Search NASA⌕ Search

SEARCH · Search NASA

Results for “Gaussian Process Models”

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 1,171 records · Page 65

Effect of Measurement vs. Counting Errors on Parameters' Covariance in Neutron Tomography Analysis

We present here a method that estimates the relative effect of the counting uncertainty and of the instrument uncertainty on that of the parameters in a parametric model for neutron time of flight. The final result, obtained independently of calculation of the parameter values from measured data, presents explicitly the ratio of the two uncertainties in terms of the choice, settings, and placement of the detector and the oscilloscope. Consequently, the method can serve as a tool in planning a measurement setup.

fusion↗

Approach to the unfolding and folding dynamics of add A-riboswitch upon adenine dissociation using a coarse-grained elastic network model

Riboswitches are noncoding mRNA segments that can regulate the gene expression via altering their structures in response to specific metabolite binding. We proposed a coarse-grained Gaussian network model (GNM) to examine the unfolding and folding dynamics of adenosine deaminase (add) A-riboswitch upon the adenine dissociation, in which the RNA is modeled by a nucleotide chain with interaction networks formed by connecting adjoining atomic contacts. It was shown that the adenine binding is critical to the folding of the add A-riboswitch while the removal of the ligand can result in drastic increase of the thermodynamic fluctuations especially in the junction regions between helix domains. Under the assumption that the native contacts with the highest thermodynamic fluctuations break first, the iterative GNM simulations showed that the unfolding process of the adenine-free add A-riboswitch starts with the denature of the terminal helix stem, followed by the loops and junctions involving ligand binding pocket, and then the central helix domains. Despite the simplified coarse-grained modeling, the unfolding dynamics and pathways are shown in close agreement with the results from atomic-level MD simulations and the NMR and single-molecule force spectroscopy experiments. Overall, the study demonstrates a new avenue to investigate the binding and folding dynamics of add A-riboswitch molecule which can be readily extended for other RNA molecules.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

ET-AL: Entropy-targeted active learning for bias mitigation in materials data

Growing materials data and data-driven informatics drastically promote the discovery and design of materials. While there are significant advancements in data-driven models, the quality of data resources is less studied despite its huge impact on model performance. In this work, we focus on data bias arising from uneven coverage of materials families in existing knowledge. Observing different diversities among crystal systems in common materials databases, we propose an information entropy-based metric for measuring this bias. To mitigate the bias, we develop an entropy-targeted active learning (ET-AL) framework, which guides the acquisition of new data to improve the diversity of underrepresented crystal systems. We demonstrate the capability of ET-AL for bias mitigation and the resulting improvement in downstream machine learning models. This approach is broadly applicable to data-driven materials discovery, including autonomous data acquisition and dataset trimming to reduce bias, as well as data-driven informatics in other scientific domains.

36 MATERIALS SCIENCE↗

Range data description based on multiple characteristics

An algorithm for describing range images based on Mean curvature (H) and Gaussian curvature (K) is presented. Range images are unique in that they directly approximate the physical surfaces of a real world 3-D scene. The curvature parameters are derived from the fundamental theorems of differential geometry and provides visible invariant pixel labels that can be used to characterize the scene. The sign of H and K can be used to classify each pixel into one of eight possible surface types. Due to the sensitivity of these parameters to noise the resulting HK-sing map does not directly identify surfaces in the range images and must be further processed. A region growing algorithm based on modeling the scene points with a Markov Random Field (MRF) of variable neighborhood size and edge models is suggested. This approach allows the integration of information from multiple characteristics in an efficient way. The performance of the proposed algorithm on a number of synthetic and real range images is discussed.

Al-Hujazi, Ezzet↗

Ensemble models for circuit topology estimation, fault detection and classification in distribution systems

