SEARCH · Engineering Papers

Results for “Data fitting”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 811 records · Page 45

All sources NASA NTRS DOE OSTI

NASA NTRS2019

A Fundamental Plane for Gamma-Ray Pulsars

We show that the γ-ray pulsar observables, i.e., their total γ-ray luminosity, L(sub γ), spectral cutoff energy, ϵ (sub cut), stellar surface magnetic field, B(sub ⋆), and spin-down power ε, obey a relation of the form L(sub γ) = f (ϵ (sub cut), B(sub ⋆), ε ), which represents a 3D plane in their 4D logspace. Fitting the data of 88 pulsars of the second Fermi pulsar catalog, we show this relation to be L(sub γ) ∝ ϵ (sub cut)(sup 1.18±0.24 B(sub ⋆)(sup 0.17± 0.05) ε(sup 0.41±0.08), a pulsar fundamental plane (FP). We show that the observed FP is remarkably close to the theoretical relation L(sub γ) ∝ ϵ (sub cut)(sup 4/3) B(sub ⋆)(sup 1/6) ε(sup 5/12) obtained assuming that the pulsar γ-ray emission is due to curvature radiation by particles accelerated at the pulsar equatorial current sheet just outside the light cylinder. Interestingly, the FP seems incompatible with emission by synchrotron radiation. The corresponding scatter about the FP is ∼0.35 dex and can only partly be explained by the observational errors while the rest is probably due to the variation of the inclination and observer angles. We predict also that ϵ (sub cut) ∝ ε (sup 7/16) toward low ε for both young and millisecond pulsars implying that the observed death line of γ-ray pulsars is due to ϵ (sub cut) dropping below the Fermi band. Our results provide a comprehensive interpretation of the observations of γ-ray pulsars, setting requirements for successful theoretical modeling.

Kalapotharakos, Constantinos↗

NASA NTRS2019

Numerical Investigation and Optimization of a Flushwall Injector for Scramjet Applications at Hypervelocity Flow Conditions

An investigation utilizing Reynolds-averaged simulations (RAS) was performed in order to find optimal designs for an interdigitated flushwall injector suitable for scramjet applications at hypervelocity conditions. The flight Mach number, duct height, spanwise width, and injection angle were the design variables selected to maximize two objective functions: the thrust potential and combustion efficiency. A Latin hypercube sampling design-of-experiments method was used to select design points for RAS. A methodology was developed that automated building geometries and generating grids for each design. The ensuing RAS analysis generated the performance database from which the two objective functions of interest were computed using a one-dimensional performance utility. The data were fitted using four surrogate models: an artificial neural network (ANN) model, a cubic polynomial, a quadratic polynomial, and a Kriging model. Variance-based decomposition showed that both objective functions were primarily driven by changes in the duct height. Multiobjective design optimization was performed for all four surrogate models via a genetic algorithm method. Optimal solutions were obtained at the upper and lower bounds of the flight Mach number range. The Kriging model obtained an optimal solution set that predicted high values for both objective functions. Additionally, three challenge points were selected to assess the designs on the Pareto fronts. Further sampling among the designs of the Pareto fronts are required in order to lower the errors and perform more accurate surrogate-based optimization. sed optimization.

Shenoy, Rajiv R.↗

NASA NTRS2019

Numerical Investigation and Optimization of a Flushwall Injector for Scramjet Applications at Hypervelocity Flow Conditions

