Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Network inference”

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 577 records · Page 32
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Tokamak divertor plasma emulation with machine learning

Abstract Future tokamak devices that aim to create conditions relevant to power plant operations must consider strategies for mitigating damage to plasma facing components in the divertor. One of the goals of MAST-U tokamak operations is to inform these considerations by researching advanced divertor configurations that aid stable plasma detachment. Machine design, scenario planning and detachment control would all greatly benefit from tools that enable rapid calculation of scenario-relevant quantities given some input parameters. This paper presents a method for generating large, simulated scrape-off layer data sets, which was applied to generate a data set of steady-state Hermes-3 simulations of the MAST-U tokamak. A machine learning model was constructed using a Bayesian approach to hyperparameter optimisation to predict diagnosable output quantities given control-relevant input features. The resulting best-performing model, which is based on a feedforward neural network, achieves high accuracy when predicting electron temperature at the divertor target and carbon impurity radiation front position and runs in around 1 ms in inference mode. Techniques for interpreting the predictions made by the model were applied, and a high-resolution parameter scan of upstream conditions was performed to demonstrate the utility of rapidly generating accurate predictions using the emulator. This work represents a step forward in the design of machine learning-driven emulators of tokamak exhaust simulation codes in operational modes relevant to divertor detachment control and plasma scenario design.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRS2003

A Comparison of Solar p-Mode Parameters from MDI and Gong: Mode Frequencies and Structure Inversions

Helioseismic analysis of solar global oscillations allows investigation of the internal structure of the Sun. One important test of the reliability of the inferences from helioseismology is that the results from independent sets of contemporaneous data are consistent with one another. Here we compare mode frequencies from the Global Oscillation Network Group and Michelson Doppler Imager on board SOHO and resulting inversion results on the Sun's internal structure. The average relative differences between the data sets are typically less than 1 x 10(exp -5) substantially smaller than the formal errors in the differences; however, in some cases the frequency differences show a systematic behavior that might nonetheless influence the inversion results. We find that the differences in frequencies are not a result of instrumental effects but are almost entirely related to the data pipeline software. Inversion of the frequencies shows that their differences do not result in any significant effects on the resulting inferences on solar structure. We have also experimented with fitting asymmetric profiles to the oscillation power spectra and find that, compared with the symmetric fits, this causes no significant change in the inversion results.

Basu, S.↗
DOE OSTI2020

Effects of social structure and management on risk of disease establishment in wild pigs

Contact heterogeneity among hosts determines invasion and spreading dynamics of infectious disease, thus its characterization is essential for identifying effective disease control strategies. Yet, little is known about the factors shaping contact networks in many 28 wildlife species and how wildlife management actions might affect contact networks. Wild pigs in North America are an invasive, socially-structured species that pose a health concern for domestic swine given their ability to transmit numerous devastating diseases 2 such as African swine fever (ASF). Using proximity loggers and GPS data from 48 wild pigs in Florida and South Carolina, USA, we employed a probabilistic framework to estimate weighted contact networks. We determined the effects of sex, social group, and spatial distribution (monthly home range overlap and distance) on wild pig contact. We also estimated the impacts of management-induced perturbations on contact and inferred their effects on ASF establishment in wild pigs with simulation. Social group membership was the primary factor influencing contacts. Between-group contacts depended primarily on space use characteristics, with fewer contacts among groups separated by >2 km and no contacts among groups >4 km apart within a month. Modeling ASF dynamics on the contact network demonstrated that indirect contacts resulting from baiting (a typical method of attracting wild pigs or game species to a site to enhance recreational hunting) increased the risk of disease establishment by ~33% relative to direct contact. Low-intensity population reduction (<5.9% of the population) had no detectable impact on contact structure but reduced predicted ASF establishment risk relative to no population reduction. Here, we demonstrate an approach for understanding the relative role of spatial, social, and individual-level characteristics in shaping contact networks and predicting their effects on disease establishment risk, thus providing insight for optimizing disease control in spatially- and socially-structured wildlife species.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

