Results for “limited data” — Page 6

DOE OSTI2022

An Application of a Modified Beta Factor Method for the Analysis of Software Common Cause Failures

This paper presents an approach for modeling software common cause failures (CCFs) within digital instrumentation and control (I&C) systems. CCFs consist of a concurrent failure between two or more components due to a shared failure cause and coupling mechanism. This work emphasizes the importance of identifying software-centric attributes related to the coupling mechanisms necessary for simultaneous failures of redundant software components. The groups of components that share coupling mechanisms are called common cause component groups (CCCGs). Most CCF models rely on operational data as the basis for establishing CCCG parameters and predicting CCFs. This work is motivated by two primary concerns: (1) a lack of operational and CCF data for estimating software CCF model parameters; and (2) the need to model single components as part of multiple CCCGs simultaneously. A hybrid approach was developed to account for these concerns by leveraging existing techniques: a modified beta factor model allows single components to be placed within multiple CCCGs, while a second technique provides software-specific model parameters for each CCCG. This hybrid approach provides a means to overcome the limitations of conventional methods while offering support for design decisions under the limited data scenario.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2024

AI-Driven Crack Detection for Remanufacturing Cylinder Heads Using Deep Learning and Engineering-Informed Data Augmentation

Detecting cracks in cylinder heads traditionally relies on manual inspection, which is time-consuming and susceptible to human error. As an alternative, automated object detection utilizing computer vision and machine learning models has been explored. However, these methods often face challenges due to a lack of sufficiently annotated training data, limited image diversity, and the inherently small size of cracks. Addressing these constraints, this paper introduces a novel automated crack-detection method that enhances data availability through a synthetic data generation technique. Unlike general data augmentation practices, our method involves copying cracks from one location to another, guided by both random and informed engineering decisions about likely crack formations due to cyclic thermomechanical loads. The innovative aspect of our approach lies in the integration of domain-specific engineering knowledge into the synthetic generation process, which substantially improves detection accuracy. We evaluate our method’s effectiveness using two metrics: the F2 score, which emphasizes recall to prioritize detecting all potential cracks, and mean average precision (MAP), a standard measure in object detection. Experimental results demonstrate that, without engineering insights, our method increases the F2 score from 0.40 to 0.65, while maintaining a stable MAP. Incorporating detailed engineering knowledge further enhances the F2 score to 0.70 and improves MAP to 0.57, representing increases of 63% and 43%, respectively. These results confirm that our approach not only mitigates the limitations of traditional data augmentation but also significantly advances the reliability and precision of crack detection in industrial settings.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2024

Bi-fidelity variational auto-encoder for uncertainty quantification

Quantifying the uncertainty of quantities of interest (QoIs) from physical systems is a primary objective in model validation. However, achieving this goal entails balancing the need for computational efficiency with the requirement for numerical accuracy. To address this trade-off, we propose a novel bi-fidelity formulation of variational auto-encoders (BF-VAE) designed to estimate the uncertainty associated with a QoI from low-fidelity (LF) and high-fidelity (HF) samples of the QoI. Here, this model allows for the approximation of the statistics of the HF QoI by leveraging information derived from its LF counterpart. Specifically, we design a bi-fidelity auto-regressive model in the latent space which is integrated within the VAE’s probabilistic encoder–decoder structure. An effective algorithm is proposed to maximize the variational lower bound of the HF log-likelihood in the presence of limited HF data, resulting in the synthesis of HF realizations with a reduced computational cost. Additionally, we introduce the concept of the bi-fidelity information bottleneck (BF-IB) to provide an information-theoretic interpretation of the proposed BF-VAE model. Our numerical results demonstrate that the BF-VAE leads to considerably improved accuracy, as compared to a VAE trained using only HF data, when limited HF data is available.

