Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “error estimation and 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.

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

At least 937 records · Page 52
All sourcesNASA NTRSDOE OSTI
NASA NTRS1998

In-Flight System Identification

A method is proposed and studied whereby the system identification cycle consisting of experiment design and data analysis can be repeatedly implemented aboard a test aircraft in real time. This adaptive in-flight system identification scheme has many advantages, including increased flight test efficiency, adaptability to dynamic characteristics that are imperfectly known a priori, in-flight improvement of data quality through iterative input design, and immediate feedback of the quality of flight test results. The technique uses equation error in the frequency domain with a recursive Fourier transform for the real time data analysis, and simple design methods employing square wave input forms to design the test inputs in flight. Simulation examples are used to demonstrate that the technique produces increasingly accurate model parameter estimates resulting from sequentially designed and implemented flight test maneuvers. The method has reasonable computational requirements, and could be implemented aboard an aircraft in real time.

Morelli, Eugene A.↗
NASA NTRS2003

Karhunen-Loeve Estimation of the Power Spectrum Parameters from the Angular Distribution of Galaxies in Early Sloan Digital Sky Survey Data

We present measurements of parameters of the three-dimensional power spectrum of galaxy clustering from 222 square degrees of early imaging data in the Sloan Digital Sky Survey (SDSS). The projected galaxy distribution on the sky is expanded over a set of Karhunen-Loeve (KL) eigenfunctions, which optimize the signal-to-noise ratio in our analysis. A maximum likelihood analysis is used to estimate parameters that set the shape and amplitude of the three-dimensional power spectrum of galaxies in the SDSS magnitude-limited sample with r* less than 21. Our best estimates are gamma = 0.188 +/- 0.04 and sigma(sub 8L) = 0.915 +/- 0.06 (statistical errors only), for a flat universe with a cosmological constant. We demonstrate that our measurements contain signal from scales at or beyond the peak of the three-dimensional power spectrum. We discuss how the results scale with systematic uncertainties, like the radial selection function. We find that the central values satisfy the analytically estimated scaling relation. We have also explored the effects of evolutionary corrections, various truncations of the KL basis, seeing, sample size, and limiting magnitude. We find that the impact of most of these uncertainties stay within the 2 sigma uncertainties of our fiducial result.

Szalay, Alexander S.↗
NASA NTRS2019

Uncertainty Quantification for JPL Retrievals

We define the notions of verification, validation, and uncertainty quantification (VVUQ), and place them in the context of retrievals of Earth science data. Uncertainty quantification (UQ) is the process of characterizing the uncertainty of a measurement system relative to the true, physical quantity it measures. UQ is needed when measurements are used for scientific inference, assimilated into models with other data, and used for decision-making and risk analysis. Many measurements come with measures of standard error and bias, but these uncertainty measures are often inaccurate, due to the operational need to assume crisp values for parameters which are in fact uncertain, as well as the failure of technical assumptions needed by operational error-estimation approaches. UQ provides a way to study which assumptions fail to hold, and a tool to improve the measurement. We describe a particular, broadly-applicable UQ setup that has been successful for JPL retrievals including OCO-2, and provide recommendations for furthering the UQ state of the practice at JPL.

Turmon, Michael↗
NASA NTRS2020

Multi-Objective Gust Load Alleviation Control Designs for an Aeroelastic Wind Tunnel Demonstration Wing

This paper presents several control and gust disturbance estimation techniques applied to a mathematical model of a physical flexible wing wind tunnel model used in ongoing tests at the University of Washington Aeronautical Laboratory's Kirsten Wind Tunnel. Three methods of gust disturbance estimation are presented, followed by three control methods: LQG, Basic Multi-Objective (BMO), and a novel Multi-Objective Prediction Correction (MOPC) controller. The latter of which augments a multi-objective controller, and attempts to correct for errors in the disturbance estimate. A simplified linear simulation of the three controllers is performed and a simple MIMO stability and robustness assessment is performed. Then, the same controllers are simulated in a higher fidelity Simulink environment that captures sampling, saturation and noise effects. This preliminary analysis indicates that the BMO controller provides the best performance and largest stability margins.