Solving Inverse Stochastic Problems from Discrete Particle Observations Using the Fokker--Planck Equation and Physics-Informed Neural Networks

The Fokker--Planck (FP) equation governing the evolution of the probability density function (PDF) is applicable to many disciplines, but it requires specification of the coefficients for each case, which can be functions of space-time and not just constants and hence require the development of a data-driven modeling approach. When the data available is directly on the PDF, there exist methods for inverse problems that can be employed to infer the coefficients and thus determine the FP equation and subsequently obtain its solution. Herein, we address a more realistic scenario, where only sparse data are given on the particles' positions at a few time instants, which are not sufficient to accurately construct directly the PDF even at those times from existing methods, e.g., kernel estimation algorithms. To this end, we develop a general framework based on physics-informed neural networks (PINNs) that introduces a new loss function using the Kullback--Leibler divergence to connect the stochastic samples with the FP equation to simultaneously learn the equation and infer the multidimensional PDF at all times. In particular, we consider two types of inverse problems, type I, where the FP equation is known but the initial PDF is unknown, and type II, in which, in addition to the unknown initial PDF, the drift and diffusion terms are also unknown. In both cases, we investigate problems with either Brownian or Lévy noise or a combination of both. Here, we demonstrate the new PINN framework in detail in the one-dimensional (1D) case, but we also provide results for up to five dimensions demonstrating that we can infer both the FP equation and dynamics simultaneously at all times with high accuracy using only very few discrete observations of the particles.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2008

Algorithm Optimally Orders Forward-Chaining Inference Rules

People typically develop knowledge bases in a somewhat ad hoc manner by incrementally adding rules with no specific organization. This often results in a very inefficient execution of those rules since they are so often order sensitive. This is relevant to tasks like Deep Space Network in that it allows the knowledge base to be incrementally developed and have it automatically ordered for efficiency. Although data flow analysis was first developed for use in compilers for producing optimal code sequences, its usefulness is now recognized in many software systems including knowledge-based systems. However, this approach for exhaustively computing data-flow information cannot directly be applied to inference systems because of the ubiquitous execution of the rules. An algorithm is presented that efficiently performs a complete producer/consumer analysis for each antecedent and consequence clause in a knowledge base to optimally order the rules to minimize inference cycles. An algorithm was developed that optimally orders a knowledge base composed of forwarding chaining inference rules such that independent inference cycle executions are minimized, thus, resulting in significantly faster execution. This algorithm was integrated into the JPL tool Spacecraft Health Inference Engine (SHINE) for verification and it resulted in a significant reduction in inference cycles for what was previously considered an ordered knowledge base. For a knowledge base that is completely unordered, then the improvement is much greater.

James, Mark↗
NASA NTRS1981

The state of the atmosphere as inferred from the FGGE satellite observing systems during SOP-1

The statistical properties, and coverage, of satellite temperature sounding data are described. Tropical regions are observed every two days, extratropics from one to four times a day. Oceans are covered two to three times a day. Asynoptic coverage is comparable to the U.S. rawinsonde network twice daily coverage. Lack of ground truth for data sparse areas makes accuracy difficult to assess. The rms differences of layer mean temperatures obtained from collocating rawinsonde observations with satellite temperature profiles in space and time differ from rms differences of layer mean satellite temperature soundings. The FGGE satellite systems can infer the three dimensional motion field and improve the representation of the large scale state of the atmosphere.