This paper presents a methodology for simultaneous fault detection, classification, and topology estimation for adaptive protection of distribution systems. The methodology estimates the probability of the occurrence of each one of these events by using a hybrid structure that combines three sub-systems, a convolutional neural network for topology estimation, a fault detection based on predictive residual analysis, and a standard support vector machine with probabilistic output for fault classification. The input to all these sub-systems is the local voltage and current measurements. A convolutional neural network uses these local measurements in the form of sequential data to extract features and estimate the topology conditions. The fault detector is constructed with a Bayesian stage (a multitask Gaussian process) that computes a predictive distribution (assumed to be Gaussian) of the residuals using the input. Since the distribution is known, these residuals can be transformed into a Standard distribution, whose values are then introduced into a one-class support vector machine. The structure allows using a one-class support vector machine without parameter cross-validation, so the fault detector is fully unsupervised. Finally, a support vector machine uses the input to perform the classification of the fault types. All three sub-systems can work in a parallel setup for both performance and computation efficiency. In conclusion, we test all three sub-systems included in the structure on a modified IEEE123 bus system, and we compare and evaluate the results with standard approaches.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Spatiotemporally dynamic implicit large eddy simulation using machine learning classifiers

In this article, we utilize machine learning to dynamically determine if a point on the computational grid requires implicit numerical dissipation for large eddy simulation (LES). The decision making process is learnt through a priori training on quantities derived from direct numerical simulation (DNS) data. In particular, we compute eddy-viscosities obtained through the coarse-graining of DNS quantities and utilize their projection onto a Gaussian distribution to categorize areas that may require dissipation. If our learning determines that closure is necessary, an upwinded scheme is utilized for computing the non-linear Jacobian. In contrast, if it is determined that closure is unnecessary, a symmetric and second-order accurate energy and enstrophy preserving Arakawa scheme is utilized instead. This results in a closure framework that precludes the specification of any model-form for the small scale contributions of turbulence but deploys an appropriate numerical dissipation from explicit closure driven hypotheses. Here, this methodology is deployed for the Kraichnan turbulence test-case and assessed through various statistical quantities such as angle-averaged kinetic energy spectra and vorticity structure functions. Our framework thus establishes a link between the use of explicit LES ideologies for closure and numerical dissipation-based modeling of turbulence leading to improved statistical fidelity of a posteriori simulations.

42 ENGINEERING↗

Machine learning inversion from scattering for mechanically driven polymers

A machine learning inversion method is developed for analyzing scattering functions of mechanically driven polymers and extracting the corresponding feature parameters, which include energy parameters and conformation variables. The polymer is modeled as a chain of fixed-length bonds constrained by bending energy, and it is subject to external forces such as stretching and shear. We generate a data set consisting of random combinations of energy parameters, including bending modulus, stretching and shear force, along with Monte Carlo-calculated scattering functions and conformation variables such as end-to-end distance, radius of gyration and off-diagonal component of the gyration tensor. The effects of the energy parameters on the polymer are captured by the scattering function, and principal component analysis ensures the feasibility of the machine learning inversion. Finally, we train a Gaussian process regressor using part of the data set as a training set and validate the trained regressor for inversion using the rest of the data. The regressor successfully extracts the feature parameters.

Gaussian process regressors↗

Digital TV processing system

Two digital video data compression systems directly applicable to the Space Shuttle TV Communication System were described: (1) For the uplink, a low rate monochrome data compressor is used. The compression is achieved by using a motion detection technique in the Hadamard domain. To transform the variable source rate into a fixed rate, an adaptive rate buffer is provided. (2) For the downlink, a color data compressor is considered. The compression is achieved first by intra-color transformation of the original signal vector, into a vector which has lower information entropy. Then two-dimensional data compression techniques are applied to the Hadamard transformed components of this last vector. Mathematical models and data reliability analyses were also provided for the above video data compression techniques transmitted over a channel encoded Gaussian channel. It was shown that substantial gains can be achieved by the combination of video source and channel coding.

Source record↗

Entanglement evolution of a two-mode Gaussian system in various thermal environments

We describe the evolution of the quantum entanglement of an open system consisting of two bosonic modes interacting with a common thermal environment, described by two different models. The initial state of the system is taken of Gaussian form. In the case of a thermal bath, characterized by temperature and dissipation constant which correspond to an asymptotic Gibbs state of the system, we show that for a zero temperature of the thermal bath an initial entangled Gaussian state remains entangled for all finite times. For an entangled initial squeezed thermal state, the phenomenon of entanglement sudden death takes place and we calculate the survival time of entanglement. For the second model of the environment, corresponding to a non-Gibbs asymptotic state, we study the possibility of generating entanglement. We show that the generation of the entanglement between two uncoupled bosonic modes is possible only for definite values of the temperature and dissipation constant, which characterize the thermal environment.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Computational Mechanics for Heterogeneous Materials

