Search NASA⌕ Search

SEARCH · Search NASA

Results for “Graph filtering”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
NASA NTRS2000

Planar Particle Imaging Doppler Velocimetry Developed

Two current techniques exist for the measurement of planar, three-component velocity fields. Both techniques require multiple views of the illumination plane in order to extract all three velocity components. Particle image velocimetry (PIV) is a high-resolution, high accuracy, planar velocimetry technique that provides valuable instantaneous velocity information in aeropropulsion test facilities. PIV can provide three-component flow-field measurements using a two-camera, stereo viewing configuration. Doppler global velocimetry (DGV) is another planar velocimetry technique that can provide three component flow-field measurements; however, it requires three detector systems that must be located at oblique angles from the measurement plane. The three-dimensional configurations of either technique require multiple (DGV) or at least large (stereo PIV) optical access ports in the facility in which the measurements are being conducted. Optical access is extremely limited in aeropropulsion test facilities. In many cases, only one optical access port is available. A hybrid measurement technique has been developed at the NASA Glenn Research Center, planar particle image and Doppler velocimetry (PPIDV), which combines elements from both the PIV and DGV techniques into a single detection system that can measure all three components of velocity across a planar region of a flow field through a single optical access port. In the standard PIV technique, a pulsed laser is used to illuminate the flow field at two closely spaced instances in time, which are recorded on a "frame-straddling" camera, yielding a pair of single-exposure image frames. The PIV camera is oriented perpendicular to the light sheet, and the processed PIV data yield the two-component velocity field in the plane of the light sheet. In the standard DGV technique, an injection-seeded Nd:YAG pulsed laser light sheet illuminates the seeded flow field, and three receiver systems are used to measure three components of velocity. The receiver systems are oriented at oblique angles to the light sheet in order to accurately resolve the three-component velocity. Each DGV receiver system contains two cameras, which share a common view of the illuminated flow through a beam-splitting cube. One camera views the illuminated flow directly (reference camera) and the second camera images the illuminated flow through an iodine vapor cell (signal camera). The laser frequency (wavelength) is adjusted so that the Doppler-shifted light from particles in the flow falls on an iodine absorption feature, see the following graph. The iodine vapor cell acts as a frequency-to-velocity filter by modulating the intensity of the transmitted light as a function of the flow velocity (Doppler shift). The ratio of the signal and reference images yields the component of the flow velocity along the bisector of the laser sheet propagation direction and the receiver system observation direction. The hybrid system employs a single-component DGV receiver system configured to simultaneously acquire PIV image data, as shown in the following diagram. The cameras used in the DGV receiver are replaced with PIV frame-straddling cameras, and the receiver system views the illuminated light sheet plane at 90 (as in the standard PIV configuration).