An investigation utilizing Reynolds-averaged simulations (RAS) was performed in order to demonstrate the use of design and analysis of computer experiments (DACE) methods in Sandia’s DAKOTA software package for surrogate modeling and optimization. These methods were applied to a flow- path fueled with an interdigitated flushwall injector suitable for scramjet applications at hyper- velocity conditions and ascending along a constant dynamic pressure flight trajectory. The flight Mach number, duct height, spanwise width, and injection angle were the design variables selected to maximize two objective functions: the thrust potential and combustion efficiency. Because the RAS of this case are computationally expensive, surrogate models are used for optimization. To build a surrogate model a RAS database is created. The sequence of the design variables comprising the database were generated using a Latin hypercube sampling (LHS) method. A methodology was also developed to automatically build geometries and generate structured grids for each design point. The ensuing RAS analysis generated the simulation database from which the two objective functions were computed using a one-dimensionalization (1D) of the three-dimensional simulation data. The data were fitted using four surrogate models: an artificial neural network (ANN), a cubic polynomial, a quadratic polynomial, and a Kriging model. Variance-based decomposition showed that both objective functions were primarily driven by changes in the duct height. Multiobjective design optimization was performed for all four surrogate models via a genetic algorithm method. Optimal solutions were obtained at the upper and lower bounds of the flight Mach number range. The Kriging model predicted an optimal solution set that exhibited high values for both objective functions. Additionally, three challenge points were selected to assess the designs on the Pareto fronts. Further sampling among the designs of the Pareto fronts may be required to lower the surrogate model errors and perform more accurate surrogate-model-based optimization.

Shenoy, Rajiv R.↗

NASA NTRS2018

Automatic Design of Missions to Small Bodies

The new JPL Small-Body Automatic Mission-Design System comprises two main elements: a database of pre-computed mission options to all known asteroids and comets, and an interactive web interface that can be used to design transfers to each small body. The system is kept current with the JPL Small Body orbit catalog, is publicly available, and can be accessed from the JPL Solar System Dynamics Group website. The missions computed by the automatic system are impulsive. However, a low-thrust v estimate is also provided. The database of pre-computed missions can be filtered to find potential targets with certain orbital and physical properties, and that meet specific mission-design constraints. In addition to describing the system in detail, this paper presents empirical analytic expressions to approximate the impulsive v requirements of missions to comets and to each family of asteroids, obtained by fitting statistical data. To show how the interactive interface works, we consider mission options to asteroid 99942 Apophis. We also include optimal mission opportunities to other selected small bodies.

Farnocchia, Davide↗

NASA NTRS2017

Juno Orbit Determination Experience During First Year At Jupiter

The Juno spacecraft successfully inserted into a polar orbit around Jupiter on 5- July-2016. Since the Jupiter Orbit Insertion (JOI) maneuver, Juno has completed six orbits around Jupiter. The mission plan at the time of JOI was for Juno to perform two 53.5-day capture orbits before executing a Period Reduction Maneuver (PRM) to place the spacecraft into its intended 14-day science orbit. This maneuver was canceled due to a concern with the propulsion system. As a result, the Juno spacecraft will remain in its longer orbit period for rest of its mission. This paper discusses the Navigation Team’s experience: the orbit determination strategy and how it changed due to the cancellation of the PRM, challenges fitting the data during perijove, and how we reconstructed the trajectory during Juno’s first year in orbit.

Thompson, Paul↗

NASA NTRS2022

Recommended Direct Simulation Monte Carlo Collision Model Parameters for Reacting Methane Flows

While hydrocarbon combustion and gas-surface interactions have been simulated for a variety of applications in continuum solvers, such processes have been rarely investigated in Direct Simulation Monte Carlo (DSMC). Thus Variable Hard/Soft Sphere (VHS/VSS)collision parameters are not available for most molecules of interest in the decomposition of methane, a significant pyrolysis gas product and potential atmospheric species for Titan re-entries. As these properties are essential for accurate modeling of gas transport and thermochemical effects, a method must be devised to compute them for use in DSMC. In this study, the methodology outlined in Stephani et al is utilized to compute the VHS/VSS parameters by fitting the Ω(1,1)and Ω(2,2) collision integrals [1]. These collision integrals are typically computed from intermolecular potentials and are used in CFD simulations to calculate transport properties such as viscosity, thermal conductivity, and diffusion. This approach ensures consistency between micro-scale collisions and macro-scale transport properties. To produce the necessary collision integrals needed for the fitting routine, we utilize the method described by Laricchiuta et al, where collision integrals are computed based on the constituent particles’ polarizability and effective electron number [2].Preliminary computations of pure-species Ω(1,1)and Ω(2,2)collision integrals produced from the coupled approach of Stephani et al and Laricchiuta et al for methane show good agreement with those computed via available ab-initio Potential Energy Surface Computations, as seen in Figure 1. Above 300K, no more than 12% deviation is observed between the two results over the temperature range of the fitted PES data [3].Current results of this study include full collision-specificVSS transport properties for a complex-species mixture involving major byproducts of pure methane decomposition. Targeted final results will include a comprehensive list of VSS parameters relevant toDSMC simulations of Titan atmospheric entry, based on the reduced kinetic model of Savajano et al [4].