The subject of this work is the development of models for the numerical simulation of matter, momentum, and energy balance in heterogeneous materials. These are materials that consist of multiple phases or species or that are structured on some (perhaps many) scale(s). By computational mechanics we mean to refer generally to the standard type of modeling that is done at the level of macroscopic balance laws (mass, momentum, energy). We will refer to the flow or flux of these quantities in a generalized sense as transport. At issue here are the forms of the governing equations in these complex materials which are potentially strongly inhomogeneous below some correlation length scale and are yet homogeneous on larger length scales. The question then becomes one of how to model this behavior and what are the proper multi-scale equations to capture the transport mechanisms across scales. To address this we look to the area of generalized stochastic process that underlie the transport processes in homogeneous materials. The archetypal example being the relationship between a random walk or Brownian motion stochastic processes and the associated Fokker-Planck or diffusion equation. Here we are interested in how this classical setting changes when inhomogeneities or correlations in structure are introduced into the problem. Aspects of non-classical behavior need to be addressed, such as non-Fickian behavior of the mean-squared-displacement (MSD) and non-Gaussian behavior of the underlying probability distribution of jumps. We present an experimental technique and apparatus built to investigate some of these issues. We also discuss diffusive processes in inhomogeneous systems, and the role of the chemical potential in diffusion of hard spheres is considered. Also, the relevance to liquid metal solutions is considered. Finally we present an example of how inhomogeneities in material microstructure introduce fluctuations at the meso-scale for a thermal conduction problem. These fluctuations due to random microstructures also provide a means of characterizing the aleatory uncertainty in material properties at the mesoscale.

36 MATERIALS SCIENCE↗

MODEL OF THE FIELD LINE RANDOM WALK EVOLUTION AND APPROACH TO ASYMPTOTIC DIFFUSION IN MAGNETIC TURBULENCE

The turbulent random walk of magnetic field lines plays an important role in the transport of plasmas and energetic particles in a wide variety of astrophysical situations, but most theoretical work has concentrated on determination of the asymptotic field line diffusion coefficient. Here we consider the evolution with distance of the field line random walk using a general ordinary differential equation (ODE), which for most cases of interest in astrophysics describes a transition from free streaming to asymptotic diffusion. By challenging theories of asymptotic diffusion to also describe the evolution, one gains insight on how accurately they describe the random walk process. Previous theoretical work has effectively involved closure of the ODE, often by assuming Corrsin's hypothesis and a Gaussian displacement distribution. Approaches that use quasilinear theory and prescribe the mean squared displacement ({Delta}x {sup 2}) according to free streaming (random ballistic decorrelation, RBD) or asymptotic diffusion (diffusive decorrelation, DD) can match computer simulation results, but only over specific parameter ranges, with no obvious 'marker' of the range of validity. Here we make use of a unified description in which the ODE determines ({Delta}x {sup 2}) self-consistently, providing a natural transition between the assumptions of RBD and DD. We find that the minimum kurtosis of the displacement distribution provides a good indicator of whether the self-consistent ODE is applicable, i.e., inaccuracy of the self-consistent ODE is associated with non-Gaussian displacement distributions.

79 ASTRONOMY AND ASTROPHYSICS↗

The Navstar GPS master control station's Kalman filter experience

The Navstar Global Positioning System (GPS) is a highly accurate space based navigation system providing all weather, 24 hour a day service to both military and civilian users. The system provides a Gaussian position solution with four satellites, each providing its ephemeris and clock offset with respect to GPS time. The GPS Master Clock Station (MCS) is charged with tracking each Navstar spacecraft and precisely defining the ephemeris and clock parameters for upload into the vehicle's navigation message. Briefly described here are the Navstar system and the Kalman filter estimation process used by MCS to determine, predict, and ensure quality control for each of the satellite's ephemeris and clock states. Routine performance is shown. Kalman filter reaction and response is discussed for anomalous clock behavior and trajectory perturbations. Particular attention is given to MCS efforts to improve orbital adjust modeling. The satellite out of service time due to orbital maneuvering has been reduced in the past year from four days to under twelve hours. The planning, reference trajectory model, and Kalman filter management improvements are explained.