Wernet, Mark P.↗
DOE OSTI2024

LENS: Learning Enabled Network Synthesis

RTRC and UMD have developed novel machine learning based methods under the ARPA-E DIFFERENTIATE program for rapid acceleration of hypothesis generation in complex architecture design spaces involving both discrete choices of component inclusion and interconnection and continuous parametric decisions. The project named Learning Enabled Network Synthesis (LENS) further demonstrated the developed methods on challenging electrical power converter design problems by identifying the most suitable circuit topologies and simultaneously selecting the most appropriate components to achieve optimized design of power converter with improved performances. We demonstrated that LENS could enable exploration of very large design space of circuit topologies and components by addressing the limitations of conventional design process in non-linear, high switching speed, multi-dimensional power converter design and optimization. The key innovation developed in LENS is the seamless integration of statistical learning and logical reasoning techniques and building on the individual strengths of these techniques for rapid hypothesis discovery. The main component of LENS comprises of: 1) Graph Reasoning Engine (GRE) to enforce composition rules that rapidly reject all discrete architectures that are composed incorrectly and generates an adaptive database of feasible designs which can be used by ML modules, 2) Graph Generative Learning module which is a deep neural network based generative model for graph architectures which can enable design space exploration beyond the dataset generated by the GRE, 3) Graph Reduced Order Model (ROM) for graph domains for accelerating computation of output metrics, and 4) Active learning and Rule Discovery module for sample efficient learning and extracting logical rules from the learned ML models which will be integrated in the GRE to enhance the filtering effectiveness. LENS approach can be applied to any design domains where designs can be represented as multi-attribute graphs. The LENS team integrated the various technical innovations listed above into an optimization pipeline and exercised the optimization pipeline on the converter design problem. The LENS project demonstrated that the developed AI/ML technologies can be used to generate novel converter circuits >45x faster than experts on chosen use-cases. This can enable faster design space exploration and identification of new designs which are not considered by experts due to the increasing design space complexity. This has significant potential impact on the public and energy needs of the country. It is currently estimated that 30% of all electrical powers generated passes through power converters. The future estimate is that 80% of all power generated would be passing through converters. LENS fills a critical gap in this space since by accelerating the design process the designers would be able to generate more efficient converters which can lead to significant energy savings for the country.

42 ENGINEERING↗
DOE OSTI2022

Improved Distributed-memory Triangle Counting by Exploiting the Graph Structure

Graphs are ubiquitous in modeling complex systems and representing interactions between entities to uncover structural information of the domain. Traditionally, graph analytics workloads are challenging to efficiently scale (both strong and weak cases) on distributed memory due to the irregular memory-access driven nature (with little or no computations) of the methods. The structure of graphs and their relative distribution over the processing elements poses another level of complexity, making it difficult to attain sustainable scalability across platforms. In this paper, we discuss enhancements to TriC, a distributed-memory implementation of graph triangle counting using Message Passing Interface (MPI), which was featured in the 2020 Graph Challenge competition. We have made some incremental enhancements to TriC, primarily adopting a user-defined buffering strategy to overcome the startup problem for large graphs (by fixing the memory for intermediate data), and experimenting with probabilistic data structures such as bloom filter to improve the query response time for assessing edge existence, at the expense of increasing the overall false positive rate. These adjustments have led to a modest improvements in most cases, as compared to the previous version.

Graph Analytics, HPC↗
NASA NTRS1988

New insights in the numerical reliability properties of existing Kalman filter implementations

The convergence properties of Kalman filter algorithms are investigated analytically. A theoretical error analysis is performed on four types of algorithms, as defined by Anderson and Moore (1979): (1) a conventional Kalman filter, (2) a square-root (SR) covariance filter, (3) the Chandrasekhar SR filter, and (4) an SR information filter. The derivations are given in detail, and numerical results for the flight-path reconstruction problem studied by Verhaegen (1987) are presented in tables and graphs. It is shown that error propagation in algorithms (1) and (2) is sensitive to the condition number of the innovation-signal covariance matrix and the spectral norm of the filter state-transition matrix, whereas other parameters are dominant in (3) and (4). Filter (2) is found to be the most reliable for the class of problems studied.

Verhaegen, M. H.↗
DOE OSTI2021

Distance-Weighted Graph Neural Networks on FPGAs for Real-Time Particle Reconstruction in High Energy Physics

Graph neural networks have been shown to achieve excellent performance for several crucial tasks in particle physics, such as charged particle tracking, jet tagging, and clustering. An important domain for the application of these networks is the FGPA-based first layer of real-time data filtering at the CERN Large Hadron Collider, which has strict latency and resource constraints. We discuss how to design distance-weighted graph networks that can be executed with a latency of less than one μs on an FPGA. To do so, we consider a representative task associated to particle reconstruction and identification in a next-generation calorimeter operating at a particle collider. We use a graph network architecture developed for such purposes, and apply additional simplifications to match the computing constraints of Level-1 trigger systems, including weight quantization. Using the hls4ml library, we convert the compressed models into firmware to be implemented on an FPGA. Performance of the synthesized models is presented both in terms of inference accuracy and resource usage.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
NASA NTRS1986

