Search NASA⌕ Search

SEARCH · Search NASA

Results for “Model interpretation”

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 523 records · Page 29

Dynamics of particle network in composite battery cathodes

We report that improving composite battery electrodes requires a delicate control of active materials and electrode formulation. The electrochemically active particles fulfill their role as energy exchange reservoirs through interacting with the surrounding conductive network. We formulate a network evolution model to interpret the regulation and equilibration between electrochemical activity and mechanical damage of these particles. Through statistical analysis of thousands of particles using x-ray phase contrast holotomography in a LiNi 0.8 Mn 0.1 Co 0.1 O 2 -based cathode, we found that the local network heterogeneity results in asynchronous activities in the early cycles, and subsequently the particle assemblies move toward a synchronous behavior. Our study pinpoints the chemomechanical behavior of individual particles and enables better designs of the conductive network to optimize the utility of all the particles during operation.

25 ENERGY STORAGE↗

Effect of Te on the self-activated emission of ZnSe

An interpretation of the optical properties of ZnSe is suggested. The interpretation is based on the anticrossing band theory, which specifies the oxygen-induced splitting of the conduction band. It is shown that the new approach provides a means for understanding the specific features of the self-activated luminescence in ZnSe:O and (ZnSe:O):Te. The effect of small (<1%) Te additives on the luminescence of ZnSe:Te in the edge spectral region is interpreted. The model of electronic transitions is presented.

36 MATERIALS SCIENCE↗

Digital Twin for Chemical Science (DTCS) v0.01

Directly visualizing the trajectories of chemistry can unravel novel insights into the behavior of catalysts, gas phase reactions, photo-induced dynamics, and building blocks for quantum information processing. The ability of explicitly identifying, tracking, and tagging the exchange of matter, hence the annihilation and creation of new chemical species, can be best realized through a close coupling of theory and experiment. While the synchrotron-based characterization facilities propelled rapidly in its hardware, providing higher brightness, better resolution, and more precision, the software infrastructure is lagging. We developed DTCS (Digital Twin for Chemical Science) v.01, a central platform that faithfully mimics advanced instrumentations in Scientific User Facilities, by solving a variety of technical challenges in data acquisition, analysis, and model-driven interpretation. Rooted in physics and accelerated by AI, we validated this concept by direct comparison with precise experimental X-ray Photoelectron Spectroscopy (XPS) observations using a ubiquitous metal-water interfacial scenario, i.e., Ag/H2O as our main narrative. The DTCS v.01 input mirrors how the bench chemists work, with the output directly linked to the end station computer, thereby providing a user-friendly, knowledge-driven, and accessible user experience with mechanistic insights standardized in a way that are ready to be published, versioned, and transferred flexibly.

Qian, Jin↗

Jupyter Notebook Code for “Data-Driven Insights to Accelerate Advanced Biomanufacturing”

This page contains the datasets and code #O5097 Jupyter Notebook Code for “Data-Driven Insights to Accelerate Advanced Biomanufacturing”. Data literature-derived cultivation experiments for polyhydroxybutyrate (PHB) production in Synechocystis sp. PCC 6803 and were used for ML model development, interpretation, and experimental validation.

Lalonde, Jessica N. [Los Alamos National Laborator↗

Development of a Full-Scale Connected U-Net for Reflectivity Inpainting in Spaceborne Radar Blind Zones

CloudSat’s Cloud Profiling Radar is a valuable tool for remotely monitoring high-latitude snowfall, but its ability to observe hydrometeor activity near the Earth’s surface is limited by a radar blind zone caused by ground clutter contamination. This study presents the development of a deeply supervised U-Net-style convolutional neural network to predict cold season reflectivity profiles within the blind zone at two Arctic locations. The network learns to predict the presence and intensity of near-surface hydrometeors by coupling latent features encoded in blind zone-aloft clouds with additional context from collocated atmospheric state variables (i.e., temperature, specific humidity, and wind speed). Results show that the U-Net predictions outperform traditional linear extrapolation methods, with low mean absolute error, a 38% higher Sørensen–Dice coefficient, and vertical reflectivity distributions 60% closer to observed values. The U-Net is also able to detect the presence of near-surface cloud with a critical success index (CSI) of 72% and cases of shallow cumuliform snowfall and virga with 18% higher CSI values compared to linear methods. An explainability analysis shows that reflectivity information throughout the scene, especially at cloud edges and at the 1.2-km blind zone threshold, along with atmospheric state variables near the tropopause, are the most significant contributors to model skill. This surface-trained generative inpainting technique has the potential to enhance current and future remote sensing precipitation missions by providing a better understanding of the nonlinear relationship between blind zone reflectivity values and the surrounding atmospheric state.

