the workload distributions over the extreme-scale computing platforms; At the algorithm assembly level, we derive a unified learning framework for unsupervised transfer learning and dynamic growing capabilities. Novel training methods are also exploited to enhance the training efficiency of the proposed framework; At the algorithm acceleration level, we will develop a series of techniques that can accelerate the computation of sparse matrix operations, which are one of the core executions in deep learning and optimize memory access of the concerned platforms. Our proposed techniques attack the fundamental problems in machine learning algorithms running on extreme-scale computing platforms by vertically integrating the solutions at three closely entangled layers, paving the long-term scaling path of machine learning applications under DOE context. Three tasks corresponding to the above respective research orientations are performed during the three-year project period with our collaborators at ORNL. The outcome of the proposed project is anticipated to form a holistic solution set of novel algorithms and network topologies, efficient training techniques, and fast acceleration methods to promote the computing scalability of the machine learning applications of particular interest to DOE.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

A Quantum-Inspired Tensor Network Algorithm for Constrained Combinatorial Optimization Problems

Combinatorial optimization is of general interest for both theoretical study and real-world applications. Fast-developing quantum algorithms provide a different perspective on solving combinatorial optimization problems. In this paper, we propose a quantum-inspired tensor-network-based algorithm for general locally constrained combinatorial optimization problems. Our algorithm constructs a Hamiltonian for the problem of interest, effectively mapping it to a quantum problem, then encodes the constraints directly into a tensor network state and solves the optimal solution by evolving the system to the ground state of the Hamiltonian. We demonstrate our algorithm with the open-pit mining problem, which results in a quadratic asymptotic time complexity. Our numerical results show the effectiveness of this construction and potential applications in further studies for general combinatorial optimization problems.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2023

Defect Recognition for Eddy Current Testing of Spent Nuclear Fuel Canister using Convolutional Neural Network

This paper proposes an accurate and robust defect detection solution for 304L and 306L stainless steel (SS) weld. In the proposed solution, Eddy current testing (ECT) is employed to generate 2-dimensional (2D) data for samples under test with defects. The 2D data can be treated as images for deep learning-based defect detection. Since convolutional neural networks (CNNs) are powerful in processing images, CNN is employed in this study for defect detection. Experiments are conducted on a submerged arc welding (SAW) 304L SS weld sample with an artificial crack generated by waterjet cutting. The ECT data on this seeded fault sample is utilized to verify the proposed solution. For this purpose, the ECT measurement are separated as from Fault area and Normal area, which are used for CNN training. After training, the testing data is used for verification. Experimental results demonstrate the feasibility and effectiveness of the proposed solution.

Niu, Guangxing↗
DOE OSTI2026

Geologic stress modulates fluid mixing at fracture intersections

Fracture intersections are critical links that enable flow and transport in subsurface fracture networks, and their behavior strongly influences fluid mixing in a network. Although all subsurface fractures are subjected to geological stress, we lack a fundamental understanding of how fracture intersection geometry evolves under stress and how these changes influence fluid mixing. Here, we combine 3D printing, 3D X-ray tomographic imaging, and 3D pore-scale numerical simulations to reveal stress-induced changes in intersection geometry and their impact on mixing. Mixing is found to be strongly affected by partial closure of an intersection under stress. As an intersection closes, the void area for fluid flow and diffusion decreases leading to substantial deviations between conventional mixing models and full pore-scale modeling. To address this, we propose a modified mixing model that accounts for intersection deformation, which is essential for accurate modeling of solute transport and mixing through fracture networks.

15 GEOTHERMAL ENERGY↗
NASA NTRS1991

Sub-nanosecond clock synchronization and precision deep space tracking

Interferometric spacecraft tracking is accomplished at the NASA Deep Space Network (DSN) by comparing the arrival time of electromagnetic spacecraft signals to ground antennas separated by baselines on the order of 8000 km. Clock synchronization errors within and between DSN stations directly impact the attainable tracking accuracy, with a 0.3 ns error in clock synchronization resulting in an 11 nrad angular position error. This level of synchronization is currently achieved by observing a quasar which is angularly close to the spacecraft just after the spacecraft observations. By determining the differential arrival times of the random quasar signal at the stations, clock synchronization and propagation delays within the atmosphere and within the DSN stations are calibrated. Recent developments in time transfer techniques may allow medium accuracy (50-100 nrad) spacecraft observations without near-simultaneous quasar-based calibrations. Solutions are presented for a global network of GPS receivers in which the formal errors in clock offset parameters are less than 0.5 ns. Comparisons of clock rate offsets derived from GPS measurements and from very long baseline interferometry and the examination of clock closure suggest that these formal errors are a realistic measure of GPS-based clock offset precision and accuracy. Incorporating GPS-based clock synchronization measurements into a spacecraft differential ranging system would allow tracking without near-simultaneous quasar observations. The impact on individual spacecraft navigation error sources due to elimination of quasar-based calibrations is presented. System implementation, including calibration of station electronic delays, is discussed.