Learned pattern recognition using synthetic-discriminant-functions

A method of using synthetic-discriminant-functions to facilitate learning in a pattern recognition system is discussed. Learning is accomplished by continually adding images to the training set used for synthetic discriminant functions (SDF) construction. Object identification is performed by efficiently searching a library of SDF filters for the maximum optical correlation. Two library structures are discussed - binary tree and multilinked graph - along with maximum ascent, back-tracking, perturbation, and simulated annealing searching techniques. By incorporating the distortion invariant properties of SDFs within a library structure, a robust pattern recognition system can be produced.

Jared, David A.↗
NASA NTRS1973

Analytical study to define a helicoper stability derivative extraction method, volume 2

The data generated during tests to determine helicopter stability derivatives are presented in the form of graphs. The data are based on CH-53A helicopters and CH-54B helicopters with various digital filters operating at various airspeeds. Curves are plotted for a time history comparison of identified derivative models against flight data for the helicopters at specific airspeeds and maneuvers.

Molusis, J. A.↗
NASA NTRS1987

Surface brightness profiles of 10 comets

CCD photometric observations of the comae of 10 comets, obtained at the 4-m and 2.1-m telescopes at KPNO during 1985-1986 using filters centered at 700.5, 650.0, or 546.0 nm, are reported. The data are presented in extensive tables and graphs and characterized in detail. The radial surface brightness profiles are shown to be steeper than predicted by an idealized spherically symmetric steady-state comet model, the steepness increasing with the projected distance from the nucleus. These profiles are attributed, on the basis of Monte Carlo simulations, to imperfect coupling between the sublimated gas and the optically dominant grains of the coma.

Jewitt, D. C.↗
NASA NTRS2011

Decentralized Observer with a Consensus Filter for Distributed Discrete-Time Linear Systems

This paper presents a decentralized observer with a consensus filter for the state observation of a discrete-time linear distributed systems. In this setup, each agent in the distributed system has an observer with a model of the plant that utilizes the set of locally available measurements, which may not make the full plant state detectable. This lack of detectability is overcome by utilizing a consensus filter that blends the state estimate of each agent with its neighbors' estimates. We assume that the communication graph is connected for all times as well as the sensing graph. It is proven that the state estimates of the proposed observer asymptotically converge to the actual plant states under arbitrarily changing, but connected, communication and sensing topologies. As a byproduct of this research, we also obtained a result on the location of eigenvalues, the spectrum, of the Laplacian for a family of graphs with self-loops.

embedded consensus↗
NASA NTRSDate not supplied

Populating a Graph Database to Run a Usage-Based Discovery Tool

Most dataset discovery tools for Earth Observation data rely on descriptions and other metadata of the datasets, using keyword searches or attribute filtering to determine relevance. However, these descriptions often do not include the potential uses of the data. Thus, a user working on floods will rarely see few if any rainfall datasets show up in such a search. The Usage Based Discovery tool, on the other hand, offers usage instances to the user, either research articles or applications, along with the datasets that those usage instances used. This allows a user, particularly one new to the world of Earth Observation data, to investigate which datasets are used in similar cases. The information that powers Usage-Based Discovery is a graph database of relationships of usage to dataset and usage to topic, allowing the user to narrow their search for similar cases. In order to scale out to a graph database rich enough to provide a satisfactory user experience, we combine manual and automated processes to populate the graph. The initial content of the graph has been seeded primarily via human-aided data curation methods, using sites like Google Scholar. To scale up this effort, we’ve employed crowdsourcing. It is easy for anyone to contribute to our graph using their Open Researcher and Contributor Identifier for authorization. We’re now experimenting with Machine Learning and Natural Language Processing to help automate population of the graph, starting with the classification of research articles by topic. Finding adequate training data in the absence of a comprehensive and open research article API continues to be a significant challenge.

