SEARCH · Search NASA

Results for “Methods: data analysis”

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 775 records · Page 43

All sources NASA NTRS DOE OSTI

DOE OSTI2021

Beyond optimization—supervised learning applications in relativistic laser-plasma experiments

We explore the applications of machine learning techniques in relativistic laser-plasma experiments beyond optimization purposes. We predict the beam charge of electrons produced in a laser wakefield accelerator given the laser wavefront change caused by a deformable mirror. Machine learning enables feature analysis beyond merely searching for an optimal beam charge, showing that specific aberrations in the laser wavefront are favored in generating higher beam charges. Supervised learning models allow characterizing the measured data quality as well as recognizing irreproducible data and potential outliers. Furthermore, we also include virtual measurement errors in the experimental data to examine the model robustness under these conditions. This work demonstrates how machine learning methods can benefit data analysis and physics interpretation in a highly nonlinear problem of relativistic laser-plasma interaction.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

NASA NTRS2019

Cloud Giovanni: Reining in Costs and Improving Performance with Analytical Data Stores Using Scalable Serverless Architecture

Giovanni is the Geospatial Interactive Online Visualization ANd aNalysis Infrastructure developed at NASA GES DISC which provides a simple and intuitive way to visualize, analyze, and access vast amounts of Earth science data. It receives large number of user requests each day for a variety of analysis and visualization services, which leads to the big data challenge of serving gradually increasing large data volumes with diverse statistical algorithms. We hereby propose a multi-dimensional accumulation method which provides fast and cost-efficient cloud analysis for diverse services including both area averaging and time averaging. This method involves the weighted volume integration over multiple variable dimensions (time and space), and is implemented in AWS using Athena providing serverless and highly scalable data analysis. Compared to the standard method, this approach dramatically reduced the computational time by order of magnitude with a minimal AWS cost incurred. For example, for a benchmark of 10-year area averaging over the 1x1 degree daily variable, the computational time was reduced from minutes to seconds, and the Athena cost is only $5 for 100,000 requests.

Zhang, Hailiang↗

DOE OSTI2017

Electric load shape benchmarking for small- and medium-sized commercial buildings

Small- and medium-sized commercial buildings owners and utility managers often look for opportunities for energy cost savings through energy efficiency and energy waste minimization. However, they currently lack easy access to low-cost tools that help interpret the massive amount of data needed to improve understanding of their energy use behaviors. Benchmarking is one of the techniques used in energy audits to identify which buildings are priorities for an energy analysis. Traditional energy performance indicators, such as the energy use intensity (annual energy per unit of floor area), consider only the total annual energy consumption, lacking consideration of the fluctuation of energy use behavior over time, which reveals the time of use information and represents distinct energy use behaviors during different time spans. To fill the gap, this study developed a general statistical method using 24-hour electric load shape benchmarking to compare a building or business/tenant space against peers. Specifically, the study developed new forms of benchmarking metrics and data analysis methods to infer the energy performance of a building based on its load shape. We first performed a data experiment with collected smart meter data using over 2,000 small- and medium-sized businesses in California. We then conducted a cluster analysis of the source data, and determined and interpreted the load shape features and parameters with peer group analysis. Finally, we implemented the load shape benchmarking feature in an open-access web-based toolkit (the Commercial Building Energy Saver) to provide straightforward and practical recommendations to users. The analysis techniques were generic and flexible for future datasets of other building types and in other utility territories.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

DOE OSTI2025

A cross-dimensional analysis of data-driven short-term load forecasting methods with large-scale smart meter data

Electricity load forecasting is essential to utility operation and power grid stability. A wide spectrum of data-driven methods, ranging from linear regression models to more recent deep learning models have been adopted to forecast electric load over the years. However, there still lacks a holistic evaluation of the applicability of conventional statistical and machine learning based algorithms with respect to different temporal and spatial scopes, computational requirements, and sensitivity of model-tuning. Enabled by a large-scale electricity load profile dataset of over 40,000 residential customers in a utility region, we conducted a cross-dimensional analysis of data-driven load forecasting methods. Three regression-based and seven deep learning algorithms with different model configurations were evaluated in terms of their overall and peak load prediction accuracy, and training burdens, across spatial aggregation levels ranging from the transformer, feeder, substation, to neighborhood. We found, first, the load forecasting accuracy is constrained by a predictability boundary, influenced by the forecasting horizon and spatial aggregation level. Specifically, RandomForest, XGBoost, TFT, TSMixer, and TiDE models achieved less than 10 % prediction error for up to 96-h ahead forecasting for district, substation, and feeder levels, while other models struggle at long-horizon predictions; Second, for winter and summer peak load dates, most models were able to predict the peak demand timing within ± 1 h, but the prediction percentage error varied by models, with TFT and TiDE models being the top performers; Third, models with similar prediction accuracy can differ in training burden by an order of magnitude. Therefore, choosing model configurations that balance prediction performance and computational resource is an important practical consideration for large-scale deployment of the machine learning based load forecasting. The outcome of this study can guide researchers and practitioners to choose the proper load forecasting algorithms based on their problem scope, required accuracy, and available resources. The predictability boundary can serve as a benchmark for electricity load forecasting problems with new algorithms and datasets.

