Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Inference mechanisms”

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 145 records · Page 8
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Monitoring the propagation of mechanical discontinuity using data-driven causal discovery and supervised learning

Mechanical wave transmission through a material is influenced by the mechanical discontinuity in the material. The propagation of embedded discontinuities can be monitored by analyzing the wave-transmission measurements recorded by a multipoint sensor system placed on the surface of the material. The proposed workflow monitors the propagation of mechanical discontinuity through three stages, namely initial, intermediate, and final stages, by using supervised learning followed by data-driven causal discovery. To the end, the workflow processes the multipoint waveform measurements resulting from a single impulse source, while considering the effects of wave attenuation, dispersion and multiple wave-propagation modes due to the discontinuity and material boundaries. Among various feature reduction techniques ranging from decomposition methods to manifold approximation methods, the features derived based on statistical parameterizations of the measured waveforms lead to reliable monitoring that is robust to changes in precision, resolution, and signal-to-noise ratio of the multipoint sensor measurements. The numbers of zero-crossing, negative-turning, and positive turning in the waveforms are the strongest causal signatures of the propagation of mechanical discontinuity. Higher order moments of the waveforms, such as variance, skewness and kurtosis, are also strong causal signatures of the propagation. Finally, the newly discovered causal signatures confirm that the statistical correlations and conventional feature rankings are not always statistically significant indicators of causality.

42 ENGINEERING↗
NASA NTRS1984

The size of the lunar core

A 0.2 arc second phase shift in the 18.6 year forced precession of the lunar figure was inferred from analysis of lunar laser ranging data. The source of the phase shift is either viscous friction at a lunar core mantle boundary or solid friction caused by tidal flexing of the Moon by the Earth. Core mantle viscous coupling will explain the signature if the core radius R sub c approximately equals 330 km. On the other hand, solid friction can account for the signature only if the lunar solid friction 0 approximately equals 30 is abnormally small compared with that observed for, say, Mars (approximately equals 100 - 200). Although the inferred core radius is certainly within the limits imposed by the Apollo seismic experiment, it is significantly smaller than estimates of order 400-500 km from electromagnetic sounding. How accurate is the estimate of the R sub c derived from the phase shift? The effect of uncertainties in the frictional coupling mechanism, core density and ellipticity on the inferred core radius are discussed. The excitation of lunar free librations by core turbulence vis-a-vis other mechanisms (e.g., cometary or asteroidal impacts) and the influence of changes in lunar precession in the past on lunar dynamo generation are also examined.

Yoder, C. F.↗
DOE OSTI2023

Quasi-Newton Variational Bayes v.3.0

SAND2024-09052O Quasi-Newton Variational Bayes (QNVB) is a training algorithm that performs high-dimensional variational inference on machine learning models. It provides mechanisms to calibrate and control model uncertainty during training. The PyTorch implementation of projective integral updates for Gaussian mean-fields supports the manuscript, "Projective Integral Updates for High-Dimensional Variational Inference." Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Duersch, Jed↗
DOE OSTI2024

Verification, Validation, and Calibration Through a Causal Lens

While typical validation and verification approaches focus on identifying the associations between data elements using statistical and machine learning methods, the novel methods in this paper focus instead on identifying causal relationships between data elements. Statistical and machine-learning-based approaches are strictly data-driven, meaning that they provide quantitative comparison measures between data sets without explicitly considering the hypotheses behind them. This can lead to the erroneous conclusion that, if two data sets are close enough, the models that generated them are similar. In addition, when experimental and simulated data differ to an extent that fails to meet the acceptance criteria, calibration techniques are used to tweak simulation model parameters to reduce the gap between the two types of data. This produces the false expectation that a simulation model will match reality. The methods presented in this paper move away from these strictly data-driven methods for validation and calibration toward more robust, model-driven methods based on causal inference. Causal inference aims to identify the possible mechanisms that might have generated data. Thus, this analysis targets the prediction of the effects when one (or more) of the identified mechanisms are altered. There are many approaches to identify, quantify, and illustrate causal relationships. For the scope of this paper, directed graphs are employed as causal models. If the directed graph lacks cycles, it is known as a directed acyclic graph. A node in such a graph represents an observed data element while a directed edge connecting two nodes represents a causal relationship between two variables. The developed causal methods are designed to extract causal models from simulation models and experimental data. Causal models capture the causal relationships between data elements (e.g., simulated and experimental data). In this context, validation and verification are performed by comparing causal models. The proposed approach does not only inform system analysts on how a simulation model matches real-world data, but also identifies elements of the simulation model that should be revised when discrepancies between simulation and experimental data are observed. Through these causal methods, analysts can identify the portion of the model equation(s) that are behind an edge connecting two variables. Hence, once the structural differences between causal models have been determined, model calibration can occur by changing only those model parameters that impact the identified causal relationships.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