42 ENGINEERING↗

DOE OSTI2022

Multitask Machine Learning of Collective Variables for Enhanced Sampling of Rare Events

Computing accurate reaction rates is a central challenge in computational chemistry and biology because of the high cost of free energy estimation with unbiased molecular dynamics. In this work, a data-driven machine learning algorithm is devised to learn collective variables with a multitask neural network, where a common upstream part reduces the high dimensionality of atomic configurations to a low dimensional latent space and separate downstream parts map the latent space to predictions of basin class labels and potential energies. Here, the resulting latent space is shown to be an effective low-dimensional representation, capturing the reaction progress and guiding effective umbrella sampling to obtain accurate free energy landscapes. This approach is successfully applied to model systems including a 5D Müller Brown model, a 5D three-well model, the alanine dipeptide in vacuum, and an Au(110) surface reconstruction unit reaction. It enables automated dimensionality reduction for energy controlled reactions in complex systems, offers a unified and data-efficient framework that can be trained with limited data, and outperforms single-task learning approaches, including autoencoders.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2022

State Estimation for Distribution Networks with Asynchronous Sensors Using Stochastic Descent: Preprint

This paper investigates the problem of state estimation for distribution networks with asynchronous sensors comprising of a mix of smart meters and phasor measurement units (PMUs) with multiple sampling and reporting rates. We consider two independent scenarios of state estimation and tracking, with either voltages or currents as states. With these two sets, we investigate estimation under (a) full data, assuming all measurements are available and (b) limited data, where an online algorithmic approach is adopted to estimate the possibly time-varying states by processing measurements as and when available. The proposed algorithm, inspired by the classical Stochastic Gradient Descent (SGD) approach updates the states based on the previous estimate and the newly available measurements. Finally, we demonstrate the estimation and tracking efficacy through numerical simulations on the IEEE-37 test network, while also highlighting how estimation with currents as states leads to faster convergence.

asynchronous sensors↗

NASA NTRS1981

Evaluation of high temperature structural adhesives for extended service

The long term thermal aging data initiated in Phase 1 is reported. All candidate adhesive systems have exhibited significant degradation in bond properties after 505K (450 F) 10,000 hour exposure. Failures appear to be adhesive in the oxide layer. Phase 2 chemical characterization, cure cycle studies, baseline data, preliminary specifications, and environmental exposure data generated on polyphenyquinoxaline is presented. Similar but limited data on LARC-13 and NR056X adhesives is reported.

Hill, S. G.↗

NASA NTRS1993

Proposed ground-based control of accelerometer on Space Station Freedom

This paper describes the innovative control of an accelerometer to support the needs of the scientists operating science experiments that are on-board Space Station Freedom (SSF). Accelerometers in support of science experiments on the shuttle have typically been passive, record-only devices that present data only after the mission or that present limited data to the crew or ground operators during the mission. With the advent of science experiment operations on SSF, the principal investigators will need microgravity acceleration data during, as well as after, experiment operations. Because their data requirements may change during the experiment operations, the principal investigators will be allocated some control of accelerometer parameters. This paper summarizes the general-purpose Space Acceleration Measurement System (SAMS) operation that supports experiments on the shuttle and describes the control of the SAMS for Space Station Freedom. Emphasis is placed on the proposed ground-based control of the accelerometer by the principal investigators.

Delombard, Richard↗

NASA NTRS1993

Proposed ground-based control of accelerometer on Space Station Freedom

This paper describes the innovative control of an accelerometer to support the needs of the scientists operating science experiments that are on-board Space Station Freedom (SSF). Accelerometers in support of science experiments on the shuttle have typically been passive, record-only devices that present data only after the mission or that present limited data to the crew or ground operators during the mission. With the advent of science experiment operations on SSF, the principal investigators will need microgravity acceleration data during, as well as after, experiment operations. Because their data requirements may change during the experiment operations, the principal investigators will be allocated some control of accelerometer parameters. This paper summarizes the general-purpose Space Acceleration Measurement System (SAMS) operation that supports experiments on the shuttle and describes the control of the SAMS for Space Station Freedom. Emphasis is placed on the proposed ground-based control of the accelerometer by the principal investigators.