M R Gosma↗

NASA NTRS2022

Model for Dielectric Constant of Seawater based on L-band Measurements with Conductivity by Definition

This article reports an improvement in the model for the dielectric constant of seawater used to fit laboratory measurements of the dielectric constant at the L-band. The new model (dielectric constant as a function of salinity, temperature, and frequency) is based on the response of a polar molecule proposed by Debye and fits the same measurement as reported in earlier work but uses a functional form for conductivity, σ(S,T) , that is given by the definition of salinity. The new version of this model fits the data well and has the advantages that the relaxation time constant is allowed to be a function of temperature and salinity and is well behaved when extrapolated to high salinities.

L-band↗

NASA NTRS2022

Quantifying urban, industrial, and background changes in NO2 during the COVID-19 lockdown period based on TROPOMI satellite observations

The COVID-19 lockdown had a large impact on anthropogenic emissions of air pollutants and particularly on nitrogen dioxide (NO 2 ). While the overall NO 2 decline over some large cities is well-established, understanding the details remains a challenge since multiple source categories contribute. In this study, a new method of isolation of three components (background NO 2 , NO 2 from urban sources, and NO 2 from industrial point sources) is applied to estimate the impact of the COVID-19 lockdown on each of them. The approach is based on fitting satellite data by a statistical model with empirical plume dispersion functions driven by a meteorological reanalysis. Population density and surface elevation data as well as coordinates of industrial sources were used in the analysis. The tropospheric NO 2 vertical column density (VCD) values measured by the Tropospheric Monitoring Instrument (TROPOMI) on board the Sentinel-5 Precursor over 261 urban areas for the period from 16 March to 15 June 2020 were compared with the average VCD values for the same period in 2018 and 2019. While the background NO 2 component remained almost unchanged, the urban NO 2 component declined by −18 % to −28 % over most regions. India, South America, and a part of Europe (particularly, Italy, France, and Spain) demonstrated a −40 % to −50 % urban emission decline. In contrast, the decline over urban areas in China, where the lockdown was over during the analysed period, was, on average, only -4.4± 8 %. Emissions from large industrial sources in the analysed urban areas varied greatly from region to region from -4.8 ± 6 % for China to -40 ± 10 % for India. Estimated changes in urban emissions are correlated with changes in Google mobility data (the correlation coefficient is 0.62) confirming that changes in traffic were one of the key elements in the decline in urban NO2 emissions. No correlation was found between changes in background NO2 and Google mobility data. On the global scale, the background and urban components were remarkably stable in 2018, 2019, and 2021, with averages of all analysed areas all being within ±2.5 % and suggesting that there were no substantial drifts or shifts in TROPOMI data. The 2020 data are clearly an outlier: in 2020, the mean background component for all analysed areas (without China) was -6.0 % ± 1.2 % and the mean urban component was -26.7 ± 2.6 % or 20σ below the baseline level from the other years.