Li, Han↗

NASA NTRS1990

Information and problem report usage in system saftey engineering division

Five basic problems or question areas are examined. They are as follows: (1) Evaluate adequacy of current problem/performance data base; (2) Evaluate methods of performing trend analysis; (3) Methods and sources of data for probabilistic risk assessment; and (4) How is risk assessment documentation upgraded and/or updated. The fifth problem was to provide recommendations for each of the above four areas.

Morrissey, Stephen J.↗

NASA NTRS1986

Sequential estimation and satellite data assimilation in meteorology and oceanography

The central theme of this review article is the role that dynamics plays in estimating the state of the atmosphere and of the ocean from incomplete and noisy data. Objective analysis and inverse methods represent an attempt at relying mostly on the data and minimizing the role of dynamics in the estimation. Four-dimensional data assimilation tries to balance properly the roles of dynamical and observational information. Sequential estimation is presented as the proper framework for understanding this balance, and the Kalman filter as the ideal, optimal procedure for data assimilation. The optimal filter computes forecast error covariances of a given atmospheric or oceanic model exactly, and hence data assimilation should be closely connected with predictability studies. This connection is described, and consequences drawn for currently active areas of the atmospheric and oceanic sciences, namely, mesoscale meteorology, medium and long-range forecasting, and upper-ocean dynamics.

Ghil, M.↗

DOE OSTI2014

Solar flare composition and thermodynamics from RESIK X-ray spectra

Previous estimates of the solar flare abundances of Si, S, Cl, Ar, and K from the RESIK X-ray crystal spectrometer on board the CORONAS-F spacecraft were made on the assumption of isothermal X-ray emission. We investigate the effect on these estimates by relaxing this assumption and instead determining the differential emission measure (DEM) or thermal structure of the emitting plasma by re-analyzing RESIK data for a GOES class M1.0 flare on 2002 November 14 (SOL2002-11-14T22:26) for which there was good data coverage. The analysis method uses a maximum-likelihood (Withbroe-Sylwester) routine for evaluating the DEM. In a first step, called here AbuOpt, an optimized set of abundances of Si, S, Ar, and K is found that is consistent with the observed spectra. With these abundances, the DEM evolution during the flare is found. The abundance optimization leads to revised abundances of silicon and sulfur in the flare plasma: A(S) = 6.94 ± 0.06 and A(Si) = 7.56 ± 0.08 (on a logarithmic scale with A(H) = 12). Previously determined abundances of Ar, K, and Cl from an isothermal assumption are still the preferred values. During the flare's maximum phase, the X-ray-emitting plasma has a basically two-temperature structure, with the cooler plasma with approximately constant temperature (3-6 MK) and a hotter plasma with temperature 16-21 MK. Using imaging data from the RHESSI hard X-ray spacecraft, the emission volume of the hot plasma is deduced from which lower limits of the electron density N{sub e} and the thermal content of the plasma are given.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRS2021

Analysis of Nonlinear Shrinkage for the Bound Metal Deposition Manufacturing using Multi-scale Approach

We consider problem of nonlinear shrinkage of the metal part during bound metal deposition manufacturing on the ground and in zero-G. To analyze this problem we developed multi-scale physics-based approach that spans atomistic dynamics at the scale of nanoseconds and the full part shrinkage at the time scale of hours. Using this approach we estimated the key parameters of the problem including grain boundary width, coefficient of surface diffusion, initial redistribution of particles during debinding stage, micro-structure evolution from round particles to densely packed grains and corresponding change of the total and chemical free energy, and sintering stress. The introduced method was used to predict shrinkage at the level of two particles, filament cross-section, sub-model, and the whole green, brown, and metal parts. To further improve accuracy and reliability of the shrinkage predictions we propose concept of intelligent additive manufacturing of metal powders in space that combines the strengths of both physics-based and data-driven methods of analysis of AM.