Vincent Inverso↗
DOE OSTI2021

Performance of a geometric deep learning pipeline for HL-LHC particle tracking

The Exa.TrkX project has applied geometric learning concepts such as metric learning and graph neural networks to HEP particle tracking. Exa.TrkX’s tracking pipeline groups detector measurements to form track candidates and filters them. The pipeline, originally developed using the TrackML dataset (a simulation of an LHC-inspired tracking detector), has been demonstrated on other detectors, including DUNE Liquid Argon TPC and CMS High-Granularity Calorimeter. This paper documents new developments needed to study the physics and computing performance of the Exa.TrkX pipeline on the full TrackML dataset, a first step towards validating the pipeline using ATLAS and CMS data. The pipeline achieves tracking efficiency and purity similar to production tracking algorithms. Crucially for future HEP applications, the pipeline benefits significantly from GPU acceleration, and its computational requirements scale close to linearly with the number of particles in the event.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2017

APT malware static trace analysis through bigrams and graph edit distance

Research and business organizations are vulnerable to attack by malware, particularly advanced persistent threat malware tailored for a specific target. Malware identification is made more difficult because samples can be subtly altered to avoid detection by methods that check for an identical match to known code. Different versions of an original piece of malware form a malware family. And when new malicious software is identified, reverse engineers seek to identify its origin and purpose. Knowing whether new malware is from a known family or a previously unobserved family aids the efficiency of reverse engineers. Furthermore, this article presents a three-stage method to classify new malware into a family by comparing its similarity to existing static traces, and assigning it to the most similar family. First, a fast filtering method creates a shortlist of samples with some similarity to the new malware, using a simple bigram comparison of the instructions. The second stage takes the call graph view of the shortlisted static traces and uses simulated annealing to estimate the graph edit distance, a measure of dissimilarity between graphs. Finally, a random forest classifier combines the previous two results to predict the family to which a new sample belongs. Our paper also considers how to detect when malware is from a new family.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Charged Particle Tracking via Edge-Classifying Interaction Networks

Recent work has demonstrated that geometric deep learning methods such as graph neural networks (GNNs) are well suited to address a variety of reconstruction problems in high-energy particle physics. In particular, particle tracking data are naturally represented as a graph by identifying silicon tracker hits as nodes and particle trajectories as edges, given a set of hypothesized edges, edge-classifying GNNs identify those corresponding to real particle trajectories. In this work, we adapt the physics-motivated interaction network (IN) GNN toward the problem of particle tracking in pileup conditions similar to those expected at the high-luminosity Large Hadron Collider. Assuming idealized hit filtering at various particle momenta thresholds, we demonstrate the IN’s excellent edge-classification accuracy and tracking efficiency through a suite of measurements at each stage of GNN-based tracking: graph construction, edge classification, and track building. The proposed IN architecture is substantially smaller than previously studied GNN tracking architectures; this is particularly promising as a reduction in size is critical for enabling GNN-based tracking in constrained computing environments. Furthermore, the IN may be represented as either a set of explicit matrix operations or a message passing GNN. Efforts are underway to accelerate each representation via heterogeneous computing resources towards both high-level and low-latency triggering applications.

accelerator physics↗
DOE OSTI2012

The discrete Kalman filtering approach for seismic signals deconvolution

Seismic signals are a convolution of reflectivity and seismic wavelet. One of the most important stages in seismic data processing is deconvolution process; the process of deconvolution is inverse filters based on Wiener filter theory. This theory is limited by certain modelling assumptions, which may not always valid. The discrete form of the Kalman filter is then used to generate an estimate of the reflectivity function. The main advantage of Kalman filtering is capability of technique to handling continually time varying models and has high resolution capabilities. In this work, we use discrete Kalman filter that it was combined with primitive deconvolution. Filtering process works on reflectivity function, hence the work flow of filtering is started with primitive deconvolution using inverse of wavelet. The seismic signals then are obtained by convoluting of filtered reflectivity function with energy waveform which is referred to as the seismic wavelet. The higher frequency of wavelet gives smaller wave length, the graphs of these results are presented.