54 ENVIRONMENTAL SCIENCES↗

Machine Learned Empirical Numerical Integrator from Simulated Data

Recently, a number of state-of-the-art surrogate machine learning (ML) models have been designed for global weather and climate prediction, which have been trained using reanalysis data products. Reanalysis data products are constructed using numerical model simulations that combine numerical integration of partial differential equations and parameterization schemes. These products are typically only archived and made available using coarsened spatial and temporal resolutions. This study explores the impact of the numerical generation methods used to produce the training datasets and the temporal resolution of those datasets on machine learning surrogate models. Using the nonlinear vector autoregression (NVAR) machine as an explainable ML technique, simple dynamical systems are emulated with ML models trained on data produced by three classical numerical integration schemes. NVAR is validated as a skillful ML method, capable of producing accurate predictions and, more importantly, reconstructing both the underlying dynamics and the numerical integration scheme used to generate the training data. However, the machine fails to generalize predictions on unseen test data generated by different numerical integration schemes, despite the underlying dynamical system being the same. This result provides a word of caution for the growing field of machine learning emulation of weather and climate dynamics. Furthermore, we illustrate using NVAR that training on temporally coarsened data may increase the required complexity of ML models and potentially introduce new numerical challenges. Finally, we discover that empirical integration schemes with arbitrary time-stepping sizes can be constructed directly from the data, which implies a potential for the development of empirical numerical integration schemes.

54 ENVIRONMENTAL SCIENCES↗

Elastic plastic self-consistent (EPSC) modeling of plastic deformation in fayalite olivine

Elastic plastic self-consistent (EPSC) simulations are used to model synchrotron X-ray diffraction observations from deformation experiments on fayalite olivine using the deformation DIA apparatus. Consistent with results from other in situ diffraction studies of monomineralic polycrystals, the results show substantial variations in stress levels among grain populations. Rather than averaging the lattice reflection stresses or choosing a single reflection to determine the macroscopic stress supported by the specimen, an EPSC simulation is used to forward model diffraction data and determine a macroscopic stress that is consistent with lattice strains of all measured diffraction lines. The EPSC simulation presented here includes kink band formation among the plastic deformation mechanisms in the simulation. The inclusion of kink band formation is critical to the success of the models. This study demonstrates the importance of kink band formation as an accommodation mechanism during plastic deformation of olivine as well as the utility of using EPSC models to interpret diffraction from in situ deformation experiments.

36 MATERIALS SCIENCE↗

Stress testing on silicon carbide electronic devices for prognostics and health management.

Power conversion systems for energy storage and other distributed energy resource applications are among the drivers of the important role that power electronics plays in providing reliable electricity. Wide band gap semiconductors such as silicon carbide (SiC) and gallium nitride (GaN) will help increase the performance and efficiency of power electronic equipment while condition monitoring (CM) and prognostics and health management (PHM) will increase the operational availability of the equipment and thereby make it more cost effective. Voltage and/or temperature stress testing were performed on a number of SiC devices in order to accelerate failure modes and to identify measureable shifts in electrical characteristics which may provide early indication of those failures. Those shifts can be interpreted and modeled to provide prognostic signatures for use in CM and/or PHM. Such experiments will also lead to a deeper understanding of basic device physics and the degradation mechanisms behind failure.

42 ENGINEERING↗

The Structure of Jets at Hadron Colliders

Particle physics seeks to understand the interactions and properties of the fundamental particles. To gain understanding, there is an interplay between theory and experiment. Models are proposed to explain how particles behave and interact. These models make precise predictions that can be tested. Experiments are built and executed to measure the properties of these particles, providing necessary tests for the theories that attempt to explain the realm of fundamental particles. However, there is also another level of interaction between theory and experiment; the development of new experiments demands the study of how particles will behave with respect to the measured observables toward the goal of understanding the details and idiosyncrasies of the measurements very well. Only once these are well-modeled and understood can one be con dent that the data that are measured is trustworthy. The modeling and interpretation of the physics of a proton collider, such as the LHC, is the main topic of this thesis.