VALIDATION, VERIFICATION, AND CALIBRATION THROUGH A CAUSAL LENS

This paper presents an alternative method based on causal inference to perform validation, verification, and calibration of simulation models. While classical validation and verification approaches focus on the identification of the associations between data elements using statistical and machine learning methods, the novel methods in this paper focus instead on the identification of causal relationships between data elements. Statistical and machine learning-based approaches are strictly data-driven, meaning that they provide quantitative comparison measures between datasets without explicitly considering the hypotheses behind them. This can lead to the erroneous conclusion that, if two data sets are close enough, then the models that generated them are similar. In addition, when experimental and simulated data differ to an extent that fails to meet the acceptance criteria, calibration techniques are used to tweak simulation model parameters to reduce the gap between the two types of data. This produces the false expectation that a simulation model will match reality. The methods presented in this paper move away from these strictly data-driven methods for validation and calibration toward more robust, model-driven methods based on causal inference. Causal inference aims to identify the possible mechanisms that might have generated data. Thus, this analysis targets the prediction of the effects when one (or more) of the identified mechanisms are altered. There are many approaches to identify, quantify and illustrate causal relationships. For the scope of this paper, directed graphs are employed as causal models. If the directed graph lacks cycles it is known as a directed acyclic graph (DAG). A node in such a graph represents an observed data element while a directed edge connecting two nodes represents a causal relationship between two variables. The developed causal methods are designed to extract causal models from simulation models and from experimental data. Causal models capture the causal relationships between data elements (e.g., simulated and experimental data). In this context, validation and verification are performed by comparing causal models. The proposed approach does not only inform system analysts on how a simulation model matches real-world data, but also identifies elements of the simulation model that should be revised when discrepancies between simulation and experimental data are observed. Through these causal methods, analysts have a means to identify the portion of the model equation(s) that are behind an edge connecting two variables. Hence, once the structural differences between causal models have been determined, model calibration can occur by changing only those model parameters that impact the identified causal relationships.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Cellular, molecular, and therapeutic characterization of pilocarpine-induced temporal lobe epilepsy

Animal models have expanded our understanding of temporal lobe epilepsy (TLE). However, translating these to cell-specific druggable hypotheses is not explored. Herein, we conducted an integrative insilico-analysis of an available transcriptomics dataset obtained from animals with pilocarpine-induced-TLE. A set of 119 genes with subtle-to-moderate impact predicted most forms of epilepsy with ~ 97% accuracy and characteristically mapped to upregulated homeostatic and downregulated synaptic pathways. The deconvolution of cellular proportions revealed opposing changes in diverse cell types. The proportion of nonneuronal cells increased whereas that of interneurons, except for those expressing vasoactive intestinal peptide (Vip), decreased, and pyramidal neurons of the cornu-ammonis (CA) subfields showed the highest variation in proportion. A probabilistic Bayesian-network demonstrated an aberrant and oscillating physiological interaction between nonneuronal cells involved in the blood–brain-barrier and Vip interneurons in driving seizures, and their role was evaluated insilico using transcriptomic changes induced by valproic-acid, which showed opposing effects in the two cell-types. Additionally, we revealed novel epileptic and antiepileptic mechanisms and predicted drugs using causal inference, outperforming the present drug repurposing approaches. These well-powered findings not only expand the understanding of TLE and seizure oscillation, but also provide predictive biomarkers of epilepsy, cellular and causal micro-circuitry changes associated with it, and a drug-discovery method focusing on these events.