Halem, M.↗
NASA NTRS2016

Variation in Global Chemical Composition of PM2.5: Emerging Results from SPARTAN

The Surface PARTiculate mAtter Network (SPARTAN) is a long-term project that includes characterization of chemical and physical attributes of aerosols from filter samples collected worldwide. This paper discusses the ongoing efforts of SPARTAN to define and quantify major ions and trace metals found in fine particulate matter (PM (sub 2.5). Our methods infer the spatial and temporal variability of PM (sub 2.5) in a cost-effective manner. Gravimetrically weighed filters represent multi-day averages of PM (sub 2.5), with a collocated nephelometer sampling air continuously. SPARTAN instruments are paired with AErosol RObotic NETwork (AERONET) sun photometers to better understand the relationship between ground-level PM (sub 2.5) and columnar aerosol optical depth (AOD). We have examined the chemical composition of PM (sub 2.5) at 12 globally dispersed, densely populated urban locations and a site at Mammoth Cave (US) National Park used as a background comparison. So far, each SPARTAN location has been active between the years 2013 and 2016 over periods of 2-26 months, with an average period of 12 months per site. These sites have collectively gathered over 10 years of quality aerosol data. The major PM (sub 2.5) constituents across all sites (relative contribution plus or minus Standard Deviation) are ammoniated sulfate (20 percent plus or minus 11 percent), crustal material (13.4 percent plus or minus 9.9 percent), equivalent black carbon (11.9 percent plus or minus 8.4 percent), ammonium nitrate (4.7 percent plus or minus 3.0 percent), sea salt (2.3 percent plus or minus 1.6 percent), trace element oxides (1.0 percent plus or minus 1.1 percent), water (7.2 percent plus or minus 3.3 percent) at 35 percent relative humidity, and residual matter (40 percent plus or minus 24 percent). Analysis of filter samples reveals that several PM (sub 2.5) chemical components varied by more than an order of magnitude between sites. Ammoniated sulfate ranges from 1.1 microns per cubic meter (Buenos Aires, Argentina) to 17 microns per cubic meter (Kanpur, India in the dry season). Ammonium nitrate ranged from 0.2 microns per cubic meter (Mammoth Cave, in summer) to 6.8 microns per cubic meter (Kanpur, dry season). Equivalent black carbon ranged from 0.7 microns per cubic meter (Mammoth Cave) to over 8 microns per cubic meter (Dhaka, Bangladesh and Kanpur, India). Comparison of SPARTAN vs. coincident measurements from the Interagency Monitoring of Protected Visual Environments (IMPROVE) network at Mammoth Cave yielded a high degree of consistency for daily PM (sub 2.5) (r squared equals 0.76, slope equals 1.12), daily sulfate (r squared equals 0.86, slope equals 1.03), and mean fractions of all major PM (sub 2.5) components (within 6 percent). Major ions generally agree well with previous studies at the same urban locations (e.g. sulfate fractions agree within 4 percent for 8 out of 11 collocation comparisons). Enhanced anthropogenic dust fractions in large urban areas (e.g. Singapore, Kanpur, Hanoi, and Dhaka) are apparent from high Zn to Al ratios. The expected water contribution to aerosols is calculated via the hygroscopicity parameter kappa (sub v (volume)) for each filter. Mean aggregate values ranged from 0.15 (Ilorin) to 0.28 (Rehovot). The all-site parameter mean is 0.20 plus or minus 0.04. Chemical composition and water retention in each filter measurement allows inference of hourly PM (sub 2.5) at 35 percent relative humidity by merging with nephelometer measurements. These hourly PM (sub 2.5) estimates compare favourably with a beta attenuation monitor (MetOne) at the nearby US embassy in Beijing, with a coefficient of variation r squared equals 0.67 (number equals 3167), compared to r squared equals 0.62 when v (volume) was not considered. SPARTAN continues to provide an open-access database of PM (sub 2.5) compositional filter information and hourly mass collected from a global federation of instruments.

Snider, Graydon↗
DOE OSTI2021

Autonomous nondestructive evaluation of resistance spot welded joints

The application of non-destructive evaluation approaches has attracted strong interests in modern automotive industries. Here, we present an autonomous deep-computing framework to analyze raw videos from infrared systems and to predict weld nugget shape and size with unprecedented accuracy and speed. In a comprehensive training and testing experiment with 90 videos (seven sets of welding material stack-ups), a new method was developed to assemble sufficient datasets for neural network training. Our framework successfully predicts all the nugget shapes with F1 scores that range from 0.84 to 0.92. The total training time on Nvidia DGX station takes less than 10 min for each set of welding material stack-up. The real inference time of an individual dataset (with 30 video frames) takes about 0.005 s. The procedure and methods developed in the study can be applied to other image-based weld property prediction, as well as other manufacturing processes. Furthermore, our well-trained neural networks take limited memory resources (2.3 MB) and are suitable for embedded microprocessors for in-situ welding quality control as edge computing within an intelligent welding framework.