Drew, Michael C.↗
DOE OSTI2025

Empirically-calibrated H100 node power models for accurate AI training energy estimation

Accurately quantifying the energy use of artificial intelligence (AI) training is critical for infrastructure planning, carbon accounting, and sustainable data center operation, but few studies have directly measured the power consumption of production workloads on contemporary hardware. By combining empirical measurements from Brookhaven National Laboratory during AI training on 8-graphics-processing-unit H100 systems with open-source benchmarking data, we develop statistical models relating computational intensity to node-level power consumption. We measure the gap between manufacturer-rated thermal design power (TDP) and actual power demand during AI training. Our analysis reveals that even computationally intensive workloads operate at only 76% of the 10.2 kW TDP rating. Our architecture-specific model, calibrated to floating-point operations, predicts energy consumption with 11.4% mean absolute percentage error, significantly outperforming TDP-based approaches (27%–37% error). We identified distinct power signatures between transformer and convolutional neural network architectures, with transformers showing characteristic fluctuations that may impact grid stability. These results provide a measurement-grounded basis for improving AI training energy estimates, enabling more reliable infrastructure sizing, cost projections, and environmental impact assessments.

Newkirk, Alex C↗
DOE OSTI2025

Brillouin Sensing with PCA, and PCA-Based Neural Networks for Efficient Temperature Monitoring

This work explores peak estimation techniques in Brillouin Optical Time Domain Analysis (BOTDA), emphasizing both accuracy and efficiency. Euclidean distance measurement method is applied to principal components derived from Brillouin Gain Spectrum data. It offers a major speed advantage being 180 170 times faster than traditional curve fitting methods such as Lorentzian curve fitting, while maintaining similar accuracy. Additionally, a PCA- based neural network model shows significant reduction of peak estimation time compared to Lorentzian fitting. Results show Brillouin frequency shift errors lie under 0.75 MHz in both Euclidean distance-based and neural network-based methods, both of which utilize PCA components. For large data sets and long length fibers, PCA- assisted neural network for peak estimation would be an efficient solution.

Distributed optical fiber sensing↗
NASA NTRS2003

First Year Wilkinson Microwave Anisotropy Probe(WMAP) Observations: Data Processing Methods and Systematic Errors Limits

We describe the calibration and data processing methods used to generate full-sky maps of the cosmic microwave background (CMB) from the first year of Wilkinson Microwave Anisotropy Probe (WMAP) observations. Detailed limits on residual systematic errors are assigned based largely on analyses of the flight data supplemented, where necessary, with results from ground tests. The data are calibrated in flight using the dipole modulation of the CMB due to the observatory's motion around the Sun. This constitutes a full-beam calibration source. An iterative algorithm simultaneously fits the time-ordered data to obtain calibration parameters and pixelized sky map temperatures. The noise properties are determined by analyzing the time-ordered data with this sky signal estimate subtracted. Based on this, we apply a pre-whitening filter to the time-ordered data to remove a low level of l/f noise. We infer and correct for a small (approx. 1 %) transmission imbalance between the two sky inputs to each differential radiometer, and we subtract a small sidelobe correction from the 23 GHz (K band) map prior to further analysis. No other systematic error corrections are applied to the data. Calibration and baseline artifacts, including the response to environmental perturbations, are negligible. Systematic uncertainties are comparable to statistical uncertainties in the characterization of the beam response. Both are accounted for in the covariance matrix of the window function and are propagated to uncertainties in the final power spectrum. We characterize the combined upper limits to residual systematic uncertainties through the pixel covariance matrix.

Hinshaw, G.↗
DOE OSTI2020

Leveraging Machine Learning Capabilities for the Characterization of Irradiated Uranium: A Case Study of Analysis Methods for Nuclear Safeguards and Nuclear Forensics