60 APPLIED LIFE SCIENCES↗
NASA NTRS1975

Investigation of scrubbing and impingement noise

Tests were conducted in an acoustic wind tunnel to determine surface pressure spectra and far field noise caused by turbulence impinging on an airfoil and turbulence convected past a sharp trailing edge. Measured effects of flow velocity and turbulence intensity were compared with predictions from several theories. Also, tests were conducted in an anechoic chamber to determine surface pressure spectra and far field noise caused by a deflected airfoil scrubbed by a subsonic jet. This installation simulated both an under-the-wing and an upper-surface-blowing externally blown flap, depending on the deflection angle. Surface and far field spectra, and cross correlation coherence and delay time, were utilized to infer the major noise-producing mechanisms.

Fink, M. R.↗
NASA NTRS1979

Airframe noise component interaction studies

Acoustic wind tunnel tests were conducted to examine the noise-generating processes of an airframe during approach flight. The airframe model was a two-dimensional wing section, to which highlift leading and trailing edge devices and landing gear could be added. Far field conventional microphones were utilized to determine component spectrum levels. An acoustic mirror directional microphone was utilized to examine noise source distributions on airframe components extended separately and in combination. Measured quantities are compared with predictions inferred from aircraft flyover data. Aeroacoustic mechanisms for each airframe component are identified. Component interaction effects on total radiated noise generally were small (within about 2 dB). However, some interactions significantly redistributed the local noise source strengths by changing local flow velocities and turbulence levels. Possibilities for noise reduction exist if trailing edge flaps could be modified to decrease their noise radiation caused by incident turbulent flow.

Fink, M. R.↗
NASA NTRS1979

Particle acceleration by propagating interplanetary shocks

The process of charged particle acceleration in interplanetary shocks has been simulated for typical parameters. Since space probe observations of charged particle fluxes are the principle means of inferring the action of an acceleration mechanism, the simulation was designed to follow particles backwards in time from a given observing point, usually 1 AU. This allowed a full diagnosis of which observed particles had interacted with an oncoming shock, where the interacting occurred, and by how much the energy had been changed by the shock interaction. Simple assumptions about the preshock energy spectrum allow the construction of a full temporal profile of the expected intensity, anisotropy, and energy spectrum. The simulations are apparently capable of reproducing the main features of the less than 10 MeV/nuc ion flux enhancements observed at interplanetary shocks using only a laminar interplanetary magnetic field, Archimidean spiral, and a spherical oblique shock with typical speed, strength, and shock normal-magnetic field angle.

Armstrong, T. P.↗
NASA NTRS1990

Surface characterizations of color threshold

The paper evaluates how well three different parametric shapes, ellipsoids, rectangles, and parallelograms, serve as models of three-dimentional detection contours. The constraints of the procedures for deriving the best-fitting shapes on inferences about the theoretical visual detection mechanisms are described. Results of two statistical tests show that only the parallelogram fits the data with more precision than the variance in repeated threshold measurements, and thus provides a slightly better fit than the other two shapes. Nevertheless it does not serve as a better guide than the ellipsoidal model for interpolating from the measurements to thresholds in novel color directions.

Poirson, Allen B.↗
NASA NTRS1990

Building distributed rule-based systems using the AI Bus

The AI Bus software architecture was designed to support the construction of large-scale, production-quality applications in areas of high technology flux, running heterogeneous distributed environments, utilizing a mix of knowledge-based and conventional components. These goals led to its current development as a layered, object-oriented library for cooperative systems. This paper describes the concepts and design of the AI Bus and its implementation status as a library of reusable and customizable objects, structured by layers from operating system interfaces up to high-level knowledge-based agents. Each agent is a semi-autonomous process with specialized expertise, and consists of a number of knowledge sources (a knowledge base and inference engine). Inter-agent communication mechanisms are based on blackboards and Actors-style acquaintances. As a conservative first implementation, we used C++ on top of Unix, and wrapped an embedded Clips with methods for the knowledge source class. This involved designing standard protocols for communication and functions which use these protocols in rules. Embedding several CLIPS objects within a single process was an unexpected problem because of global variables, whose solution involved constructing and recompiling a C++ version of CLIPS. We are currently working on a more radical approach to incorporating CLIPS, by separating out its pattern matcher, rule and fact representations and other components as true object oriented modules.

