Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “random networks”

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 739 records · Page 41
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Graph Sparsification by Approximate matrix Multiplication

Graphs arising in statistical problems, signal processing, large networks, combinatorial optimization, and data analysis are often dense, which causes both computational and storage bottlenecks. One way of sparsifying a weighted graph, while sharing the same vertices as the original graph but reducing the number of edges, is through spectral sparsification. We study this problem through the perspective of RandNLA. Specifically, we utilize randomized matrix multiplication to give a clean and simple analysis of how sampling according to edge weights gives a spectral approximation to graph Laplacians, without requiring spectral information. Through the CR–MM algorithm, we attain a simple and computationally efficient sparsifier whose resulting Laplacian estimate is unbiased and of minimum variance. Here, we define a new notion of additive spectral sparsifiers, which has not been considered in the literature.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

GraphAlign: Graph-Enabled Machine Learning for Seismic Event Filtering

This report summarizes results from a 2 year effort to improve the current automated seismic event processing system by leveraging machine learning models that can operated over the inherent graph data structure of a seismic sensor network. Specifically, the GraphAlign project seeks to utilize prior information on which stations are more likely to detect signals originating from particular geographic regions to inform event filtering. To date, the GraphAlign team has developed a Graphical Neural Network (GNN) model to filter out false events generated by the Global Associator (GA) algorithm. The algorithm operates directly on waveform data that has been associated to an event by building a variable sized graph of station waveforms nodes with edge relations to an event location node. This builds off of previous work where random forest models were used to do the same task using hand crafted features. The GNN model performance was analyzed using an 8 week IMS/IDC dataset, and it was demonstrated that the GNN outperforms the random forest baseline. We provide additional error analysis of which events the GNN model performs well and poorly against concluded by future directions for improvements.

58 GEOSCIENCES↗
NASA NTRS1999

Current Trend Towards Using Soft Computing Approaches to Phase Synchronization in Communication Systems

This paper surveys recent advances in communications that utilize soft computing approaches to phase synchronization. Soft computing, as opposed to hard computing, is a collection of complementary methodologies that act in producing the most desirable control, decision, or estimation strategies. Recently, the communications area has explored the use of the principal constituents of soft computing, namely, fuzzy logic, neural networks, and genetic algorithms, for modeling, control, and most recently for the estimation of phase in phase-coherent communications. If the receiver in a digital communications system is phase-coherent, as is often the case, phase synchronization is required. Synchronization thus requires estimation and/or control at the receiver of an unknown or random phase offset.

Drake, Jeffrey T.↗
NASA NTRS2018

Toward Long-Term Aquatic Science Products from Heritage Landsat Missions

This paper aims at generating a long-term consistent record of Landsat-derived remote sensing reflectance (Rrs) products, which are central for producing downstream aquatic science products (e.g., concentrations of total suspended solids). The products are derived from Landsat-5 and Landsat-7 observations leading to Landsat-8 era to enable retrospective analyses of inland and nearshore coastal waters. In doing so, the data processing was built into the SeaWiFS Data Analysis System (SeaDAS) followed by vicariously calibrating Landsat-7 and -5 data using reference in situ measurements and near-concurrent ocean color products, respectively. The derived Rrs products are then validated using (a) matchups using the Aerosol Robotic Network (AERONET) data measured by in situ radiometers, i.e., AERONET-OC, and (b) ocean color products at select sites in North America. Following the vicarious calibration adjustments, it is found that the overall biases in Rrs products are significantly reduced. The root-mean-square errors (RMSE), however, indicate noticeable uncertainties due to random and systematic noise. Long-term (since 1984) seasonal Rrs composites over 12 coastal and inland systems are further evaluated to explore the utility of Landsat archive processed via SeaDAS. With all the qualitative and quantitative assessments, it is concluded that with careful algorithm developments, it is possible to discern natural variability in historic water quality conditions using heritage Landsat missions. This requires the changes in Rrs exceed maximum expected uncertainties, i.e., 0.0015 [1/sr], estimated from mean RMSEs associated with the matchups and intercomparison analyses. It is also anticipated that Landsat-5 products will be less susceptible to uncertainties in turbid waters with Rrs(660) > 0.004 [1/sr], which is equivalent of ~1.2% reflectance. Overall, end-users may utilize heritage Rrs products with "fitness-for-purpose" concept in mind, i.e., products could be valuable for one application but may not be viable for another. Further research should be dedicated to enhancing atmospheric correction to account for non-negligible near-infrared reflectance in CDOM-rich and extremely turbid waters.