COVID-19↗

NASA NTRS2023

Accounting for Dose Uncertainty in Dose-Response Curve Estimation Using Hierarchical Bayes Models

As part of their development of the technology for low-boom supersonic flight, the National Aeronautics and Space Administration (NASA) is planning to conduct a set of supersonic aircraft annoyance surveys in select communities in the United States, to measure public perception of the reduced sonic boom. The relationship between a noise exposure level of a supersonic flight event and the probability of an individual being highly annoyed by the event is quantified by a dose-response curve, which is usually based on logistic regression modeling. Unavoidable amounts of estimation error are expected in the upcoming noise measurements, which can bias the results of the logistic regression if ignored. Hence, the development of an estimation approach that accounts for such error when estimating the dose-response curve is imperative. In this paper, an evaluation study is conducted to assess the impact of measurement error on dose-response curve estimation. For this, hierarchical Bayes models using different specifications are fit to data collected by NASA in 2018, as part of a risk-reduction study of supersonic flights affecting Galveston, Texas. It is observed that the estimated dose-response curve appears sensitive to uncertainty in dose measurements, being subject to attenuation bias.

community response↗

NASA NTRSDate not supplied

Uncertainty Estimates for Fitting Zernike Polynomials to Discrete Data

Zernike polynomials are a widely used metric in modern optical analysis. They conveniently represent surfaces as a series of weighted terms corresponding to various optical aberrations. Ideally, each term is independent of others in the series, but Zernike polynomials lose this property when working with sets of discrete data. This gives rise to uncertainty in each polynomial’s actual contribution and affects metrology and simulation estimates of their relative weights. Several factors influencing these estimates are the number and arrangement of sample locations, the method for calculating the weights, and the total number of Zernike terms used in the calculation. Discussed is the uncertainty associated with linear regression using random sampling. Other topics reviewed are complex Zernike polynomials and vector spaces of functions.

Zernike Polynomials↗

DOE OSTI2022

Reverse-mode differentiation in arbitrary tensor network format: with application to supervised learning.

This paper describes an efficient reverse-mode differentiation algorithm for contraction operations of tensor networks that may have arbitrary and unconventional network topologies. The approach leverages the tensor contraction tree of Evenbly and Pfeifer (2014), which provides an instruction set for the contraction sequence of a network. We show that this tree can be efficiently leveraged for differentiation of a full tensor network contraction using a recursive scheme that exploits (1) the bilinear property of contraction and (2) the property that trees have single path from root to leaves. While differentiation of tensor-tensor contraction is already possible in most automatic differentiation packages, we show that exploiting these two additional properties in the specific context of contraction sequences can improve efficiency. Following a description of the algorithm and computational complexity analysis, we investigate its utility for gradient-based supervised learning for low-rank function recovery and for fitting real-world unstructured datasets. We demonstrate improved performance over alternating least-squares optimization approaches and the capability to handle heterogeneous and arbitrary tensor network formats. When compared to alternating minimization algorithms, we find that the gradient-based approach requires a smaller oversampling ratio (number of samples compared to number model parameters) for recovery. This increased efficiency extends to fitting unstructured data of varying dimensionality and when employing a variety of tensor network formats. Here, we show improved learning using the hierarchical Tucker method over the tensor-train in high-dimensional settings on a number of benchmark problems.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Sterile Neutrino Search with the NOvA Detectors

NOvA is a two-detector experiment designed to observe neutrino oscillations from νµ toνe flavor states. Additionally, the location of the NOvA Near Detector is well situated to additionally observe short base line oscillations to sterile neutrino states. We expand upon previous NOvA sterile neutrino searches by using a covariance matrix fitting technique. This analysis is the first with NOvA to directly include neutrino interactions in the Near Detector while fitting the data. The NOvA Near Detector’s greater statistics to better constrain our Far Detector uncertainties in addition to increasing the range of Δ m$^2_{41}$ to which we are sensitive.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2024

Determination of Neutrino Oscillation Parameters through the Feldman-Cousins Method by the NOvA Experiment

The NOvA experiment presents new measurements of the neutrino oscillation parameters obtained through a fit to data from the one megawatt NuMI neutrino beam in the NOvA detectors. The analysis uses muon-neutrino disappearance and electron-neutrino appearance in both neutrino and antineutrino beam polarities. With the addition of $\sim$ 100%\) more neutrino-mode beam exposure over the previously reported results, this analysis employs the unified approach of Feldman and Cousins to determine the confidence level intervals for the oscillation parameters $\theta_{23}, \delta_{CP}$, and $\Delta m_{32}^2$ for both neutrino mass orderings.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2024