Nondestructively determining the initial enrichment of irradiated uranium is a complex and laborious multivariable problem due to the presence of fission products. This work demonstrates the capabilities of machine learning to analyze gamma-ray spectral data to determine initial enrichment without knowledge of the decay time of the sample. The approach developed is agnostic to the particular scenario and is applicable to a wide variety of applications in nuclear forensics and nuclear safeguards. We irradiated 5 mg uranium standard reference materials at discrete enrichment values ranging from 0.02% to 97% 235 U (weight percent) in UT Austin’s Nuclear Engineering Teaching Laboratory TRIGA Mark II 1.1 MW research reactor, allowed each to decay for 8 hours, and then measured each sample via gamma-ray spectrometry for 50 hours post-irradiation yielding 1,400 individual gamma-ray spectra discretized into 8,192 energy bins. We then trained decision trees models to analyze individual gamma-ray spectra and estimate the associated initial enrichment without knowledge of the time since end of irradiation. We evaluated the performance of the models with a reserved test set not used for training or calibrating the model. A decision tree model constructed with this procedure achieved a mean absolute error in initial enrichment determination of 2.3% (weight percent 235 U). Next, we implemented a principal component analysis pre-processing routine of the gamma-ray spectrometry data to reduce the dimensionality of the dataset from 8,192 channels in the spectrum to 10 principal components while retaining over 99% of the inherent variance in the data. Decision tree models constructed with these data demonstrated decreased mean absolute error in enrichment determination, reduced computational time, and decreased complexity. A single decision tree model constructed with this procedure achieved a mean absolute error in initial enrichment determination of 0.05% (weight percent 235 U). Furthermore, we analyzed these models with learning curves to ensure that overfitting did not occur. The capabilities provided by these models can be naturally extended to other application-focused measurements in the fields of nuclear safeguards, nuclear forensics, and nuclear non-proliferation.

Drescher, Adam↗
NASA NTRS2024

Modeling Entanglement-Based Quantum Key Distribution for the NASA Quantum Communications Analysis Suite

One of the most practical, and sought after, applications of quantum mechanics in the field of information science is the use of entanglement distribution to communicate quantum information effectively. Similar to the continued improvements of functional quantum computers over the past decade, advances in demonstrations of entanglement distribution over long distances may enable new applications in aeronautics and space communications. The existing NASA Quantum Communications Analysis Suite (NQCAS) software models such applications, but limited experimental data exists to verify the model’s theoretical results. There is, however, a large body of experimental data in the relevant literature for entanglement-based quantum key distribution (QKD). This paper details a Monte Carlo based QKD model that uses NQCAS input parameters to generate an estimated QKD link budget for verification of NQCAS. The model generates link budget statistics like key rates, error rates, and S values that can then be compared to the experimental values in the literature. Preliminary comparisons show many similarities between the simulated and experimental data, supporting the model’s validity. A verified NQCAS model will inform experimental work conducted in Glenn Research Center’s (GRC) NASA Quantum Metrology Laboratory (NQML), supporting the United States Quantum Initiative and potential NASA missions.

NASA Quantum Communications Analysis Suite↗
NASA NTRS2004

Error Analysis for Discontinuous Galerkin Method for Parabolic Problems

In the proposal, the following three objectives are stated: (1) A p-version of the discontinuous Galerkin method for a one dimensional parabolic problem will be established. It should be recalled that the h-version in space was used for the discontinuous Galerkin method. An a priori error estimate as well as a posteriori estimate of this p-finite element discontinuous Galerkin method will be given. (2) The parameter alpha that describes the behavior double vertical line u(sub t)(t) double vertical line 2 was computed exactly. This was made feasible because of the explicitly specified initial condition. For practical heat transfer problems, the initial condition may have to be approximated. Also, if the parabolic problem is proposed on a multi-dimensional region, the parameter alpha, for most cases, would be difficult to compute exactly even in the case that the initial condition is known exactly. The second objective of this proposed research is to establish a method to estimate this parameter. This will be done by computing two discontinuous Galerkin approximate solutions at two different time steps starting from the initial time and use them to derive alpha. (3) The third objective is to consider the heat transfer problem over a two dimensional thin plate. The technique developed by Vogelius and Babuska will be used to establish a discontinuous Galerkin method in which the p-element will be used for through thickness approximation. This h-p finite element approach, that results in a dimensional reduction method, was used for elliptic problems, but the application appears new for the parabolic problem. The dimension reduction method will be discussed together with the time discretization method.