Pahlevan, Nima↗
NASA NTRS2017

Shallow Lunar Seismic Activity and the Current Stress State of the Moon

A vast, global network of more than 3200 lobate thrust fault scarps has been revealed in high resolution Lunar Reconnaissance Orbiter Camera (LROC) images. The fault scarps are very young, less than 50 Ma, based on their small scale and crisp appearance, crosscutting relations with small-diameter impact craters, and rates of infilling of associated small, shallow graben and may be actively forming today. The population of young thrust fault scarps provides a window into the recent stress state of the Moon and offers insight into the origin of global lunar stresses. The distribution of orientations of the fault scarps is non-random, inconsistent with isotropic stresses from late-stage global contraction as the sole source of stress. Modeling shows that tidal stresses contribute significantly to the current stress state of the lunar crust. Tidal stresses (orbital recession and diurnal tides) superimposed on stresses from global contraction result in non-isotropic compressional stress and may produce thrust faults consistent with lobate scarp orientations. At any particular point on the lunar surface, peak compressive stress will be reached at a certain time in the diurnal cycle. Coseismic slip events on currently active thrust faults are expected to be triggered when peak stresses are reached. Analysis of the timing of the 28 the shallow moonquakes recorded by the Apollo seismic network shows that 19 indeed occur when the Moon is closer to apogee, while only 9 shallow events occur when the Moon is closer to perigee. Here we report efforts to refine the model for the current stress state of the Moon by investigating the contribution of polar wander. Progress on relocating the epicentral locations of the shallow moonquakes using an algorithm designed for sparse networks is also reported.

structure↗
DOE OSTI2025

Denoising Autoencoder for Reconstructing Sensor Observation Data and Predicting Evapotranspiration: Noisy and Missing Values Repair and Uncertainty Quantification

Abstract Machine learning (ML) methods applied in scientific research often deal with interrelated features in high‐dimensional data. Reducing data noise and redundancy is needed to increase prediction accuracy and efficiency especially when dealing with data from field sensors. We explored an unsupervised learning method, the denoising autoencoder (DAE), to extract the underlying data structure from noisy raw data in the context of predicting hydrologic quantities from multiple field sensors. These sensors have intrinsic instrumental noise and occasional malfunctions that cause missing values. Our DAE neural network reconstructed meteorological sensor data containing noise and missing values to predict evapotranspiration in a mountainous watershed. The DAE reconstructed the sensor variables with a mean coefficient of determination value of 0.77 across 15 dimensions representing individual sensors. It reduced variance and bias uncertainties compared to a classical autoencoder model. The reconstruction quality varied across dimensions depending on their cross‐correlation and alignment with the underlying data structure. Uncertainties arising from the model structure were overall higher than those resulting from data corruption. We attached the DAE structure to a downstream ET‐prediction neural network in three formats and achieved reasonably accurate ET predictions . The use of the DAE notably reduced variance uncertainty in ET prediction. However, excessive variance reduction may be accompanied by an increase in bias due to the intrinsic bias‐variance tradeoff. Our method of evaluating and reducing uncertainties in aggregated data from different sources can be used to improve predictive models, process understanding, and uncertainty quantification for better water resource management. Plain Language Summary We present a machine learning method, namely the denoising autoencoder, which reduces the effects of data noise and missing values typically present in scientific data sets collected through sensor measurements. This method selects the most relevant information from noisy raw data collected by the instruments and fills in missing values. To demonstrate the effectiveness of our method, we applied it to predict evapotranspiration, a hydrologic variable that represents the water moved from the land surface to the atmosphere through a combination of evaporation and plant water use (transpiration). We also used a random sampling technique (the Monte Carlo method) to compare the uncertainty in the predictions when using the raw and noisy data versus the reconstructed data. The denoising process produced more accurate predictions of evapotranspiration with less uncertainty. Improved predictions of evapotranspiration can lead to a better understanding and accounting of water budgets. This ML approach is broadly suitable for a wide variety of applications that involve noisy sensor data with missing values. Key Points We used a denoising autoencoder (DAE) neural network to reduce noise in meteorological and soil sensor observations by on average We used Monte Carlo sampling to estimate the bias and variance of all model outputs, including uncertainty sources from data and the model We attached the DAE component to a downstream neural network to predict ET with the variance reduced by , compared to that without the DAE