Schultz, Roger D.↗
NASA NTRS2004

Implications of Observed High Supersaturation for TTL Cloud Formation and Dehydration

In situ measurements of water vapor concentration made during the CRYSTAL-FACE and Pre-AVE missions indicate higher than expected supersaturations in both clear and cloudy air near the cold tropical tropopause: (1) steady-state ice supersaturations of 20-30% were measured within cirrus at T < 200 K; (2) supersaturations exceeding 100% (near water saturation) were observed under cloud-free conditions near 187 K. The in-cloud measurements challenge the conventional belief that any water vapor in excess of ice saturation should be depleted by crystal growth given sufficient time. The high clear-sky supersaturations imply that thresholds for ice nucleation due to homogeneous freezing of aerosols (or any other mechanism) are much higher than those inferred from laboratory measurements. We will use simulations of Tropical Tropopause Layer (TTL) transport and cloud formation throughout the tropics to show that these effects have important implications for TTL cloud frequency and freeze-drying of air crossing the tropical tropopause cold trap.

Jensen, Eric↗
NASA NTRS2003

Obliquity, Ice Sheets, and Layered Sediments on Mars: What Spacecraft Observations and Climate Models are Telling Us

The Mars Odyssey Gamma-Ray Spectrometer (GRS) data present a quandary: On the one hand, large deposits of (inferred) water ice are located where thermal models suggest they would form and best be protected, e.g., if deposited during periods of higher obliquity. On the other hand, the volume mixing ratios (approx. 70%) are so high that diffusive deposition of water in regolith pore space (which is the process assumed by these models) cannot be the primary formation mechanism. Furthermore, given that the water is inferred to be so close to the surface (less than a few 10's of cm's), it must be in communication with the atmosphere on time scales that are geologically relatively short (10(exp 3)-10(exp 6) years); therefore the water cannot be archaic. Considering the GRS data, images of mantled, fretted, and disaggregated terrain, and new climate modeling of Mars orbital cycles, we are led to an alternate conclusion about the ice deposits: that they form as subaerial ice sheets. This scenario not only provides a simple explanation for these observations, but may also help explain the formation of globally distributed, sedimentary layered deposits.

Richardson, M. I.↗
DOE OSTI2023

Iterative multi-task learning and inference from seismic images

Seismic interpretation aims to extract quantitative and interpretable attributes from a seismic image produced using some migration method to inform characteristics of a subsurface reservoir or target of interest. Current paradigms for computing seismic attributes mostly rely on single-task algorithms. We develop an iterative, multi-task machine learning method to learn and infer multiple attributes from a seismic image. This method is composed of two stages: a multi-task inference stage and a multi-modal, multi-task refinement stage. The basic mechanism of this method is that we train a multi-task inference neural network (NN) to estimate a set of attributes, including a relative geological time (RGT), a denoised higher-resolution (DHR) seismic image, and multiple fault attributes (including probability, dip, and strike), from a low-resolution, noisy seismic image; then we input the inferred attributes to a multi-task refinement NN to enhance the raw inference results iteratively. The two multi-task NNs are trained separately based on synthetic seismic images and associated attributes generated by a geological modeling algorithm. The software we intend to release is a PyTorch implementation of this multi-task learning method for both 2D and 3D cases along with scripts to run the training/validation. The algorithm and software can be a useful tool for automatic seismic interpretation.