Kaneko, Hideaki↗
DOE OSTI2025

SAXS Assistant: Automated SAXS analysis for structural discovery in biologics and polymeric nanoparticles

Small-angle x-ray scattering (SAXS) is a powerful technique for assessing macromolecular structure. High-throughput SAXS is limited by the time-consuming and, at times, subjective nature of SAXS data interpretation. Here, we present SAXS Assistant, a Python-based script that streamlines SAXS data analysis to extract features for machine learning (ML) and key structural parameters, including the Guinier radius of gyration (R g ), pair distance distribution function (PDDF)-derived R g , maximum particle dimension (D max ), and Kratky plots. The script builds upon BioXTAS RAW and validates reliability via Guinier/PDDF R g agreement, an important indicator of well-measured data sets. For assistance in D max estimation, a multilayer perceptron regressor was trained with 1940 data files from the Small Angle Scattering Biological Data Bank. The model achieved a test set performance R 2 = 0.90 and mean absolute error = 11.7 Å. Training exclusively with experimental data translates analyses from researchers, including experts in the field, to the ML model, which helps assess D max estimations from PDDF. Gaussian mixture model clustering was implemented to classify profiles into structural classes based on entries in the Small Angle Scattering Biological Data Bank. Users may therefore assess the similarity between experimental samples and known biomolecular shapes within the mapped repository entries. This probabilistic clustering aids in quantifying information from Kratky and generating shape-descriptive features. SAXS Assistant accelerates SAXS data analysis through enforced quality control, ML-ready outputs, and flags for low-confidence results. In addition to providing the ability to analyze large data sets at high throughput, this tool is versatile and may serve researchers in both biological and synthetic polymer research fields.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Decision Science for Machine Learning (DeSciML)

The increasing use of machine learning (ML) models to support high-consequence decision making drives a need to increase the rigor of ML-based decision making. Critical problems ranging from climate change to nonproliferation monitoring rely on machine learning for aspects of their analyses. Likewise, future technologies, such as incorporation of data-driven methods into the stockpile surveillance and predictive failure analysis for weapons components, will all rely on decision-making that incorporates the output of machine learning models. In this project, our main focus was the development of decision scientific methods that combine uncertainty estimates for machine learning predictions, with a domain-specific model of error costs. Other focus areas include uncertainty measurement in ML predictions, designing decision rules using multiobjecive optimization, the value of uncertainty reduction, and decision-tailored uncertainty quantification for probability estimates. By laying foundations for rigorous decision making based on the predictions of machine learning models, these approaches are directly relevant to every national security mission that applies, or will apply, machine learning to data, most of which entail some decision context.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Cladding Profilometry Analysis of Experimental Breeder Reactor-II Metallic Fuel Pins with HT9, D9, and SS316 Cladding

BISON finite element method fuel performance simulations were conducted using an existing automated process that couples the Fuels Irradiation & Physics Database (FIPD) and the Integral Fast Reactor Materials Information System database by writing input files and comparing the BISON output to post-irradiation fuel pin profilometry measurements contained within the databases. The importance of this work is to demonstrate the ability to benchmark fuel performance metallic fuel models within BISON using Experimental Breeder Reactor-II fuel pin data for a number of similar pins, while building off previous modeling efforts. Changes to the generic BISON input file include implementing pin specific axial power and flux profiles, pin specific fluences, frictional contact, and irradiation-induced volumetric swelling models for cladding. A statistical analysis of irradiation-induced volumetric swelling models for HT9, D9, and SS316 was performed for experiments X421/X421A, X441/X441A, and X486. Between these three experiments, there were 174 post-irradiation examination (PIE) profilometries used for validating the swelling models presented using a standard error of the estimate (SEE) method. Implementation of the volumetric swelling models for D9 and SS316 claddings was found to have a significant impact on the BISON profilometry simulated, where HT9 clad pins had an insignificant change due to low fluence values. BISON profilometry simulated for HT9, D9, and SS316 fuel pins agreed with PIE profilometry measurements, with assembly SEE values being 4.4 × 10−3 for X421A, 2.0 × 10−3 for X441A, and 2.8 × 10−3 for X486. D9 clad pins in X421/X421A had the highest SEE values, which is due to the BISON simulated profilometry being shifted axially. While this work accomplished its purpose to demonstrate the modeling of multiple fuel pins from the databases to help validate models, the results suggest that the continued development of metallic fuel models is necessary for qualifying new metallic fuel systems to better capture some physical performance phenomena, such as the hot pressing of U-Pu-Zr and the fuel cladding chemical interaction.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
NASA NTRS1978