42 ENGINEERING↗
DOE OSTI2020

Radiation Source Localization Using Surrogate Models Constructed from 3-D Monte Carlo Transport Physics Simulations

Recent research has focused on the development of surrogate models for radiation source localization in a simulated urban domain. We employ the Monte Carlo N-Particle (MCNP) code to provide high- delity simulations of radiation transport within an urban domain. The model is constructed to employ a source location (x, y, z) as input and return the estimated count rate for a set of speci ed detector locations. Because MCNP simulations are computationally expensive, we develop e cient and accurate surrogate models of the detector responses. We construct surrogate models using Gaussian processes (GP) and neural networks (NN) that we train and verify using the MCNP simulations. The trained surrogate models provide an e cient framework for Bayesian inference and experimental design. We employ Delayed Rejection Adaptive Metropolis (DRAM), a Markov Chain Monte Carlo (MCMC) algorithm, to infer the location and intensity of an unknown source. The DRAM results yield a posterior probability distribution for the source's location conditioned on the observed detector count rates. The posterior distribution exhibits regions of high and low probability within the simulated environment identifying potential source locations. In this manner, we can quantify the source location to within at least one of these regions of high probability in the considered cases. Employing these methods, we are able to reduce the space of potential source locations by at least 60%.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2023

Deducing neutron star equation of state parameters directly from telescope spectra with uncertainty-aware machine learning

Neutron stars provide a unique laboratory for studying matter at extreme pressures and densities. While there is no direct way to explore their interior structure, X-rays emitted from these stars can indirectly provide clues to the equation of state (EOS) of the superdense nuclear matter through the inference of the star's mass and radius. However, inference of EOS directly from a star's X-ray spectra is extremely challenging and is complicated by systematic uncertainties. The current state of the art is to use simulation-based likelihoods in a piece-wise method which relies on certain theoretical assumptions and simplifications about the uncertainties. It first infers the star's mass and radius to reduce the dimensionality of the problem, and from those quantities infer the EOS. Here we demonstrate a series of enhancements to the state of the art, in terms of realistic uncertainty quantification and a path towards circumventing the need for theoretical assumptions to infer physical properties with machine learning. We also demonstrate novel inference of the EOS directly from the high-dimensional spectra of observed stars, avoiding the intermediate mass-radius step. Our network is conditioned on the sources of uncertainty of each star, allowing for natural and complete propagation of uncertainties to the EOS.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Hindered segmental dynamics in associative protein hydrogels studied by neutron spin-echo spectroscopy