bound metal deposition↗

NASA NTRS1990

Buckling analysis and test correlation of hat stiffened panels for hypersonic vehicles

The paper discusses the design, analysis, and test of hat stiffened panels subjected to a variety of thermal and mechanical load conditions. The panels were designed using data from structural optimization computer codes and finite element analysis. Test methods included the grid shadow moire method and a single gage force stiffness method. The agreement between the test data and analysis provides confidence in the methods that are currently being used to design structures for hypersonic vehicles. The agreement also indicates that post buckled strength may potentially be used to reduce the vehicle weight.

Percy, Wendy C.↗

DOE OSTI2013

Validation of FSP Reactor Design with Sensitivity Studies of Beryllium-Reflected Critical Assemblies

The baseline design for space nuclear power is a fission surface power (FSP) system: sodium-potassium (NaK) cooled, fast spectrum reactor with highly-enriched-uranium (HEU)-O 2 fuel, stainless steel (SS) cladding, and beryllium reflectors with B4C control drums. Previous studies were performed to evaluate modeling capabilities and quantify uncertainties and biases associated with analysis methods and nuclear data. Comparison of Zero Power Plutonium Reactor (ZPPR)-20 benchmark experiments with the FSP design indicated that further reduction of the total design model uncertainty requires the reduction in uncertainties pertaining to beryllium and uranium cross-section data. Further comparison with three beryllium-reflected HEU-metal benchmark experiments performed at the Oak Ridge Critical Experiments Facility (ORCEF) concluded the requirement that experimental validation data have similar cross section sensitivities to those found in the FSP design. A series of critical experiments was performed at ORCEF in the 1960s to support the Medium Power Reactor Experiment (MPRE) space reactor design. The small, compact critical assembly (SCCA) experiments were graphite- or beryllium-reflected assemblies of SS-clad, HEU-O 2 fuel on a vertical lift machine. All five configurations were evaluated as benchmarks. Two of the five configurations were beryllium reflected, and further evaluated using the sensitivity and uncertainty analysis capabilities of SCALE 6.1. Validation of the example FSP design model was successful in reducing the primary uncertainty constituent, the Be(n,n) reaction, from 0.28 %$\delta$k/k to 0.0004 %$\delta$k/k. Further assessment of additional reactor physics measurements performed on the SCCA experiments may serve to further validate FSP design and operation.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

DOE OSTI2018

First‐principles study in an inter‐granular glassy film model of silicon nitride

Abstract Based on a previously constructed intergranular glassy film ( IGF ) model for bulk silicon nitride, a large periodic model of 3864 atoms containing 2 grains of different orientations from the main bulk β‐Si 3 N 4 and 2 IGF s was fully relaxed using the ab initio density‐functional theory package VASP . The relaxed structure was then used to calculate the electronic structure, density of states, interatomic bonding, partial charge distribution, and electron localization using the OLCAO method. Analysis of the data focuses on the interfacial regions between bulk β‐Si 3 N 4 and the Si‐O‐N glass layer with different orientations. The total bond order density ( TBOD ) is evaluated in different interfacial and bulk regions. We show minor differences in the internal cohesion between crystalline grains of different facets. However, the overall charges in the bulk crystal grains and in the glassy regions are electropositive which are balanced by the negatively charged interfacial region between the two. The presence of a less rigid glassy layer is the reason for structural flexibility in ceramics without a huge penalty in the steric energy. The optimization of the interfacial structure and bonding via the creation of defective sites is the atomic origin for the existence of the IGF in silicon nitride. The insights obtained from this detailed quantum mechanical analysis of a realistic IGF model are discussed including implications on the strength, fracture toughness, and processing methods. We also discuss the potential applications of our method to other complex materials systems.