Delombard, Richard↗

NASA NTRS2006

NASA's Program to Monitor Orbital Debris in the GEO Belt and the General Problem of Measuring Near-Earth Object Orbits: Similarities and Differences

One of the goals for NASA s Orbital Debris Program Office has been to accurately characterize the population of debris in the geosynchronous Earth orbit (GEO) environment. Most objects larger than about 1 meter in size are regularly tracked and catalogued by the US Space Surveillance System in the GEO regime. The consequence has been that most large intact GEO objects are tracked, but the vast majority of GEO debris fragments are not. Only in recent years have observations been dedicated to characterize the GEO debris population. NASA s efforts have concentrated on using wide field-of-view telescopes to make complete surveys of the GEO regime to better our statistical understanding of the GEO debris population. These telescopes operate in a staring mode, and only make limited short-arc measurements of the orbits. This information, while limited, allows the possibility of debiasing the observations and constructing statistical distributions of orbits in inclination and ascending node. Recent work suggests that we may be able to use statistical methods to estimate better orbit parameters despite the limited data. Both of these types of studies estimating statistical orbit distributions, and estimating accurate orbits using limited short-arc data have direct analogues in ongoing studies of near-Earth objects (NEO) such as asteroids and comets. This talk will describe the GEO study methods in use and being developed at NASA, and will discuss how such methods may or may not be applicable for NEO studies as well.

Matney, Mark↗

NASA NTRSDate not supplied

Evaluation of Composite Airframe Dynamic Impact Modeling Using Hawker 4000 Fuselage Drop Test Data

Two drop tests of partial Hawker 4000 fuselage sections were conducted at the National Aeronautics and Space Administration (NASA) Langley Research Center (LaRC) to characterize the response of representative composite aerospace structure to dynamic impact loads. Test conditions were selected to induce damage into the composite structure in order to study material failure within a composite fuselage and evaluate the capability of finite element (FE) model analysis to predict that failure. The tests were simulated using FE models which were generated to isolate the effect of developmental data availability on predictive capability. FE Models of the tested fuselage sections were generated using two limited data sets. The first model configuration was reverse engineered from the test article with no information related to design or fabrication details which would be known only by the manufacturer. The second model was generated from data provided by the manufacturer but without additional material characterization test data. Models were developed using these methodologies for both fuselage sections tested. Correlation of each model to the tests conducted was evaluated in terms of damage, deformation, and cabin acceleration measurements. Correlation between the developed models and the tested fuselage sections showed that the reverse engineered model predicted the composite damage and cabin acceleration measured during test though it was limited due to lack of detail in the composite layup changes through the structure. The model developed using manufacturer specifications did not predict damage, due to limited material and component model characterization data, but it did predict acceleration on par with the reverse engineered model. Model capability and limitation sources identified were verified through correlation of a final model which was developed by combining the individual data sets. The combined model demonstrated that the addition of calibrated composite material models to accurate composite layup definitions and detailed geometry led to improved correlation of damage and acceleration response within the composite fuselage structures.

Crashworthiness↗

NASA NTRSDate not supplied

Evaluation of Composite Airframe Dynamic Impact Modeling Using Hawker 4000 Fuselage Drop Test Data

Two drop tests of partial Hawker 4000 fuselage sections were conducted at the National Aeronautics and Space Administration (NASA) Langley Research Center (LaRC) to characterize the response of representative composite aerospace structure to dynamic impact loads. Test conditions were selected to induce damage into the composite structure in order to study material failure within a composite fuselage and evaluate the capability of finite element (FE) model analysis to predict that failure. The tests were simulated using FE models which were generated to isolate the effect of developmental data availability on predictive capability. FE Models of the tested fuselage sections were generated using two limited data sets. The first model configuration was reverse engineered from the test article with no information related to design or fabrication details which would be known only by the manufacturer. The second model was generated from data provided by the manufacturer but without additional material characterization test data. Models were developed using these methodologies for both fuselage sections tested. Correlation of each model to the tests conducted was evaluated in terms of damage, deformation, and cabin acceleration measurements. Correlation between the developed models and the tested fuselage sections showed that the reverse engineered model predicted the composite damage and cabin acceleration measured during test though it was limited due to lack of detail in the composite layup changes through the structure. The model developed using manufacturer specifications did not predict damage, due to limited material and component model characterization data, but it did predict acceleration on par with the reverse engineered model. Model capability and limitation sources identified were verified through correlation of a final model which was developed by combining the individual data sets. The combined model demonstrated that the addition of calibrated composite material models to accurate composite layup definitions and detailed geometry led to improved correlation of damage and acceleration response within the composite fuselage structures.