Gao, Kai↗
NASA NTRSDate not supplied

In Vitro Selections with RNAs of Variable Length Converge on a Robust Catalytic Core

In vitro selection is a powerful tool that can be used to understand basic principles of molecular evolution. We used in vitro selection to understand how changes in length and the accumulation of point mutations enable the evolution of functional RNAs. Using RNA populations of various lengths, we performed a series of in vitro experiments to select for ribozymes with RNA ligase activity. We identified a core ribozyme structure that was robust to changes in RNA length, high levels of mutagenesis, and increased selection pressure. Elaboration on this core structure resulted in improved activity which we show is consistent with a larger trend among functional RNAs in which increasing motif size can lead to an exponential improvement in fitness. We conclude that elaboration on conserved core structures is a preferred mechanism in RNA evolution. This conclusion, drawn from selections of RNAs from random sequences, is consistent with proposed evolutionary histories of specific biological RNAs. More generally, our results indicate that modern RNA structures can be used to infer ancestral structures. Our observations also suggest a mechanism by which structural outcomes of early RNA evolution would be largely reproducible even though RNA fitness landscapes consist of disconnected clusters of functional sequences.

In vitro evolution↗
NASA NTRS2014

Inferred Variable FeO Content in Medium-sized Lunar Pyroclastic Deposits from LRO Diviner Data

Lunar pyroclastic deposits (LPDs) are low albedo features that mantle underlying terrain (Gaddis et al. 1985). They are high priority targets for science and exploration as they are believed to originate from and therefore reflect the composition of the deep lunar interior (NRC, 2011). They are also the best potential resource of oxygen out of any Apollo samples (Allen et al. 1996). Historically, LPDs have been divided into regional versus local categories (Gaddis et al. 2003). The large (>1000 km2 area) regional deposits are deeply sourced (>400 km deep) and result from fire fountaining. Small (<1000 km2) local deposits are thought to result from Vulcanian eruptions in which magma is slowly emplaced beneath the surface until enough volatiles exsolve and the high pressure causes an explosion. Bennett et al. (2013) identified a local deposit (674 km2 area) that may have resulted from both Vulcanian activity and fire fountaining. This deposit potentially represents a new intermediate class of LPDs that straddles the interface between the two formation mechanisms. The deposit also exhibits the highest inferred FeO wt.% of any known lunar glass. In this work we investigate the inferred FeO abundances of other medium-sized deposits to characterize this potential new class of deposits and understand the magnitude of variations in inferred FeO among pyroclastic deposits. We use the method of Greenhagen et al. (2010) to calculate the wavelength of the Christiansen Feature (CF) from Lunar Reconnaissance Orbiter Diviner Lunar Radiometer instrument thermal-infrared observations for four medium-sized deposits. From the CF values, we estimate each deposit's FeO abundance using the method of Allen et al. (2012). The four LPDs that we examined (Oppenheimer South, Beer, Cleomedes, and J. Herschel) all have average CF values from 8.22-8.28 microns, corresponding to FeO abundances of approx. 10-15 wt.%. All of these values are within the range and uncertainties of FeO abundances measured in Apollo samples. As previously identified, the Oppenheimer South deposit exhibits an area of enhanced CF values (8.49 microns) that, if the methods of Allen et al. (2012) can be extrapolated, correspond to a highest observed approx. 30 wt.% FeO. Moon Mineralogy Mapper near-infrared spectra indicate that this area is glass-rich as opposed to olivine-rich. While we are still investigating the nature of the high CF wavelength in Oppenheimer South, spatially-resolved observations there and (to a smaller degree) in our other study sites, shows that FeO wt.% can vary within LPDs. Thus, obtaining only the average FeO abundance over a large area may not be adequate to understand global variation. The magnitude of Oppenheimer South's CF variability, if due to actual surface variations rather than calibration artifacts or spectral mixing, could indicate that it is a unique deposit and not part of a new mid-sized class of deposits. The higher value could be a result of its location within the South Pole Aitken Basin and exsolution of more deeply sourced magma due to the thin crust there.

Bennett, K. A.↗
NASA NTRS1971

A case for catastrophic early degassing of the earth.

Evidence is given for the occurrence in early earth history of massive escape of volatiles from the interior of the earth to its surface, which must have had catastrophic proportions. The mechanism and time of this event are inferred from the abundances and distribution of volatiles in the atmosphere, hydrosphere and crust, and from meteorite studies, and geophysical and astronomical observations. This event is linked to the melting of the earth during accretion which resulted in the concentration of U, Th, K and Pb in the outermost portion of the earth and in virtually complete expulsion of nonradiogenic Ar, Ne and Kr from its interior.

Fanale, F. P.↗