58 GEOSCIENCES↗
DOE OSTI2023

Convoluted filtering for process cycle modeling

Principles of materials science and engineering, physics, mathematics, and information science are used to extract knowledge and insights from the process-structure–property-performance relationships hidden in materials data. The process-structure modeling can be accelerated without loss of interpretability, with artificial intelligence tools that mimic the salient features of the process and process-structure relations. In this work, a novel convoluted model-filtering technique was exploited to build and successfully train the Convoluted Filter (CoFi) artifacts for Fe-based alloy heat treatment cycles. The artifacts were pre-trained to filter out deep models that change the surrogate microstructure state after the heat treatment at ambient conditions. Direct representation of the thermal cycle features within knowledge Graph facilitated development of meaningful data models for microstructure evolution, which reduce overfitting to limited datasets.

36 MATERIALS SCIENCE↗
NASA NTRS2020

Customer Avionics Interface Development and Analysis Development Activity Tracking System

The Customer Avionics Interface Development and Analysis (CAIDA) Development Activity (DA)Tracking System is a Microsoft Access Database that tracks, organizes, and analyzes data about DAs from the work management tool. The CAIDA DA Tracking System takes data imported from the work management system. Once in the system, the data is filtered to generate each DA’s Asset. From there, many different queries are run on the data and their results are imported into forms to create graphs to get and display a wide range of metrics. These graphs are automatically updated with each new data import and over time. They can be easily exported for use in presentations, documents, etc. Additionally, the system is highly customizable and can be added upon to include more data members, generate new graphs, and much more.

Tara Conti↗
NASA NTRS1985

Accelerated stress testing of amorphous silicon solar cells

A technique for performing accelerated stress tests of large-area thin a-Si solar cells is presented. A computer-controlled short-interval test system employing low-cost ac-powered ELH illumination and a simulated a-Si reference cell (seven individually bandpass-filtered zero-biased crystalline PIN photodiodes) calibrated to the response of an a-Si control cell is described and illustrated with flow diagrams, drawings, and graphs. Preliminary results indicate that while most tests of a program developed for c-Si cells are applicable to a-Si cells, spurious degradation may appear in a-Si cells tested at temperatures above 130 C.

Stoddard, W. G.↗
NASA NTRS2012

Improving Spacecraft Data Visualization Using Splunk

EPOXI, like all spacecraft missions, receives large volumes of telemetry data from its spacecraft, DIF. It is extremely important for this data to be updated quickly and presented in a readable manner so that the flight team can monitor the status of the spacecraft. Existing DMD pages for monitoring spacecraft telemetry, while functional, are limited and do not take advantage of modern search technology. For instance, they only display current data points from instruments on the spacecraft and have limited graphing capabilities, making it difficult to see historical data. The DMD pages have fixed refresh rates so the team must often wait several minutes to see the most recent data, even after it is received on the ground. The pages are also rigid and require an investment of time and money to update. To more easily organize and visualize spacecraft telemetry, the EPOXI team has begun experimenting with Splunk, a commercially-available data mining system. Splunk can take data received from the spacecraft's different data channels, often in different formats, and index all the data into a common format. Splunk allows flight team members to search through the different data formats from a single interface and to filter results by time range and data field to make finding specific spacecraft events quick and easy. Furthermore, Splunk provides functions to create custom interfaces which help team members visualize the data in charts and graphs to show how the health of the spacecraft has changed over time.One of the goals of my internship with my mentor, Victor Hwang, was to develop new Splunk interfaces to replace the DMD pages and give the spacecraft team access to historical data and visualizations that were previously unavailable. The specific requirements of these pages are discussed in the next section.

spacecraft missions↗