Scardera, Michael P.↗

Analysis of moire fringes by Wiener filtering: An extension to the Fourier method

In X-ray Talbot interferometry, tilting the phase grating with respect to the absorption grating results in the formation of spatial fringes. The analysis of this moire pattern, classically performed by the Fourier method, allows the extraction of the sample phase shift information from a single image. In this context, an extension to the Fourier method is proposed. The filter used to extract the fringe information is chosen optimally in the least-squares sense, given models for the zeroth and first order modes, noise and the modulation transfer function. The latter is obtained by measuring the detector response to moire fringes with increasing frequencies. The obtained Wiener filter allows a better reconstruction of the phase information at all fringe frequencies, compared to the usual box or gaussian filters. This is demonstrated quantitatively by experiments using synchrotron radiation.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Electronically nonadiabatic wave packet propagation using frozen Gaussian scattering

We present an approach, which allows to employ the adiabatic wave packet propagation technique and semiclassical theory to treat the nonadiabatic processes by using trajectory hopping. The approach developed generates a bunch of hopping trajectories and gives all additional information to incorporate the effect of nonadiabatic coupling into the wave packet dynamics. This provides an interface between a general adiabatic frozen Gaussian wave packet propagation method and the trajectory surface hopping technique. The basic idea suggested in [A. D. Kondorskiy and H. Nakamura, J. Chem. Phys. 120, 8937 (2004)] is revisited and complemented in the present work by the elaboration of efficient numerical algorithms. We combine our approach with the adiabatic Herman-Kluk frozen Gaussian approximation. The efficiency and accuracy of the resulting method is demonstrated by applying it to popular benchmark model systems including three Tully’s models and 24D model of pyrazine. It is shown that photoabsorption spectrum is successfully reproduced by using a few hundreds of trajectories. We employ the compact finite difference Hessian update scheme to consider feasibility of the ab initio “on-the-fly” simulations. It is found that this technique allows us to obtain the reliable final results using several Hessian matrix calculations per trajectory.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Investigation of Marangoni convection with high-fidelity simulations for metal melt pool dynamics

The goal of this research is to investigate and characterize the dynamics of a molten metal pool through modeling and simulation. Two manufacturing processes are of interest, laser spot welding and selective laser melting. These are laser-based techniques for joining parts together and for building up parts. The laser power impacts the melt pool temperature and therefore the dynamics. Moreover, the metals used are often alloys, containing impurities, where the concentration of these impurities can have a large effect on surface tension and therefore influence the melt pool dynamics. The application space considered in this project is for laser spot welding as a case study, and is also relevant to 3D metal printing. The thermal evolution and fluid dynamics in a 2D metal block exposed to a Gaussian distributed laser source is numerically simulated using a high-order, fully-implicit code. Results are compared to published experimental data and other published numerical simulations for laser spot welding. Results confirm that unstable structures appear at higher laser powers with higher impurity concentrations compared to simulations with no impurities and lower laser powers.