Urine monitoring system failure analysis and operational verification test report

Failure analysis and testing of a prototype urine monitoring system (UMS) are reported. System performance was characterized by a regression formula developed from volume measurement test data. When the volume measurement test data. When the volume measurement data was imputted to the formula, the standard error of the estimate calculated using the regression formula was found to be within 1.524% of the mean of the mass of the input. System repeatability was found to be somewhat dependent upon the residual volume of the system and the evaporation of fluid from the separator. The evaporation rate was determined to be approximately 1cc/minute. The residual volume in the UMS was determined by measuring the concentration of LiCl in the flush water. Observed results indicated residual levels in the range of 9-10ml, however, results obtained during the flushing efficiency test indicated a residual level of approximately 20ml. It is recommended that the phase separator pumpout time be extended or the design modified to minimize the residual level.

Glanfield, E. J.↗
NASA NTRS1983

The use of LANDSAT-4 MSS digital data in temporal data sets and the evaluation of scene-to-scene registration accuracy

The MSS sensor on LANDSAT 4 is, in certain performance aspects, different from those on LANDSATS 1 through 3. These differences created some concern in the NASA research community as to whether individual data sets can be registered accurately enough to produce acceptable data sets for multitemporal data analysis. The use of LANDSAT 4 MSS digital data in temporal data sets is examined and a method is presented for estimating temporal registration accuracy based on the use of an X-Y digitizer and grey tone electrostatic plots. Results indicate that the RMS temporal registration errors are not significantly different from the temporal data sets generated using LANDSAT 4 and LANDSAT 2 data (33.35 meters) and the temporal data set constructed from two LANDSAT 2 data sets (33.61 meters). A derivation of the model used to evaluate the temporal registration is included.

Anderson, J. E.↗
NASA NTRS1985

Identification of dynamic systems, theory and formulation

The problem of estimating parameters of dynamic systems is addressed in order to present the theoretical basis of system identification and parameter estimation in a manner that is complete and rigorous, yet understandable with minimal prerequisites. Maximum likelihood and related estimators are highlighted. The approach used requires familiarity with calculus, linear algebra, and probability, but does not require knowledge of stochastic processes or functional analysis. The treatment emphasizes unification of the various areas in estimation in dynamic systems is treated as a direct outgrowth of the static system theory. Topics covered include basic concepts and definitions; numerical optimization methods; probability; statistical estimators; estimation in static systems; stochastic processes; state estimation in dynamic systems; output error, filter error, and equation error methods of parameter estimation in dynamic systems, and the accuracy of the estimates.

Maine, R. E.↗
NASA NTRS1985

The use of Landsat-4 MSS digital data in temporal data sets and the evaluation of scene-to-scene registration accuracy

The MSS sensor on Landsat 4 is, in certain performance aspects, diferent from those of Landsats 1 through 3. These differences created some concern in the NASA research community as to whether individual data sets can be registered accurately enough to produce acceptable data sets for multitemporal data analysis. The use of Landsat 4 MSS digital data in temporal data sets is examined and a method is presented for estimating temporal registration accuracy based on the use of an X-Y digitizer and grey tone electrostatic plots. Results indicate that the RMS temporal registration errors are not significantly different from the temporal data sets generated using Landsat 4 and Landsat 2 data (33.35 meters) and the temporal data set constructed from two Landsat 2 data sets (33.61 meters). A derivation of the model used to evaluate the temporal registration is included.

Anderson, J. E.↗
NASA NTRS1988

Radiometric calibration analysis of SIR-B imagery

The second flight of the NASA Spaceborne Imaging Radar (SIR-B) collected nearly 8 hours of digital synthetic-aperture radar (SAR) data of the earth's surface. This paper analyzes the performance of the SAR to determine the accuracy of the relative radiometric calibration of these data. Procedures are presented for deriving accuracy estimates as well as limitations under which these estimates are valid. The impact on calibration of an RF breakdown in the primary antenna feed system cable is evaluated. An analysis of the platform stability based on the SAR echo data is also presented in conjunction with its resultant effect on the calibration accuracy. Finally, numerical error bounds are derived with guidelines on their utilization.

Wall, Stephen D.↗