Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Function Estimation”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

Contaminant point source localization error estimates as functions of data quantity and model quality

We develop empirically-grounded error envelopes for localization of a point contamination release event in the saturated zone of a previously uncharacterized heterogeneous aquifer into which a number of plume-intercepting wells have been drilled. We assume that flow direction in the aquifer is known exactly and velocity is known to within a factor of two of our best guess from well observations prior to source identification. Other aquifer and source parameters must be estimated by interpretation of well breakthrough data via the advection-dispersion equation. We employ high performance computing to generate numerous random realizations of aquifer parameters and well locations, simulate well breakthrough data, and then employ unsupervised machine optimization techniques to estimate the most likely spatial (or space-time) location of the source. Tabulating the accuracy of these estimates from the multiple realizations, we relate the size of 90% and 95% confidence envelopes to the data quantity (number of wells) and model quality (fidelity of ADE interpretation model to actual concentrations in a heterogeneous aquifer with channelized flow). We find that for purely spatial localization of the contaminant source, increased data quantities can make up for reduced model quality. For space-time localization, we find similar qualitative behavior, but significantly degraded spatial localization reliability and less improvement from extra data collection. Since the space-time source localization problem is much more challenging, we also tried a multiple-initial-guess optimization strategy. Furthermore, this greatly enhanced performance, but gains from additional data collection remained limited.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2019

Temporal changes of seismic velocities in the San Jacinto Fault zone associated with the 2016 M w 5.2 Borrego Springs earthquake

We study temporal changes of seismic velocities associated with the 10 June 2016 Mw 5.2 Borrego Springs earthquake in the San Jacinto fault zone, using nine component Green's function estimates reconstructed from daily cross correlations of ambient noise. The analysed data are recorded by stations in two dense linear arrays, at Dry Wash (DW) and Jackass Flat (JF), crossing the fault surface trace ~3 km northwest and southeast of the event epicentre. The two arrays have 9 and 12 stations each with instrument spacing of 25–100 m. Relative velocity changes (δv/v) are estimated from arrival time changes in the daily correlation coda waveforms compared to a reference stack. The obtained array-average δv/v time-series exhibit changes associated with the Borrego Springs event, superposed with seasonal variations. The earthquake-related changes are characterized by a rapid coseismic velocity drop followed by a gradual recovery. This is consistently observed at both arrays using time- and frequency-domain δv/v analyses with data from different components in various frequency bands. Larger changes at lower frequencies imply that the variations are not limited to the near surface material. A decreasing coseismic velocity reduction with coda wave lapse time indicates larger coseismic structural perturbations in the fault zone and near-fault environment compared to the surrounding rock. Observed larger changes at the DW array compared to the JF array possibly reflect the northwestward rupture directivity of the Borrego Springs earthquake.

Geochemistry & Geophysics↗
DOE OSTI2018

Estimates for Functionals of Solutions to Higher-Order Heat-Type Equations with Random Initial Conditions

In the present paper we continue the investigation of solutions to higher-order heat-type equations with random initial conditions, which play the important role in many applied areas. We consider the random initial conditions given by harmonizable φ-sub-Gaussian processes. The main results are the bounds for the distributions of the suprema over bounded and unbounded domains for solutions of such equations. The results obtained in the paper hold, in particular, for the case of Gaussian initial condition.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2014

Signal/noise enhancement strategies for stochastically estimated correlation functions

We develop here strategies for enhancing the signal/noise ratio for stochastically sampled correlation functions. The techniques are general and offer a wide range of applicability. We demonstrate the potential of the approach with a generic two-state system and then explore the practical applicability of the method for single hadron correlators in lattice quantum chromodynamics. In the latter case, we determine the ground state energies of the pion, proton, and delta baryon, as well as the ground and first excited state energy of the rho meson using matrices of correlators computed on an exemplary ensemble of anisotropic gauge configurations. In the majority of cases, we find a modest reduction in the statistical uncertainties on extracted energies compared to conventional variational techniques. However, in the case of the delta baryon, we achieve a factor of 3 reduction in statistical uncertainties. The variety of outcomes achieved for single hadron correlators illustrates an inherent dependence of the method on the properties of the system under consideration and the operator basis from which the correlators are constructed.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2021