denoising autoencoder↗
DOE OSTI2025

Data-Driven Closures and Assimilation for Stiff Multiscale Random Dynamics

Here, we introduce a data-driven and physics-informed framework for propagating uncertainty in stiff, multiscale random ordinary differential equations (RODEs) driven by correlated (colored) noise. Unlike systems subjected to Gaussian white noise, a deterministic equation for the joint probability density function (PDF) of RODE state variables does not exist in closed form. Moreover, such an equation would require as many phase-space variables as there are states in the RODE system. To alleviate this curse of dimensionality, we instead derive exact, albeit unclosed, reduced-order PDF (RoPDF) equations for low-dimensional observables/quantities of interest. The unclosed terms take the form of state-dependent conditional expectations, which are directly estimated from data at sparse observation times. However, for systems exhibiting stiff, multiscale dynamics, data sparsity introduces regression discrepancies that compound during RoPDF evolution. This is overcome by introducing a kinetic-like defect term to the RoPDF equation, which is learned by assimilating in sparse, low-fidelity RoPDF estimates. Two assimilation methods are considered, namely nudging and deep neural networks, which are successfully tested against Monte Carlo simulations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Microstructure analysis of laser beam weldments performed on neutron-irradiated 304L steel containing 3 and 8 appm helium

We report that AISI 304 L austenitic stainless steel is one of the major structural materials used in light water reactors (LWRs). In the future, 304 L components may require repair, and several welding techniques have been proposed as candidates. In this study, laser beam welding using the low-energy contribution approach was performed in ~2016 on neutron-irradiated AISI 304 L steel with nominal values of 3 and 8 appm helium (He). The goal was to investigate the impact of helium on the irradiated material weldability and evaluate helium-associated damage. The weld and heat-affected zone (HAZ) microstructure was studied using scanning electron microscopy, electron backscatter diffraction, and transmission electron microscopy (TEM). Analysis of the weldment cross-sections did not reveal severe cracking. Still, evidence of incipient helium-associated damage was observed in the HAZ, manifesting as degraded grain boundaries (GBs) “decorated” by pore chains and mostly small (i.e., ≤ 25-30 µm) scattered cracks. The largest observed crack was ~50 µm in length. Helium-associated damage was localized within ~200-300 µm of the weld pool boundary, and the fraction of the compromised GBs, as a rule, was below ~9-11% of the total GB network. TEM analysis showed significant annealing of the radiation-induced defects at distances up to ~300 µm from the weld pool boundary. Inside the HAZ, the cavities (likely, helium bubbles) tended to form at GBs and inclusions, such as manganese sulfide precipitates in the grain interior. Crystallography analysis of the HAZ damage showed that random high-angle boundaries were most susceptible to helium-associated damage. In contrast, low-angle random boundaries and twin boundaries appeared to be strongly resistant to degradation from the combined effects of helium and welding.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Tracking structural solvent reorganization and recombination dynamics following e – photoabstraction from aqueous I – with femtosecond x-ray spectroscopy and scattering

Here we present a sub-picosecond resolved investigation of the structural solvent reorganization and geminate recombination dynamics following 400 nm two-photon excitation and photodetachment of a valence p electron from the aqueous atomic solute, I–(aq). The measurements utilized time-resolved X-ray Absorption Near Edge Structure (TR-XANES) spectroscopy and X-ray Solution Scattering (TR-XSS) at the Linac Coherent Light Source x-ray free electron laser in a laser pump/x-ray probe experiment. The XANES measurements around the L 1 -edge of the generated nascent iodine atoms (I 0 ) yield an average electron ejection distance from the iodine parent of 7.4 ± 1.5 Å with an excitation yield of about 1/3 of the 0.1M NaI aqueous solution. The kinetic traces of the XANES measurement are in agreement with a purely diffusion-driven geminate iodine–electron recombination model without the need for a long-lived (I 0 :e–) contact pair. Nonequilibrium classical molecular dynamics simulations indicate a delayed response of the caging H 2 O solvent shell and this is supported by the structural analysis of the XSS data: We identify a two-step process exhibiting a 0.1 ps delayed solvent shell reorganization time within the tight H-bond network and a 0.3 ps time constant for the mean iodine–oxygen distance changes. The results indicate that most of the reorganization can be explained classically by a transition from a hydrophilic cavity with a well-ordered first solvation shell (hydrogens pointing toward I–) to an expanded cavity around I 0 with a more random orientation of the H 2 O molecules in a broadened first solvation shell.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS1973

The International Planetary Patrol Program - An assessment of the first three years.