42 ENGINEERING↗

Tuning and Robustness Analysis for the Orion Absolute Navigation System

The Orion Multi-Purpose Crew Vehicle (MPCV) is currently under development as NASA's next-generation spacecraft for exploration missions beyond Low Earth Orbit. The MPCV is set to perform an orbital test flight, termed Exploration Flight Test 1 (EFT-1), some time in late 2014. The navigation system for the Orion spacecraft is being designed in a Multi-Organizational Design Environment (MODE) team including contractor and NASA personnel. The system uses an Extended Kalman Filter to process measurements and determine the state. The design of the navigation system has undergone several iterations and modifications since its inception, and continues as a work-in-progress. This paper seeks to show the efforts made to-date in tuning the filter for the EFT-1 mission and instilling appropriate robustness into the system to meet the requirements of manned space ight. Filter performance is affected by many factors: data rates, sensor measurement errors, tuning, and others. This paper focuses mainly on the error characterization and tuning portion. Traditional efforts at tuning a navigation filter have centered around the observation/measurement noise and Gaussian process noise of the Extended Kalman Filter. While the Orion MODE team must certainly address those factors, the team is also looking at residual edit thresholds and measurement underweighting as tuning tools. Tuning analysis is presented with open loop Monte-Carlo simulation results showing statistical errors bounded by the 3-sigma filter uncertainty covariance. The Orion filter design uses 24 Exponentially Correlated Random Variable (ECRV) parameters to estimate the accel/gyro misalignment and nonorthogonality. By design, the time constant and noise terms of these ECRV parameters were set to manufacturer specifications and not used as tuning parameters. They are included in the filter as a more analytically correct method of modeling uncertainties than ad-hoc tuning of the process noise. Tuning is explored for the powered-flight ascent phase, where measurements are scarce and unmodelled vehicle accelerations dominate. On orbit, there are important trade-off cases between process and measurement noise. On entry, there are considerations about trading performance accuracy for robustness. Process Noise is divided into powered flight and coasting ight and can be adjusted for each phase and mode of the Orion EFT-1 mission. Measurement noise is used for the integrated velocity measurements during pad alignment. It is also used for Global Positioning System (GPS) pseudorange and delta- range measurements during the rest of the flight. The robustness effort has been focused on maintaining filter convergence and performance in the presence of unmodeled error sources. These include unmodeled forces on the vehicle and uncorrected errors on the sensor measurements. Orion uses a single-frequency, non-keyed GPS receiver, so the effects due to signal distortion in Earth's ionosphere and troposphere are present in the raw measurements. Results are presented showing the efforts to compensate for these errors as well as characterize the residual effect for measurement noise tuning. Another robustness tool in use is tuning the residual edit thresholds. The trade-off between noise tuning and edit thresholds is explored in the context of robustness to errors in dynamics models and sensor measurements. Measurement underweighting is also presented as a method of additional robustness when processing highly accurate measurements in the presence of large filter uncertainties.

Holt, Greg N.↗

Development and Modeling of a Novel Self-Assembly Process for Polymer and Polymeric Composite Nanoparticles

Extensive computational simulations and experiments have been used to investigate the structure, dynamics and resulting photophysical properties of a number para-phenylenevinylene (PPV) based polymers and oligomers. These studies have shown how the morphology and structure are controlled to a large extent by the nature of the solute-solvent interactions in the initial solution phase preparation. A good solvent such as dichloromethane generates non-compact structures with more of a defect-extended chain like morphology while a bad solvent such as toluene leads to compact organized and folded structures with rod-like morphologies. Secondary structural organization is induced by using the solution phase structures to generate solvent-free single molecule nanoparticles. These nanoparticles are very compact and rod shaped, consisting of near-cofacial ordering of the conjugated PPV chain backbones between folds located at tetrahedral defects (sp3 C-C bonds). The resulting photophysical properties exhibit a significant enhancement in the photoluminescence quantum yield, lifetime, and stability. In addition, the single molecule nanoparticles have Gaussian-like emission spectra with discrete center frequencies that are correlated to a conjugation length, allowing the design of nanoparticles which luminesces at a particular frequency. We followed a similar approach and applied a comparable methodology in our recent work on polythiophenes in order to study the effect of polymer architecture on nanoscale assembly. Unlike linear chains of comparable size, we observed aggregation of the bottlebrush architecture of poly(norbornene)-g-poly(3-hexylthiophene) (PNB-g-P3HT) after the freeze-drying and dissolution processes. The behavior can be attributed to a significant enhancement in the number of π-π interactions between grafted P3HT side chains.

Sumpter, Bobby G.↗

Discriminative Dimensionality Reduction using Deep Neural Networks for Clustering of LIGO Data

In this paper, leveraging the capabilities of neural networks for modeling the non-linearities that exist in the data, we propose several models that can project data into a low dimensional, discriminative, and smooth manifold. The proposed models can transfer knowledge from the domain of known classes to a new domain where the classes are unknown. A clustering algorithm is further applied in the new domain to find potentially new classes from the pool of unlabeled data. The research problem and data for this paper originated from the Gravity Spy project which is a side project of Advanced Laser Interferometer Gravitational-wave Observatory (LIGO). The LIGO project aims at detecting cosmic gravitational waves using huge detectors. However non-cosmic, non-Gaussian disturbances known as "glitches", show up in gravitational-wave data of LIGO. This is undesirable as it creates problems for the gravitational wave detection process. Gravity Spy aids in glitch identification with the purpose of understanding their origin. Since new types of glitches appear over time, one of the objective of Gravity Spy is to create new glitch classes. Towards this task, we offer a methodology in this paper to accomplish this.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