Transient binding between associating macromolecules can cause qualitative changes to chain dynamics, including modes of conformational relaxation and diffusion, through tethering effects imparted by long-range connectivity. Here, in this work, the role of binding on short-time segmental dynamics in associative polymer gels is investigated by neutron spin-echo (NSE) measurements on a class of model artificial coiled-coil proteins with a systematically varied architecture, probing timescales of 0.1–130 ns, and length scales close to the molecular radius of gyration. The results illustrate effects of transient cross-linking on chain dynamics on different timescales, manifested in changes in segmental relaxation behavior with variations in strand length, chain concentration, and sticker distribution (endblock- vs midblock-functionalized). In all gels, a short-time cooperative diffusion mode is seen over all wave vectors, analogous to a semidilute solution, with no transitions seen at any known structural length scale. However, the diffusion coefficients are found to decrease with increasing junction density across all gels, with the strand length and number of stickers per chain in each gel appearing to play a relatively minor role. The slowing of cooperative diffusion with junction density contrasts with classical predictions of a greater restoring force for fluctuation dissipation due to the increased elasticity, suggesting additional effects of the coiled-coil junctions such as an enhancement in local viscosity that slows dynamics. Notably, the relaxation rates for all gels can be rescaled by the interjunction spacing inferred from small-angle neutron scattering, where they collapse onto a master curve suggestive of self-similar dynamics even in networks with different strand lengths and chain architectures. On long timescales (but shorter than the junction exchange time), a slowing of network relaxation is observed, resulting in a nondecaying plateau in the spin-echo amplitude attributed to a freezing of chain dynamics due to tethering. A characteristic length scale corresponding to the extent of dynamic fluctuations is estimated for each gel, which appears to be smaller than the interjunction spacing but similar to the correlation blob size of the overlapping strands. The results indicate an important role of transient binding on molecular-scale dynamics in associative polymer gels, even on timescales shorter than the junction exchange time, in addition to its effects on long-range self-diffusion previously observed.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Improving Subsurface Stress Characterization for Carbon Dioxide Storage Projects

Researchers at New Mexico Tech, Los Alamos National Laboratory (LANL), Sandia National Laboratory (SNL), and other collaborators are developing a methodology to improve characterization of stress in the subsurface by means of a model-based inversion of reservoir engineering data, time lapse seismic measurements, and microseismicity. Historically, various geophysical techniques have been used in efforts to understand the state of stress in the subsurface through direct and indirect imaging of stress sensitive features (faults and fractures) and observations of transient stress related observations (time variant elastic moduli and microseismicity). Standard direct and indirect techniques for seismic fault and fracture imaging suffer from detectability limits due to reliance on high data quality and multiplicity. Variations in effective elastic properties from inversion of high quality seismic (Vertical Seismic Profile (VSP)) data may be used to infer stress near the wellbore through integration with independent experimental characterization of the stress-velocity relationship. Given a sufficiently robust observation network, microseismic emissions may be inverted to characterize focal mechanisms which, together with supporting assumptions and constraints, inform estimates of in-situ stress. While each of these techniques contributes in part to the characterization of stress within limited spatial and temporal domains, no one method provides an unambiguous stress measurement or a predictive capability over a site scale spatial extent. Challenges associated with solution non-uniqueness, measurement ambiguity, and irregular sampling may be greatly minimized through combination of one or more of independent measurements within a common framework in which realistic geological, hydrodynamic, geomechanical, and seismological constituent process models may act as constraints.

02 PETROLEUM↗
NASA NTRS2003

Maximum Rates of Olivine Dissolution on Mars

This work is a first step in constraining olivine alteration rates on Mars. By bounding dissolution rates of olivine, we can make inferences about the temporal extent of aqueous alteration on the surface. Several hypothesized low-temperature surface aqueous alteration processes include valley network formation from release of groundwater during impact or gully formation from snow melt, both of which may occur over a period of several thousand years. Outflow channels may have formed during brief floods lasting less than a year. If large bodies of water were present during a warm, wet period, large standing bodies of water may have been present for 104 to 109 years. High-temperature alteration processes include hydrothermal activity associated with impacts and magmatic features. These features may remain active for 104 to 106 years. Table 2 summarizes speculated duration times of these aqueous processes. For comparison, we have determined minimum olivine residence times (time until complete dissolution) for 1 mm diameter particles to range from much less than one year to approximately 5,000 years.

J D Stopar↗
NASA NTRS2018

Tracer Transport Differences: Challenges and Implications for Flux Inversions

This poster demonstrates that different chemistry transport models (CTMs), each extensively validated, can have significant differences in the predicted transport of long-lived trace gases. For carbon dioxide (CO2), this difference is 0.5 ppm or greater in the total column. The value exceeds the nominal retrieval error requirements of the Total Carbon Column Observing Network (TCCON) and the Orbiting Carbon Observatory 2 (OCO-2), which were chosen based on the understanding of the accuracy necessary to infer surface fluxes of CO2 on regional and seasonal scales. This suggests that the transport errors of CTMs play a considerable role in the surface flux inversion of satellite-based measurements of long-lived trace gases and the interpretation of the inferred fluxes requires a careful understanding of this role.