Ching, Wai‐Yim↗

DOE OSTI2013

Measurement of baryon acoustic oscillations in the Lyman-α forest fluctuations in BOSS data release 9

We use the Baryon Oscillation Spectroscopic Survey (BOSS) Data Release 9 (DR9) to detect and measure the position of the Baryonic Acoustic Oscillation (BAO) feature in the three-dimensional correlation function in the Lyman-α flux fluctuations at a redshift z{sub eff} = 2.4. The feature is clearly detected at significance between 3 and 5 sigma (depending on the broadband model and method of error covariance matrix estimation) and is consistent with predictions of the standard ΛCDM model. We assess the biases in our method, stability of the error covariance matrix and possible systematic effects. We fit the resulting correlation function with several models that decouple the broadband and acoustic scale information. For an isotropic dilation factor, we measure 100 × (α{sub iso} − 1) = −1.6{sup +2.0+4.3+7.4}{sub −2.0−4.1−6.8} (stat.) ±1.0 (syst.) (multiple statistical errors denote 1,2 and 3 sigma confidence limits) with respect to the acoustic scale in the fiducial cosmological model (flat ΛCDM with Ω{sub m} = 0.27, h = 0.7). When fitting separately for the radial and transversal dilation factors we find marginalised constraints 100 × (α{sub ||} − 1) = −1.3{sup +3.5+7.6+12.3}{sub −3.3−6.7−10.2} (stat.) ±2.0 (syst.) and 100 × (α{sub p}erpendicular − 1) = −2.2{sup +7.4+17}{sub −7.1−15} (stat.) ±3.0 (syst.). The dilation factor measurements are significantly correlated with cross-correlation coefficient of ∼ −0.55. Errors become significantly non-Gaussian for deviations over 3 standard deviations from best fit value. Because of the data cuts and analysis method, these measurements give tighter constraints than a previous BAO analysis of the BOSS DR9 Lyman-α sample, providing an important consistency test of the standard cosmological model in a new redshift regime.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRS2017

Methods for Probabilistic Uncertainty Analysis and Bayesian Analysis with Examples of Statistically Analyzing Data to Revise MMOD Risk Estimates and Compare Models

Probabilistic methods are presented for characterizing and quantifying uncertainties in models and model predictions. Techniques are given for constructing specific uncertainty distributions based on available information. Alternative techniques are given for propagating uncertainties in model inputs to obtain the uncertainty in the model result or prediction. Bayesian techniques are also described for utilizing data and information to update and revise model results and predictions. The focus is on applications with numerous specific examples given. The use of data to revise Micrometeoroid and Orbital Debris (MMOD) risk prediction models are among the examples given.

Risk↗

DOE OSTI2018

Relationships between physical–geographical factors and soil degradation on agricultural land

Highlights: • Connection of soil degradation to seven physical-geographical variables was revealed. • New method of threshold data analysis was proposed. • Precise treshold values where soil degradation starts and ends were revealed. • Study area covered the entire area of Czech agricultural land – 13027 cadastres (78,866 km{sup 2}). It is a well-known fact that soil degradation is dramatically increasing and currently threatens agricultural soils all around the world. The objective of this study was to reveal the possible connection between soil degradation and seven physical-geographical factors - slope steepness, altitude, elevation differences, rainfall, temperature, soil texture and solar radiation – in the form of threshold values (if these exist), where soil degradation begins and ends. The analysis involved the whole area of the Czech Republic which consists of 13,027 cadasters (78,866 km{sup 2}). The greatest total degradation threat occurs in areas with slope steepness >7 degrees, average annual temperature 10.54, altitude >766 m a.s.l. Similarly, the results for water erosion, wind erosion, soil compaction, loss of organic matter, acidification and heavy metal contamination were processed. The results enable us to identify the relationships of different levels of threats which could consequently be used in various ways – for classification of threatened areas, for more effective implementation of anti-degradation measures, or purely for a better understanding of the role of physical geographical factors in soil degradation in the Czech Republic, and thus could increase the chances of reducing vulnerability to land degradation not only in the Czech Republic.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2023

X-ray Absorption Spectroscopy Studies of a Molecular CO 2 -Reduction Catalyst Deposited on Graphitic Carbon Nitride

Metal-ligand complexes have been extensively explored as well-defined molecular catalysts in small molecule activation reactions such as carbon dioxide (CO 2 ) reduction. Many hybrid photocatalysts have been prepared by coupling such complexes with photoactive surfaces for use in solar CO 2 reduction. In this work, we employ X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) spectroscopies, density functional theory (DFT) and computational XANES modeling to interrogate the structure of a hybrid photocatalyst consisting of a macrocyclic cobalt complex deposited on graphitic carbon nitride (C 3 N 4 ). Results show that the cobalt complex binds on C 3 N 4 through surface OH or NH 2 groups. By refining the local geometry and binding sites of this well-defined molecular cobalt complex on C 3 N 4 , here we established an important benchmark for modeling a large class of molecular catalysts that can be adapted to in situ/operando studies and further enhanced by applying chemometrics-based approaches and machine learning methods of XANES data analysis.