Charles Dunn↗
NASA NTRS1992

Subnanosecond GPS-based clock synchronization and precision deep-space tracking

Interferometric spacecraft tracking is accomplished by the Deep Space Network (DSN) by comparing the arrival time of electromagnetic spacecraft signals at ground antennas separated by baselines on the order of 8000 km. Clock synchronization errors within and between DSN stations directly impact the attainable tracking accuracy, with a 0.3-nsec error in clock synchronization resulting in an 11-nrad angular position error. This level of synchronization is currently achieved by observing a quasar which is angularly close to the spacecraft just after the spacecraft observations. By determining the differential arrival times of the random quasar signal at the stations, clock offsets and propagation delays within the atmosphere and within the DSN stations are calibrated. Recent developments in time transfer techniques may allow medium accuracy (50-100 nrad) spacecraft tracking without near-simultaneous quasar-based calibrations. Solutions are presented for a worldwide network of Global Positioning System (GPS) receivers in which the formal errors for DSN clock offset parameters are less than 0.5 nsec. Comparisons of clock rate offsets derived from GPS measurements and from very long baseline interferometry (VLBI), as well as the examination of clock closure, suggest that these formal errors are a realistic measure of GPS-based clock offset precision and accuracy. Incorporating GPS-based clock synchronization measurements into a spacecraft differential ranging system would allow tracking without near-simultaneous quasar observations. The impact on individual spacecraft navigation-error sources due to elimination of quasar-based calibrations is presented. System implementation, including calibration of station electronic delays, is discussed.

Dunn, C. E.↗
NASA NTRS1997

64x64 Analog Input Array for 3-Dimensional Neural Network Processor

In pattern recognition and classification for spatio-temporal problems, one of the most challenging tasks is to provide a good and valid solution in real-time. Because of time constraints, software-based neural network approaches may not be suitable for practical use. Hardware solutions seem to be good candidates for this class of problems. Currently, the Three Dimensional Analog Neural Network (3-DANN) is an effective approach to solving spatio-temporal problems in three-dimensional hardware.

spatio-temporal 3-DANN Three Dimensional Analog Ne↗
DOE OSTI2021

Resilience-based Recovery Scheduling of Transportation Network in Mixed Traffic Environment: A Deep-Ensemble-Assisted Active Learning Approach

Devising effective post-hazard recovery strategies is critical in enhancing the resilience of transportation networks (TNs). However, existing work does not consider the multiclass users’ travel behavior in network functionality quantification and the metaheuristic solution procedures often suffer from extensive computational burden due to the exploration need in large solution space and the expensive functionality quantification. Here, this study develops a bilevel decision-making framework for the resilience-based recovery scheduling of the TN in a mixed traffic environment with connected and autonomous vehicles (CAVs) and human-driven vehicles (HDVs). The lower level quantifies the TN's functionality over time considering different travel behavior of CAV and HDV users arisen from their different levels of information perception. The upper level presents a novel deep-ensemble-assisted active learning approach to balance optimization performance and computational cost. This framework can help decision makers better quantify the TN's functionality to support effective recovery scheduling of TN with different mixed traffic scenarios ranging from HDV-only to future CAV-dominant traffic. The optimization approach bears the potential to be extended to solving general large-scale network recovery scheduling problems effectively and efficiently. The proposed methodology is demonstrated using a real-world traffic network in Southern California under earthquake considering deterministic and stochastic repair durations.

42 ENGINEERING↗
DOE OSTI2021

Exactness of OPF Relaxation on Three-phase Radial Networks with Delta Connections

Simulations have shown that while semi-definite relaxations of AC optimal power flow (AC-OPF) on three-phase radial networks with only wye connections tend to be exact, the presence of delta connections seem to render them inexact. This paper shows that such inexactness originates from the non-uniqueness of relaxation solutions and numerical errors amplified by the non-uniqueness. This finding motivates two algorithms to recover the exact solution of AC-OPF in unbalanced distribution networks featuring both wye and delta connections. In simulations using IEEE 13, 37 and 123-bus systems, the proposed algorithms provide exact optimal solutions up to numerical precision.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2026