Day-to-day and hour-to-hour changes in the large-scale atmospheric and surface features of the planets can now be studied more effectively than previously possible. Since 1969 a network of observatories has obtained almost uninterrupted photographic coverage during all apparitions of Jupiter and Mars, plus some of Venus. Patrol films and catalogs are available to the scientific community. Recent or current analyses include the distribution and motion of clouds on Mars, the development and decay of Martian dust storms, the seasonal, diurnal, and random fluctuations in contrast between adjacent light and dark regions on Mars, the detection of vertical shear in the Jovian atmosphere, the longitudinal oscillation of the Red Spot, the dependence of rotation period on xenographic latitude and on time, the eruption and spread of south equatorial belt disturbances, and the retrograde circulation of the Venus cloud deck.

Baum, W. A.↗
NASA NTRS1990

An alternative resource sharing scheme for land mobile satellite services

A preliminary comparison between the two competing channelization concepts for the Land Mobile Satellite Services (LMSS), namely frequency division (FD) and code division (CD), is presented. Both random access and demand-assigned approaches are considered under these concepts. The CD concept is compared with the traditional FD concept based on the system consideration and a projected traffic model. It is shown that CD is not particularly attractive for the first generation Mobile Satellite Services because of the spectral occupancy of the network bandwidth. However, the CD concept is a viable alternative for future systems such as the personal access satellite system (PASS) in the Ka-band spectrum where spectral efficiency is not of prime concern. The effects of power robbing and voice activity factor are incorporated. It was shown that the traditional rule of thumb of dividing the number of raw channels by the voice activity factor to obtain the effective number of channels is only valid asymptotically as the aggregated traffic approaches infinity.

Yan, Tsun-Yee↗
NASA NTRS2004

Improving Search Algorithms by Using Intelligent Coordinates

We consider algorithms that maximize a global function G in a distributed manner, using a different adaptive computational agent to set each variable of the underlying space. Each agent eta is self-interested; it sets its variable to maximize its own function g (sub eta). Three factors govern such a distributed algorithm's performance, related to exploration/exploitation, game theory, and machine learning. We demonstrate how to exploit alI three factors by modifying a search algorithm's exploration stage: rather than random exploration, each coordinate of the search space is now controlled by a separate machine-learning-based player engaged in a noncooperative game. Experiments demonstrate that this modification improves simulated annealing (SA) by up to an order of magnitude for bin packing and for a model of an economic process run over an underlying network. These experiments also reveal interesting small-world phenomena.

Wolpert, David H.↗
DOE OSTI2020

Noise-Based Selection of Robust Inherited Model for Accurate Continual Learning

There is a growing demand for an intelligent system to continually learn knowledge from a data stream. Continual learning requires both the preservation of previous knowledge (i.e., avoiding catastrophic forgetting) and the acquisition of new knowledge. Different from previous works that focus only on model adaptation (e.g., regularization, network expansion, memory rehearsal, etc.), we propose a novel training scheme named acquisitive learning (AL), which emphasizes both the knowledge inheritance and knowledge acquisition. AL starts from an elaborately selected model with pre-trained knowledge (the inherited model) and then adapts it to new data using segmented training. The selection is achieved by injecting random noise to various inherited models for better model robustness, which promises higher accuracy in further knowledge acquisition. The approach is validated by the visualization of the loss landscape and quantitative roughness measurement. The combination of the selective inherited model and knowledge acquisition reduces catastrophic forgetting by 10X on the CIFAR-100 dataset.

Du, Xiaocong↗
DOE OSTI2022

A survey of unsupervised learning methods for high-dimensional uncertainty quantification in black-box-type problems

Constructing surrogate models for uncertainty quantification (UQ) on complex partial differential equations (PDEs) having inherently high-dimensional O(10 n ), n ≥ 2, stochastic inputs (e.g., forcing terms, boundary conditions, initial conditions) poses tremendous challenges. The “curse of dimensionality” can be addressed with suitable unsupervised learning techniques used as a pre-processing tool to encode inputs onto lower-dimensional subspaces while retaining its structural information and meaningful properties. In this work, we review and investigate thirteen dimension reduction methods including linear and nonlinear, spectral, blind source separation, convex and non-convex methods and utilize the resulting embeddings to construct a mapping to quantities of interest via polynomial chaos expansions (PCE). Here, we refer to the general proposed approach as manifold PCE (m-PCE), where manifold corresponds to the latent space resulting from any of the studied dimension reduction methods. To investigate the capabilities and limitations of these methods we conduct numerical tests for three physics-based systems (treated as black-boxes) having high-dimensional stochastic inputs of varying complexity modeled as both Gaussian and non-Gaussian random fields to investigate the effect of the intrinsic dimensionality of input data. We demonstrate both the advantages and limitations of the unsupervised learning methods and we conclude that a suitable m-PCE model provides a cost-effective approach compared to alternative algorithms proposed in the literature, including recently proposed expensive deep neural network-based surrogates and can be readily applied for high-dimensional UQ in stochastic PDEs.