Crashworthiness↗

DOE OSTI2025

Deriving Stable Peak Models to Fit Complex XPS Data From Cu Contaminated Pt Electrocatalysts

X-ray Photoelectron Spectroscopy spectra peak models, designed to partition photoemission signals emanating from different elements or chemical states within an atom, are fitted to data limited to an energy interval over which inelastically scattered photoemission signal can be estimated. While the choice of background approximation and line shapes of components to the peak model requires careful consideration, the energy interval used to define the data to which the peak model is optimized has a significant impact on the final peak model. The relationship between the background intensity and data intensity at the start and end of the energy interval dictates the line shapes used in the peak model. In this work, we devise a method to peak fit a complex overlapping Cu 3p and Pt 4f XPS peak structure to perform the elemental quantification. We first use an Al 2s peak to illustrate how background curves approach data at the limits of the energy interval over which the background is defined, influencing the analysis of XPS spectra. Next, we demonstrate the nature of interactions between specific line shapes (Voigt and pseudo-Voigt profiles) suitable for photoemission peaks and a specific background curve (Shirley) and a peak model is presented that includes components to the peak model that accommodates background intensity during fitting of the peak model to data. The peak model allowed for quantification of the contributions of Pt 4f peaks emanating from the substrate that exhibits strong asymmetry in the presence of the inhomogeneously distributed Cu species, mostly of Lorentzian character.

XPS↗

NASA NTRS2006

Lithium-Ion Verification Test Program

Need for technology verification for aerospace applications. Structure flexible program that will allow assessment of current technology capabilities. Provide information about various vendors. Provide for assessment of technology developments. Developed statistical DOE to interpret relationships in data and to address program test goals and resource limitations. Data will be used to develop a model to predict life of cells as a function of DOD, temperature, and EOCV.

McKissock, Barbara↗

NASA NTRS2008

Reliability and Confidence Interval Analysis of a CMC Turbine Stator Vane

High temperature ceramic matrix composites (CMC) are being explored as viable candidate materials for hot section gas turbine components. These advanced composites can potentially lead to reduced weight, enable higher operating temperatures requiring less cooling and thus leading to increased engine efficiencies. However, these materials are brittle and show degradation with time at high operating temperatures due to creep as well as cyclic mechanical and thermal loads. In addition, these materials are heterogeneous in their make-up and various factors affect their properties in a specific design environment. Most of these advanced composites involve two- and three-dimensional fiber architectures and require a complex multi-step high temperature processing. Since there are uncertainties associated with each of these in addition to the variability in the constituent material properties, the observed behavior of composite materials exhibits scatter. Traditional material failure analyses employing a deterministic approach, where failure is assumed to occur when some allowable stress level or equivalent stress is exceeded, are not adequate for brittle material component design. Such phenomenological failure theories are reasonably successful when applied to ductile materials such as metals. Analysis of failure in structural components is governed by the observed scatter in strength, stiffness and loading conditions. In such situations, statistical design approaches must be used. Accounting for these phenomena requires a change in philosophy on the design engineer s part that leads to a reduced focus on the use of safety factors in favor of reliability analyses. The reliability approach demands that the design engineer must tolerate a finite risk of unacceptable performance. This risk of unacceptable performance is identified as a component's probability of failure (or alternatively, component reliability). The primary concern of the engineer is minimizing this risk in an economical manner. The methods to accurately determine the service life of an engine component with associated variability have become increasingly difficult. This results, in part, from the complex missions which are now routinely considered during the design process. These missions include large variations of multi-axial stresses and temperatures experienced by critical engine parts. There is a need for a convenient design tool that can accommodate various loading conditions induced by engine operating environments, and material data with their associated uncertainties to estimate the minimum predicted life of a structural component. A probabilistic composite micromechanics technique in combination with woven composite micromechanics, structural analysis and Fast Probability Integration (FPI) techniques has been used to evaluate the maximum stress and its probabilistic distribution in a CMC turbine stator vane. Furthermore, input variables causing scatter are identified and ranked based upon their sensitivity magnitude. Since the measured data for the ceramic matrix composite properties is very limited, obtaining a probabilistic distribution with their corresponding parameters is difficult. In case of limited data, confidence bounds are essential to quantify the uncertainty associated with the distribution. Usually 90 and 95% confidence intervals are computed for material properties. Failure properties are then computed with the confidence bounds. Best estimates and the confidence bounds on the best estimate of the cumulative probability function for R-S (strength - stress) are plotted. The methodologies and the results from these analyses will be discussed in the presentation.