Make the Fastest Faster: Importance Mask Synthesis for Interactive Volume Visualization using Reconstruction Neural Networks

Visualizing a large-scale volumetric dataset with high resolution is challenging due to the substantial computational time and space complexity. Recent deep learning-based image inpainting methods significantly improve rendering latency by reconstructing a high-resolution image for visualization in constant time on GPU from a partially rendered image where only a portion of pixels go through the expensive rendering pipeline. However, existing solutions need to render every pixel of either a predefined regular sampling pattern or an irregular sample pattern predicted from a low-resolution image rendering. Both methods require a significant amount of expensive pixel-level rendering. In this work, we provide Importance Mask Learning (IML) and Synthesis (IMS) networks, which are the first attempts to directly synthesize important regions of the regular sampling pattern from the user’s view parameters, to further minimize the number of pixels to render by jointly considering the dataset, user behavior, and the downstream reconstruction neural network. Our solution is a unified framework to handle various types of inpainting methods through the proposed differentiable compaction/decompaction layers. Experiments show our method can further improve the overall rendering latency of state-of-the-art volume visualization methods using reconstruction neural network for free when rendering scientific volumetric datasets. Our method can also directly optimize the off-the-shelf pre-trained reconstruction neural networks without elongated retraining.

Large-scale data↗
DOE OSTI2021

Implementing a neural network interatomic model with performance portability for emerging exascale architectures

The two main thrusts of computational science are increasingly accurate predictions and faster calculations; to this end, the zeitgeist in molecular dynamics (MD) simulations is pursuing machine learned and data driven interatomic models, e.g. neural network potentials, and novel hardware architectures, e.g. GPUs. Current implementations of neural network potentials are orders of magnitude slower than traditional interatomic models and while looming exascale computing offers the ability to run large, accurate simulations with these models, achieving portable performance for MD with new and varied exascale hardware requires rethinking traditional algorithms, using novel data structures, and library solutions. We re-implement a neural network interatomic model in CabanaMD, an MD proxy application, built on libraries developed for performance portability. Our implementation shows significantly improved thread scaling in this complex kernel as compared to a current LAMMPS implementation, across both strong and weak scaling. Our single-source solution enables simulations up to 20 million atoms on a single CPU node and 4 million atoms with improved performance on a single GPU. Furthermore, we also explore parallelism and data layout choices (using flexible data structures called AoSoAs) and their effect on performance, seeing up to ~50% and ~5% improvements in performance on a GPU by choosing the right level of parallelism and data layout respectively.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1988

How to cluster in parallel with neural networks

Partitioning a set of N patterns in a d-dimensional metric space into K clusters - in a way that those in a given cluster are more similar to each other than the rest - is a problem of interest in astrophysics, image analysis and other fields. As there are approximately K(N)/K (factorial) possible ways of partitioning the patterns among K clusters, finding the best solution is beyond exhaustive search when N is large. Researchers show that this problem can be formulated as an optimization problem for which very good, but not necessarily optimal solutions can be found by using a neural network. To do this the network must start from many randomly selected initial states. The network is simulated on the MPP (a 128 x 128 SIMD array machine), where researchers use the massive parallelism not only in solving the differential equations that govern the evolution of the network, but also by starting the network from many initial states at once, thus obtaining many solutions in one run. Researchers obtain speedups of two to three orders of magnitude over serial implementations and the promise through Analog VLSI implementations of speedups comensurate with human perceptual abilities.

Kamgar-Parsi, Behzad↗
DOE OSTI2023

Perivascular network segmentations derived from high-field MRI and their implications for perivascular and parenchymal mass transport in the rat brain

A custom segmentation workflow was applied to ex vivo high-field MR images of rat brains acquired following in vivo intraventricular contrast agent infusion to generate maps of the perivascular spaces (PVS). The resulting perivascular network segmentations enabled analysis of perivascular connections to the ventricles, parenchymal solute clearance, and dispersive solute transport within PVS. Numerous perivascular connections between the brain surface and the ventricles suggest the ventricles integrate into a PVS-mediated clearance system and raise the possibility of cerebrospinal fluid (CSF) return from the subarachnoid space to the ventricles via PVS. Assuming rapid solute exchange between the PVS and CSF spaces primarily by advection, the extensive perivascular network decreased the mean clearance distance from parenchyma to the nearest CSF compartment resulting in an over 21-fold reduction in the estimated diffusive clearance time scale, irrespective of solute diffusivity. This corresponds to an estimated diffusive clearance time scale under 10 min for amyloid-beta which suggests that the widespread distribution of PVS may render diffusion an effective parenchymal clearance mechanism. Additional analysis of oscillatory solute dispersion within PVS indicates that advection rather than dispersion is likely the primary transport mechanism for dissolved compounds greater than 66 kDa in the long (> 2 mm) perivascular segments identified here, although dispersion may be significant for smaller compounds in shorter perivascular segments.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRSDate not supplied