Signal Preconditioning to Minimize Impulse Response Contribution

A study was performed to identify a method to minimize the effect of a linear time-invariant (LTI) system impulse response on an input. Three methods were studied: Wiener filter, the N4SID algorithm and transfer function estimation, the latter two using functions from MATLAB’s System Identification Toolbox. Although all three methods were able estimate an unknown forward impulse response given an input/output time series pair, only the Wiener filter was able to estimate a system inverse which satisfactorily solved the problem using a cosine similarity measure.

42 ENGINEERING↗
DOE OSTI2013

Effect of low and high storage temperatures on head space gas concentrations and physical properties of wood pellets

Headspace gas concentrations and wood pellet properties were studied in sealed glass canisters at 5–40 degrees C storage temperatures. CO2 and CO concentrations at 5, 10, 20 and 40 degrees C at the end of 23–28 days of storage were 1600 and 200, 4700 and 1200, and 31 200 and 15 800 parts per million by volume (ppmv) respectively. Corresponding O2 concentration was about 19•49, 19•20, 18•0 and 2•07% respectively. Non-linear regression equations adequately described the gas concentrations in the storage container as a function of time. Safe level estimation functions developed were linear for O2 and logarithmic for CO and CO2 concentrations. Measured pellet properties moisture, length, diameter, unit, bulk and tapped density, durability, calorific value, ash content and per cent fines were in the range of 4•6–5•02%, 14–15 mm, 6•4–6•5 mm, 1125–1175 kg m-3, 750–770 kg m-3, 825–840 kg m-3, 73–74%, 18•32–18•78 MJ kg-1, 0•65–0•74% and 0•13–0•15%. Durability values of pellets decreased by 13% at 40 degrees C storage temperature and other properties changed marginally.

09 BIOMASS FUELS↗
DOE OSTI2014

X-ray-selected galaxy groups in Boötes

We present the X-ray and optical properties of the galaxy groups selected in the Chandra X-Boötes survey. We used follow-up Chandra observations to better define the group sample and their X-ray properties. Group redshifts were measured from the AGN and Galaxy Evolution Survey spectroscopic data. We used photometric data from the NOAO Deep Wide Field Survey to estimate the group richness (N {sub gals}) and the optical luminosity (L {sub opt}). Our final sample comprises 32 systems at z < 1.75 with 14 below z = 0.35. For these 14 systems, we estimate velocity dispersions (σ {sub gr}) and perform a virial analysis to obtain the radii (R {sub 200} and R {sub 500}) and total masses (M {sub 200} and M {sub 500}) for groups with at least 5 galaxy members. We use the Chandra X-ray observations to derive the X-ray luminosity (L{sub X} ). We examine the performance of the group properties σ{sub gr}, L {sub opt}, and L{sub X} , as proxies for the group mass. Understanding how well these observables measure the total mass is important to estimate how precisely the cluster/group mass function is determined. Exploring the scaling relations built with the X-Boötes sample and comparing these with samples from the literature, we find a break in the L{sub X} -M {sub 500} relation at approximately M {sub 500} = 5 × 10{sup 13} M {sub ☉} (for M {sub 500} > 5 × 10{sup 13} M {sub ☉}, M{sub 500}∝L{sub X}{sup 0.61±0.02}, while for M {sub 500} ≤ 5 × 10{sup 13} M {sub ☉}, M{sub 500}∝L{sub X}{sup 0.44±0.05}). Thus, the mass-luminosity relation for galaxy groups cannot be described by the same power law as galaxy clusters. A possible explanation for this break is the dynamical friction, tidal interactions, and projection effects that reduce the velocity dispersion values of the galaxy groups. By extending the cluster luminosity function to the group regime, we predict the number of groups that new X-ray surveys, particularly eROSITA, will detect. Based on our cluster/group luminosity function estimates, eROSITA will identify ∼1800 groups (L{sub X} = 10{sup 41}-10{sup 43} erg s{sup –1}) within a distance of 200 Mpc. Since groups lie in large-scale filaments, this group sample will map the large-scale structure of the local universe.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2017