Murthy, Pappu L. N.↗

DOE OSTI2025

Graphene SHDMC Data

The data used to produce all of the figures and tables in the manuscript titled "Highly Accurate Many-Body Theory Reaches 2D Materials" can be found here. This data set includes: -SHDMC results for graphene -selected CI with and without re-normalized second-order perturbation (rPT2) theory corrections for graphene -data demonstrating that SHDMC displays an exponential rate of convergence -data used for sCI + rPT2 complete basis set extrapolation -data used to extrapolate SHDMC energies to the infinite basis limit -data used to demonstrate compactness of SHDMC wavefunction

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2023

Natural Language Processing-Enhanced Nuclear Industry Operating Experience Data Analysis: Aggregation and Interpretation of Multi-Report Analysis Results

Industry-wide operating experience is a critical source of raw data for reliability and risk model parameter estimations for nuclear power plants. A large portion of operating experience data are failure events stored as reports that contain unstructured data, such as narratives. In current practice, a failure report is usually reviewed and manually coded by analysts. The coding is based on extracting several event characteristics such as system name, component type, sub-part type, failure mode, and failure cause. Event narratives are mostly used to help understand events and extract their characteristics. In this line of research, we aim to maximize the usage of event narratives by leveraging natural language processing (NLP) methods to automatically convert an event narrative to a causal graph. This research has promise to improve physical understanding of failure initiation and propagation and to facilitate use of non-failure data (e.g., near-misses and degradations) to complement the limited data pool of failures. In our previous work, we developed an NLP tool and applied it to analyze a number of licensee event reports submitted by U.S. nuclear power plants to the Nuclear Regulatory Commission. In this paper, we will report our recent research progress in aggregating the results of multiple reports, developing network model(s), and drawing statistical insights.

99 GENERAL AND MISCELLANEOUS↗

NASA NTRS1984

A First Evaluation of LANDSAT TM Data to Monitor Suspended Sediments in Lakes

A comparison was made between ground data collected from Lake Chicot, Arkansas, and Thematic Mapper (TM) data collected on September 23, 1982. A preliminary analysis of limited data indicate tht Thematic Mapper data may be useful in monitoring suspended sediment and chlorophyll in a lake with high suspended sediment loads. Total suspended loads ranged from 168 to 508 mg/l. TM Band 3 appears to be most useful with Bands 1, 2 and 4 also containing useful information relative to suspended sediments. Considering water data only, Bands 1, 2 and 3 appear to provide similar information. Bands 3 and 4 are also significantly related. Bands 5 and 7 appear to have independent information content relative to the presence or absence of water. Insufficient range of water temperature ground truth data made an evaluation of TM Band 6 difficult.

Schiebe, F. R.↗

NASA NTRS1992

Linear regression in astronomy. II

A wide variety of least-squares linear regression procedures used in observational astronomy, particularly investigations of the cosmic distance scale, are presented and discussed. The classes of linear models considered are (1) unweighted regression lines, with bootstrap and jackknife resampling; (2) regression solutions when measurement error, in one or both variables, dominates the scatter; (3) methods to apply a calibration line to new data; (4) truncated regression models, which apply to flux-limited data sets; and (5) censored regression models, which apply when nondetections are present. For the calibration problem we develop two new procedures: a formula for the intercept offset between two parallel data sets, which propagates slope errors from one regression to the other; and a generalization of the Working-Hotelling confidence bands to nonstandard least-squares lines. They can provide improved error analysis for Faber-Jackson, Tully-Fisher, and similar cosmic distance scale relations.

Feigelson, Eric D.↗