Weir, B.↗
NASA NTRS2019

Diurnal Cycles in SST: Coupled Data Assimilation and Future Observational Requirements

Most operational centers are developing coupled (atmosphere-ocean) data assimilation systems as an alternative to uncoupled counterparts (atmosphere- or ocean-only). The Sea Surface Temperature (SST) is one of the key variables that tightly connects the atmosphere and ocean states and also air-sea fluxes. However, current prototype coupled data assimilation systems rely on external (L4) gridded SST or along-track (L3 or L2) SST retrievals as observed data or relaxation field. But in reality, SST are measurements are available from sparse in-situ network of ships, moorings, and buoys; all of them combined together are far less than those from satellites. However, satellites do not directly measure temperature, and inferring SST from satellite measured radiances requires a radiative transfer model, its calibration and also bias correction.The NASA Global Modeling and Assimilation Office (GMAO) is developing a coupled data assimilation system which assimilates SST directly from the raw observations, i.e., satellite radiances and in-situ observations. The methodology to directly assimilate radiances for SST became operational in Jan, 2017 in the GMAO's near-real time Weather Analysis and Prediction System. There were many modifications to the GMAO system in order to implement SST assimilation, most of which generally improved the predictability of the system. In order to maintain and further improve this system, we advocate for the availability of a microwave satellite radiometer in future beyond the currently operational GPM- GMI and AMSR-2 missions. For improved modeling of the near-surface temperature, salinity and mixing processes, we suggest adding more than one temperature sensor and salinity sensors to the drifting buoy network.

Akella, Santha↗
DOE OSTI2021

Identifiability and predictability of integer- and fractional-order epidemiological models using physics-informed neural networks

Here we analyze a plurality of epidemiological models through the lens of physics-informed neural networks (PINNs) that enable us to identify time-dependent parameters and data-driven fractional differential operators. In particular, we consider several variations of the classical susceptible-infectious-removed (SIR) model by introducing more compartments and fractional-order and time-delay models. We report the results for the spread of COVID-19 in New York City, Rhode Island and Michigan states and Italy, by simultaneously inferring the unknown parameters and the unobserved dynamics. For integer-order and time-delay models, we fit the available data by identifying time-dependent parameters, which are represented by neural networks. In contrast, for fractional differential models, we fit the data by determining different time-dependent derivative orders for each compartment, which we represent by neural networks. We investigate the structural and practical identifiability of these unknown functions for different datasets, and quantify the uncertainty associated with neural networks and with control measures in forecasting the pandemic.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

Simple and sophisticated models of Taylor’s cylinder impact

We present a study based on the simplest possible models for Taylor’s cylinder impact problem, in addition to examina-tion of using convolutional neural networks (CNNs) to map cylinder profiles to strength calibrations. We find that the approximate treatments of Taylor and Hawkyard compare well with hydrodynamic simulations using an equivalent assumption of constant flow stress. The CNN models prove to be well suited to successfully infer parameterizations of the Preston-Tonks-Wallace model of plastic deformation based on the deformed profile of an impacted cylinder.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Scaling Inference Using Triton to Accelerate Particle Physics at the LHC and DUNE

DUNE and the LHC experiments consist of unique and cutting-edge particle detectors that create massive, complex, and rich datasets with billions of events. They require sophisticated algorithms to reconstruct and interpret the data. Modern machine learning algorithms provide a powerful toolset to detect and classify particles, from familiar image processing convolutional neural networks to newer graph neural network architectures. A full reconstruction of these particle collisions requires novel approaches to handle the computing challenge of processing so much raw data. In a series of studies, physicists from Fermilab, CERN, and university groups explored how to accelerate their data processing using the Triton Inference Server.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