Galaxy–galaxy lensing estimators and their covariance properties

Here, we study the covariance properties of real space correlation function estimators – primarily galaxy–shear correlations, or galaxy–galaxy lensing – using SDSS data for both shear catalogues and lenses (specifically the BOSS LOWZ sample). Using mock catalogues of lenses and sources, we disentangle the various contributions to the covariance matrix and compare them with a simple analytical model. We show that not subtracting the lensing measurement around random points from the measurement around the lens sample is equivalent to performing the measurement using the lens density field instead of the lens overdensity field. While the measurement using the lens density field is unbiased (in the absence of systematics), its error is significantly larger due to an additional term in the covariance. Therefore, this subtraction should be performed regardless of its beneficial effects on systematics. Comparing the error estimates from data and mocks for estimators that involve the overdensity, we find that the errors are dominated by the shape noise and lens clustering, which empirically estimated covariances (jackknife and standard deviation across mocks) that are consistent with theoretical estimates, and that both the connected parts of the four-point function and the supersample covariance can be neglected for the current levels of noise. While the trade-off between different terms in the covariance depends on the survey configuration (area, source number density), the diagnostics that we use in this work should be useful for future works to test their empirically determined covariances.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2014

Physics-based, Bayesian sequential detection method and system for radioactive contraband

A distributed sequential method and system for detecting and identifying radioactive contraband from highly uncertain (noisy) low-count, radionuclide measurements, i.e. an event mode sequence (EMS), using a statistical approach based on Bayesian inference and physics-model-based signal processing based on the representation of a radionuclide as a monoenergetic decomposition of monoenergetic sources. For a given photon event of the EMS, the appropriate monoenergy processing channel is determined using a confidence interval condition-based discriminator for the energy amplitude and interarrival time and parameter estimates are used to update a measured probability density function estimate for a target radionuclide. A sequential likelihood ratio test is then used to determine one of two threshold conditions signifying that the EMS is either identified as the target radionuclide or not, and if not, then repeating the process for the next sequential photon event of the EMS until one of the two threshold conditions is satisfied.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2022

Multi-variance replica exchange SGMCMC for inverse and forward problems via Bayesian PINN

Physics-informed neural network (PINN) has been successfully applied in solving a variety of nonlinear non-convex forward and inverse problems. However, the training is challenging because of the non-convex loss functions and the multiple optima in the Bayesian inverse problem. In this work, we propose a multi-variance replica exchange stochastic gradient Langevin dynamics method to tackle the challenge of the multiple local optima in the optimization and the challenge of the multiple modal posterior distribution in the inverse problem. Replica exchange methods are capable of escaping from the local traps and accelerating the convergence; two chains with different temperatures are designed where the low temperature chain aims for the local convergence, and the target of the high temperature chain is to travel globally and explore the whole loss function entropy landscape. However, it may not be efficient to solve mathematical inversion problems by using the vanilla replica method directly since the method doubles the computational cost in evaluating the forward solvers (likelihood functions) in the two chains. To address this issue, we propose to make different assumptions on the energy function estimation and this facilities one to use solvers of different fidelities in the likelihood function evaluation. More precisely, one can use a solver with low fidelity in the high temperature chain while using a solver with high fidelity in the low temperature chain. Our proposed method significantly lowers the computational cost in the high temperature chain, meanwhile preserving the accuracy and converging very fast. Here we give an unbiased estimate of the swapping rate and give an estimation of the discretization error of the scheme. To verify our idea, we design and solve four inverse problems which have multiple modes. The proposed method is also employed to train the Bayesian PINN to solve the forward and inverse problems; faster and more accurate convergence has been observed when compared to the stochastic gradient Langevin dynamics (SGLD) method and vanilla replica exchange methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