36 MATERIALS SCIENCE↗

DOE OSTI2019

Convolutional neural networks on the HEALPix sphere: a pixel-based algorithm and its application to CMB data analysis

We describe a novel method for the application of convolutional neural networks (CNNs) to fields defined on the sphere, using the Hierarchical Equal Area Latitude Pixelization scheme (HEALPix). Specifically, we have developed a pixel-based approach to implement convolutional and pooling layers on the spherical surface, similarly to what is commonly done for CNNs applied to Euclidean space. The main advantage of our algorithm is to be fully integrable with existing, highly optimized libraries for NNs (e.g., PyTorch, TensorFlow, etc.). We present two applications of our method: (i) recognition of handwritten digits projected on the sphere; (ii) estimation of cosmological parameter from simulated maps of the cosmic microwave background (CMB). The latter represents the main target of this exploratory work, whose goal is to show the applicability of our CNN to CMB parameter estimation. We have built a simple NN architecture, consisting of four convolutional and pooling layers, and we have used it for all the applications explored herein. Concerning the recognition of handwritten digits, our CNN reaches an accuracy of ~95%, comparable with other existing spherical CNNs, and this is true regardless of the position and orientation of the image on the sphere. For CMB-related applications, we tested the CNN on the estimation of a mock cosmological parameter, defining the angular scale at which the power spectrum of a Gaussian field projected on the sphere peaks. We estimated the value of this parameter directly from simulated maps, in several cases: temperature and polarization maps, presence of white noise, and partially covered maps. For temperature maps, the NN performances are comparable with those from standard spectrum-based Bayesian methods. For polarization, CNNs perform about a factor four worse than standard algorithms. Nonetheless, our results demonstrate, for the first time, that CNNs are able to extract information from polarization fields, both in full-sky and masked maps, and to distinguish between E and B -modes in pixel space. Lastly, we have applied our CNN to the estimation of the Thomson scattering optical depth at reionization ( τ ) from simulated CMB maps. Even without any specific optimization of the NN architecture, we reach an accuracy comparable with standard Bayesian methods. This work represents a first step towards the exploitation of NNs in CMB parameter estimation and demonstrates the feasibility of our approach.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2014

Shock timing measurements and analysis in deuterium-tritium-ice layered capsule implosions on NIF

Recent advances in shock timing experiments and analysis techniques now enable shock measurements to be performed in cryogenic deuterium-tritium (DT) ice layered capsule implosions on the National Ignition Facility (NIF). Previous measurements of shock timing in inertial confinement fusion implosions [Boehly et al., Phys. Rev. Lett. 106, 195005 (2011); Robey et al., Phys. Rev. Lett. 108, 215004 (2012)] were performed in surrogate targets, where the solid DT ice shell and central DT gas were replaced with a continuous liquid deuterium (D2) fill. These previous experiments pose two surrogacy issues: a material surrogacy due to the difference of species (D2 vs. DT) and densities of the materials used and a geometric surrogacy due to presence of an additional interface (ice/gas) previously absent in the liquid-filled targets. This report presents experimental data and a new analysis method for validating the assumptions underlying this surrogate technique. Comparison of the data with simulation shows good agreement for the timing of the first three shocks, but reveals a considerable discrepancy in the timing of the 4th shock in DT ice layered implosions. Electron preheat is examined as a potential cause of the observed discrepancy in the 4th shock timing.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2016

Application of photon Doppler velocimetry to direct impact Hopkinson pressure bars

Direct impact Hopkinson pressure bar systems offer many potential advantages over split Hopkinson pressure bars, including access to higher strain rates, higher strains for equivalent striker velocity and system length, lower dispersion, and faster achievement of force equilibrium. Currently, these advantages are gained at the expense of all information about the striker impacted specimen face, preventing the experimental determination of force equilibrium, and requiring approximations to be made on the sample deformation history. In this paper, we discuss an experimental method and complementary data analysis for using photon Doppler velocimetry to measure surface velocities of the striker and output bars in a direct impact bar experiment, allowing similar data to be recorded as in a split bar system. We discuss extracting velocity and force measurements, and the precision of measurements. Results obtained using the technique are compared to equivalent split bar tests, showing improved stress measurements for the lowest and highest strains in fully dense metals, and improvement for all strains in slow and non-equilibrating materials.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