Carrier Dynamics and Photoluminescence in Antimony-Doped CdTe and CdSeTe

Arsenic (As)-doped graded Cd(Se,Te) photovoltaics have achieved record efficiency of 22.4%, but further increases in efficiency have been limited by doping activation of only ~1% and radiative voltage losses of ~100 meV. Antimony (Sb) can also act as a p-type dopant of Cd(Se,Te) but has received relatively little attention. We measured carrier dynamics and photoluminescence of Sb-doped CdTe and CdSeTe films grown by vapor-transport deposition, with dopant activation up to ~20% Time resolved terahertz (TRTS) and photoluminescence (TRPL) data were fit by simulating the semiconductor equations, revealing bulk lifetimes of >20 ns in CdTe:Sb and >65 ns in CdSeTe:Sb and surface recombination velocities <1000 cm/s at alumina interfaces. We also qualitatively compare transients and PL spectra of doped and undoped films, with no apparent variation associated with hole concentrations up to 1015 cm-3. Preliminary variable temperature steady state photoluminescence (VT-SSPL) indicates dominant band-gap emission in CdSeTe:Sb and no evidence of potential fluctuations with Sb-doping up to this hole concentration.

14 SOLAR ENERGY↗

NASA NTRS1994

Calibration of the COBE FIRAS instrument

The Far-Infrared Absolute Spectrophotometer (FIRAS) instrument on the Cosmic Background Explorer (COBE) satellite was designed to accurately measure the spectrum of the cosmic microwave background radiation (CMBR) in the frequency range 1-95/cm with an angular resolution of 7 deg. We describe the calibration of this instrument, including the method of obtaining calibration data, reduction of data, the instrument model, fitting the model to the calibration data, and application of the resulting model solution to sky observations. The instrument model fits well for calibration data that resemble sky condition. The method of propagating detector noise through the calibration process to yield a covariance matrix of the calibrated sky data is described. The final uncertainties are variable both in frequency and position, but for a typical calibrated sky 2.6 deg square pixel and 0.7/cm spectral element the random detector noise limit is of order of a few times 10(exp -7) ergs/sq cm/s/sr cm for 2-20/cm, and the difference between the sky and the best-fit cosmic blackbody can be measured with a gain uncertainty of less than 3%.

Fixsen, D. J.↗

NASA NTRS2001

Wilkinson Microwave Anisotropy Probe (WMAP) First Year Observations: TE Polarization

The Wilkinson Microwave Anisotropy Probe (WMAP) has mapped the full sky in Stokes I, Q, and U parameters at frequencies 23, 33, 41, 61, and 94 GHz. We detect correlations between the temperature and polarization maps significant at more than 10 standard deviations. The correlations are inconsistent with instrument noise and are significantly larger than the upper limits established for potential systematic errors. The correlations are present in all WAMP frequency bands with similar amplitude from 23 to 94 GHz, and are consistent with a superposition of a CMB signal with a weak foreground. The fitted CMB component is robust against different data combinations and fitting techniques. On small angular scales (theta less than 5 deg), the WMAP data show the temperature-polarization correlation expected from adiabatic perturbations in the temperature power spectrum. The data for l greater than 20 agree well with the signal predicted solely from the temperature power spectra, with no additional free parameters. We detect excess power on large angular scales (theta greater than 10 deg) compared to predictions based on the temperature power spectra alone. The excess power is well described by reionization at redshift 11 is less than z(sub r) is less than 30 at 95% confidence, depending on the ionization history. A model-independent fit to reionization optical depth yields results consistent with the best-fit ACDM model, with best fit value t = 0.17 +/- 0.04 at 68% confidence, including systematic and foreground uncertainties. This value is larger than expected given the detection of a Gunn-Peterson trough in the absorption spectra of distant quasars, and implies that the universe has a complex ionization history: WMAP has detected the signal from an early epoch of reionization.