revengc: An R package to Reverse Engineer Summarized Data

Decoupled (e.g. separate averages) and censored (e.g. > 100 species) variables are continually reported by many well-established organizations, such as the World Health Organization (WHO), Centers for Disease Control and Prevention (CDC), and World Bank. The challenge therefore is to infer what the original data could have been given summarized information. We present an R package that reverse engineers censored and/or decoupled data with two main functions. The cnbinom.pars() function estimates the average and dispersion parameter of a censored univariate frequency table. The rec() function reverse engineers summarized data into an uncensored bivariate table of probabilities.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Out of Distribution Detection with Neural Network Anchoring

This is code to reproduce and build on OOD detection from the paper "Out of Distribution Detection with Neural Network Anchoring". Our goal here is to exploit heteroscedastic temperature scaling as a calibration strategy for out of distribution (OOD) detection. Heteroscedasticity here refers to the fact that the optimal temperature parameter for each sample can be different, as opposed to conventional approaches that use the same value for the entire distribution. To enable this, we propose a new training strategy called anchoring that can estimate appropriate temperature values for each sample, leading to state-of-the-art OOD detection performance across several benchmarks. Using NTK theory, we show that this temperature function estimate is closely linked to the epistemic uncertainty of the classifier, which explains its behavior. In contrast to some of the best-performing OOD detection approaches, our method does not require exposure to additional outlier datasets, custom calibration objectives, or model ensembling. Through empirical studies with different OOD detection settings - far OOD, near OOD, and semantically coherent OOD - we establish a highly effective OOD detection approach.

Thiagarajan, Jayaraman↗
DOE OSTI2014

Planck 2013 results. XXIV. Constraints on primordial non-Gaussianity

The Planck nominal mission cosmic microwave background (CMB) maps yield unprecedented constraints on primordial non-Gaussianity (NG). Using three optimal bispectrum estimators, separable template-fitting (KSW), binned, and modal, we obtain consistent values for the primordial local, equilateral, and orthogonal bispectrum amplitudes, quoting as our final result f NL local = 2.7 ± 5.8, f NL equil = -42 ± 75, and f NL orth = -25 ± 39 (68% CL statistical). Non-Gaussianity is detected in the data; using skew-C &ell; statistics we find a nonzero bispectrum from residual point sources, and the integrated-Sachs-Wolfe-lensing bispectrum at a level expected in the ΛCDM scenario. The results are based on comprehensive cross-validation of these estimators on Gaussian and non-Gaussian simulations, are stable across component separation techniques, pass an extensive suite of tests, and are confirmed by skew-C &ell; , wavelet bispectrum and Minkowski functional estimators. Beyond estimates of individual shape amplitudes, we present model-independent, three-dimensional reconstructions of the Planck CMB bispectrum and thus derive constraints on early-Universe scenarios that generate primordial NG, including general single-field models of inflation, excited initial states (non-Bunch-Davies vacua), and directionally-dependent vector models. We provide an initial survey of scale-dependent feature and resonance models. These results bound both general single-field and multi-field model parameter ranges, such as the speed of sound, c s ≥ 0.02 (95% CL), in an effective field theory parametrization, and the curvaton decay fraction r D ≥ 0.15 (95% CL). The Planck data significantly limit the viable parameter space of the ekpyrotic/cyclic scenarios. The amplitude of the four-point function in the local model τ NL < 2800 (95% CL). Taken together, these constraints represent the highest precision tests to date of physical mechanisms for the origin of cosmic structure.

79 ASTRONOMY AND ASTROPHYSICS↗