42 ENGINEERING↗
DOE OSTI2021

Mitigating Catastrophic Forgetting in Deep Learning in a Streaming Setting Using Historical Summary

Recent advancements in scientific equipment and the adaptation of electronics and the Internet of Things (IoT) in our everyday lives resulted in large and complex data production at a high rate. Making meaningful and timely knowledge discovery at a modest cost from this big data is difficult for computing power and storage limitations. Training deep learning models incrementally in a streaming setting can help us with overcoming these limitations. However, in a well-known phenomenon named catastrophic forgetting, incrementally trained models increasingly perform poorly on the past data. To mitigate catastrophic forgetting in training in a streaming setting, we propose constructing a historical summary over time and use the summary with newly arrived data during incremental training. We propose various data summarization techniques such as random sampling, micro clustering, coreset computation, and Auto Encoders to counteract catastrophic forgetting. We built a pipeline for incremental training with a historical summary for training deep learning models for streaming data. We demonstrate the effectiveness of historical summary in mitigating catastrophic forgetting using three case studies involving three different deep learning applications: an Artificial Neural Network (ANN) for classification task on MNIST dataset, a language model (RNN-LM) on the WikiText2 dataset, and a Convolutional Neural Network (CNN), ResNet50 to classify the ImageNet dataset. Through the training of the models, we observe that catastrophic forgetting is evident in ANN and CNN but not in an RNN. For the first task, our method recovers up to 47.9% lost accuracy due to catastrophic forgetting. For the third task, the historical summary recovers classification accuracy by up to 25%. For the second task, though there is not proof of catastrophic forgetting, the training performance (PPL) improves by up to 26% with historical summary.

Dash, Sajal↗
DOE OSTI2022

MAD: Self-Supervised Masked Anomaly Detection Task for Multivariate Time Series

In this paper, we introduce Masked Anomaly Detection (MAD), a general self-supervised learning task for multivariate time series anomaly detection. With the increasing availability of sensor data from industrial systems, being able to detecting anomalies from streams of multivariate time series data is of significant importance. Given the scarcity of anomalies in real-world applications, the majority of literature has been focusing on modeling normality. The learned normal representations can empower anomaly detection as the model has learned to capture certain key underlying data regularities. A typical formulation is to learn a predictive model, i.e., use a window of time series data to predict future data values. In this paper, we propose an alternative self-supervised learning task. By randomly masking a portion of the inputs and training a model to estimate them using the remaining ones, MAD is an improvement over the traditional left-to-right next step prediction (NSP) task. Our experimental results demonstrate that MAD can achieve better anomaly detection rates over traditional NSP approaches when using exactly the same neural network (NN) base models, and can be modified to run as fast as NSP models during test time on the same hardware, thus making it an ideal upgrade for many existing NSP-based NN anomaly detection models.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Intelliquench: An Adaptive Machine Learning System for Detection of Superconducting Magnet Quenches

In superconducting magnets, the irreversible transition of a portion of the conductor to resistive state is called a “quench.” Having large stored energy, magnets can be damaged by quenches due to localized heating, high voltage, or large force transients. Unfortunately, current quench protection systems can only detect a quench after it happens, and mitigating risks in Low Temperature Superconducting (LTS) accelerator magnets often requires fast response (down to ms). Additionally, protection of High Temperature Superconducting (HTS) magnets is still suffering from prohibitively slow quench detection. In this study, we lay the groundwork for a quench prediction system using an auto-encoder fully-connected deep neural network. After dynamically trained with data features extracted from acoustic sensors around the magnet, the system detects anomalous events seconds before the quench in most of our data. While the exact nature of the events is under investigation, we show that the system can “forecast” a quench before it happens under magnet training conditions through a randomized experiment. This opens up the way of integrated data processing, potentially leading to faster and better diagnostics and detection of magnet quenches

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