Kogut, A.↗

NASA NTRS2007

Numerical Technique for Analyzing Rotating Rake Mode Measurements in a Duct With Passive Treatment and Shear Flow

A technique is presented for the analysis of measured data obtained from a rotating microphone rake system. The system is designed to measure the interaction modes of ducted fans. A Fourier analysis of the data from the rotating system results in a set of circumferential mode levels at each radial location of a microphone inside the duct. Radial basis functions are then least-squares fit to this data to obtain the radial mode amplitudes. For ducts with soft walls and mean flow, the radial basis functions must be numerically computed. The linear companion matrix method is used to obtain both the eigenvalues of interest, without an initial guess, and the radial basis functions. The governing equations allow for the mean flow to have a boundary layer at the wall. In addition, a nonlinear least-squares method is used to adjust the wall impedance to best fit the data in an attempt to use the rotating system as an in-duct wall impedance measurement tool. Simulated and measured data are used to show the effects of wall impedance and mean flow on the computed results.

Dahl, Milo D.↗

DOE OSTI2022

Deep learning to estimate permeability using geophysical data

Time-lapse electrical resistivity tomography (ERT) is a popular geophysical method to estimate three-dimensional (3D) permeability fields from electrical potential difference measurements. Traditional inversion and data assimilation methods are used to ingest this ERT data into hydrogeophysical models to estimate permeability. Due to ill-posedness and the curse of dimensionality, existing inversion strategies provide poor estimates and low resolution of the 3D permeability field. Recent advances in deep learning provide us with powerful algorithms to overcome this challenge. This paper presents a deep learning (DL) framework to estimate the 3D subsurface permeability from time-lapse ERT data. To test the feasibility of the proposed framework, we train DL-enabled inverse models on simulation data. Each measurement in both synthetic and field data is standardized by removing the mean and scaling the time-series to unit variance. This pre-processing step is necessary to bring simulation data closer to field observations. Subsurface process models based on hydrogeophysics are used to generate this synthetic data. Training performed on limited simulation data resulted in the DL model over-fitting. An advanced data augmentation based on mixup is implemented to generate additional training samples to overcome this issue. This mixup technique creates weakly labeled (low-fidelity) samples from strongly labeled (high-fidelity) data. The weakly labeled training data is then used to develop DL-enabled inverse models and reduce over-fitting. As both time-lapse ERT (1133048 features/realization) and 3D permeability (585453 features/realization) data samples are from a high-dimensional space, principal component analysis (PCA) is employed to reduce dimensionality. Encoded ERT and encoded permeability are generated using the trained PCA estimators. A deep neural network is then trained to map the encoded ERT to encoded permeability. This mixup training and unsupervised learning allowed us to build a fast and reasonably accurate DL-based inverse model under limited simulation data. Results show that proposed weak supervised learning can capture salient spatial features in the 3D permeability field. Quantitatively, the average mean squared error (in terms of the natural log) on the strongly labeled training, validation, and test datasets is less than 0.5. The R 2 -score (global metric) is greater than 0.75, and the percent error in each cell (local metric) is less than 10%. Finally, an added benefit in terms of computational cost is that the proposed DL-based inverse model is at least O(10 4 ) times faster than running a forward model once it is trained. Data generation, DL model training, and hyperparameter tuning to identify optimal neural network architectures utilized high-performance computing resources while the DL inference is performed on a standard laptop. Approximately, O(10 5 ) processor hours are used for generating data and DL tuning and training. We acknowledge that the data generation and DL model development are expensive. But once a DL model is trained, it can be re-used for inversion rapidly for the given system, with set physics and domain. Note that traditional inversion may require multiple forward model simulations (e.g., in the order of 10 to 1000), which are very expensive. This computational savings ≈ O(10 5 ) – O(10 7 )) makes the proposed DL-based inverse model attractive for subsurface imaging and real-time ERT monitoring applications due to fast and yet reasonably accurate estimations of permeability field.

58 GEOSCIENCES↗