Run Time Improvement Efforts for the Roman Space Telescope Thermal Analysis

"Observatory thermal models for large, complex missions, such as the Wide Field InfraRed Survey Telescope (WFIRST) mission, produce an immense amount of data to be processed. Configuration management of the model throughout the project life cycle has mainly focused on which versions of the subsystem models form the current observatory level configuration. However, the results produced by the model are not nearly as well The Roman Space Telescope (RST), formerly known as the Wide Field InfraRed Survey Telescope, is the next great astrophysics observatory mission to follow the James Webb Space Telescope with a planned launch in 2026. As a large scale, flagship mission for NASA with challenging wave front error stability requirements, a single model approach for both thermal discipline analysis and thermo-optical distortion analysis has been used since the early days of the project. In alleviating the need to maintain two separate models for different analysis types, it imposes run time penalties on the thermal analysis with a large model with significant radiation heat exchange. Throughout the lifecycle of the project, the component models have steadily grown in size, resulting in a continuous growth of the overall observatory model with each update and consequently a considerable increase in the model run time. While ongoing efforts to reduce run time are continuously investigated, previous efforts had primarily focused on timestep size and total simulation time to reach quasi-equilibrium. More recently, studies were performed on the total number of radiation couplings (radks) included in the model, which has a nearly linear impact on run time, but increases exponentially with node count. As standard practice for spacecraft analysis, small radks were excluded from the temperature solution based on the assumption that their interchange/view factors have a negligible impact on heat flow. Four approaches were investigated to reduce the model run time while minimizing the impact on accuracy: (1) the Equivalent Radiation Network node, (2) Progressive Radk Inclusion as solution proceeds, (3) Targeted Radk Filtering for critical/non critical areas, and lastly (4) Representation of culled radks with Backloads. Furthermore, the investigation of model run time also revealed that cold cases took noticeably longer to run than hot cases; the root computational inefficiencies were explored along with the computation penalty of linearization of the external radks and recalculation of temperature dependent linear couplings at each timestep. This paper outlines the details of each of the above approaches and their impact on run time and model accuracy.

Thermal Analysis↗
NASA NTRSDate not supplied

Run Time Improvement Efforts for the Roman Space Telescope Thermal Analysis

Observatory thermal models for large, complex missions, such as the Wide Field InfraRed Survey Telescope (WFIRST) mission, produce an immense amount of data to be processed. Configuration management of the model throughout the project life cycle has mainly focused on which versions of the subsystem models form the current observatory level configuration. However, the results produced by the model are not nearly as well The Roman Space Telescope (RST), formerly known as the Wide Field InfraRed Survey Telescope, is the next great astrophysics observatory mission to follow the James Webb Space Telescope with a planned launch in 2026. As a large scale, flagship mission for NASA with challenging wave front error stability requirements, a single model approach for both thermal discipline analysis and thermo-optical distortion analysis has been used since the early days of the project. In alleviating the need to maintain two separate models for different analysis types, it imposes run time penalties on the thermal analysis with a large model with significant radiation heat exchange. Throughout the lifecycle of the project, the component models have steadily grown in size, resulting in a continuous growth of the overall observatory model with each update and consequently a considerable increase in the model run time. While ongoing efforts to reduce run time are continuously investigated, previous efforts had primarily focused on timestep size and total simulation time to reach quasi-equilibrium. More recently, studies were performed on the total number of radiation couplings (radks) included in the model, which has a nearly linear impact on run time, but increases exponentially with node count. As standard practice for spacecraft analysis, small radks were excluded from the temperature solution based on the assumption that their interchange/view factors have a negligible impact on heat flow. Four approaches were investigated to reduce the model run time while minimizing the impact on accuracy: (1) the Equivalent Radiation Network node, (2) Progressive Radk Inclusion as solution proceeds, (3) Targeted Radk Filtering for critical/non critical areas, and lastly (4) Representation of culled radks with Backloads. Furthermore, the investigation of model run time also revealed that cold cases took noticeably longer to run than hot cases; the root computational inefficiencies were explored along with the computation penalty of linearization of the external radks and recalculation of temperature dependent linear couplings at each timestep. This paper outlines the details of each of the above approaches and their impact on run time and model accuracy.

Thermal Analysis↗