43 PARTICLE ACCELERATORS↗

AmeriFlux Measurement Network: Science Team Research

Research involves analysis and field direction of AmeriFlux operations, and the PI provides scientific leadership of the AmeriFlux network. Activities include the coordination and quality assurance of measurements across AmeriFlux network sites, synthesis of results across the network, organizing and supporting the annual Science Team Meeting, and communicating AmeriFlux results to the scientific community and other users. Objectives of measurement research include (i) coordination of flux and biometric measurement protocols (ii) timely data delivery to the Carbon Dioxide Information and Analysis Center (CDIAC); and (iii) assurance of data quality of flux and ecosystem measurements contributed by AmeriFlux sites. Objectives of integration and synthesis activities include (i) integration of site data into network-wide synthesis products; and (ii) participation in the analysis, modeling and interpretation of network data products. Communications objectives include (i) organizing an annual meeting of AmeriFlux investigators for reporting annual flux measurements and exchanging scientific information on ecosystem carbon budgets; (ii) developing focused topics for analysis and publication; and (iii) developing data reporting protocols in support of AmeriFlux network goals.

54 ENVIRONMENTAL SCIENCES↗

Mechanisms of sputter ripple formation: coupling among energetic ions, surface kinetics, stress and composition (Final Report)

Self-organized pattern formation enables the creation of nanoscale surface structures over large areas based on fundamental physical processes rather than an applied template. Low energy ion bombardment is one such method that induces the spontaneous formation of a wide variety of interesting morphological features (e.g., sputter ripples and/or quantum dots). This program focused on the processes controlling sputter ripple formation and the kinetics controlling the evolution of surfaces and nanostructures in high flux environments. This was done by using systematic, quantitative experiments to measure ripple formation under a variety of processing conditions coupled with modeling to interpret the results.

36 MATERIALS SCIENCE↗

Strategies for measuring fuel rho-R and mix using gamma-rays

In this paper we will describe several γ-ray based approaches for determining quantities relevant to achieving ignition at NIF: ρRfuel and hot spot-ablator mix. These approaches use the Gamma Reaction History (GRH) diagnostic and another potential quickly designed γ-ray detector. In addition, a preliminary partial evaluation of the quality of the existing γ-ray production nuclear databases used in modeling and interpreting the data for the documented approaches.

42 ENGINEERING↗

Crump Geyser Exploration and Drilling Project. High Precision Geophysics and Detailed Structural Exploration and Slim Well Drilling

The Crump Geyser Exploration and Drilling Project – High Precision Geophysics and Detailed Structural Exploration and Slim Well Drilling ran from January 29, 2010 to September 30, 2013. During Phase 1 of the project, collection of all geophysical surveys was completed as outlined in the Statement of Project Objectives. In addition, a 5000-foot full sized exploration well was drilled by Ormat, and preexisting drilling data was discovered for multiple temperature gradient wells within the project area. Three dimensional modeling and interpretation of results from the geophysical surveys and drilling data gave confidence to move to the project into Phase 2 drilling. Geological and geophysical survey interpretations combined with existing downhole temperature data provided an ideal target for the first slim-hole drilled as the first task in Phase 2. Slim-hole 35-34 was drilled in September 2011 and tested temperature, lithology, and permeability along the primary range-bounding fault zone near its intersection with buried northwest-trending faults that have been identified using geophysical methods. Following analysis of the results of the first slim-hole 35-34, the second slim hole was not drilled and subsequent project tasks, including flowing differential self-potential (FDSP) surveys that were designed to detail the affect of production and injection on water flow in the shallow aquifer, were not completed. NGP sold the Crump project to Ormat in August 2014, afterwards, there was insufficient time and interest from Ormat available to complete the project objectives. NGP was unable to continue managing the award for a project they did not own due to liability issues and Novation of the award was not a viable option due to federal award timelines. NGP submitted a request to mutually terminate the award on February 18, 2015. The results of all of the technical surveys and drilling are included in this report. Fault interpretations from surface geology, aeromag, seismic, and gravity data sets are in good agreement, illustrating two or more major range-bounding faults and buried northwest trending faults. The intersections of these fault systems provide the primary targets for drilling.

15 GEOTHERMAL ENERGY↗

Analysis of Cavern and Well Stability at the West Hackberry SPR Site Using a Full-Dome Model

This report presents computational analyses that simulate the structural response of caverns at the Strategic Petroleum Reserve (SPR) West Hackberry site. The cavern field comprises 22 caverns. Five caverns (6, 7, 8, 9, 11) were acquired from industry and have unusual shapes and a history dating back to 1946. The other 17 caverns (101-117) were leached according to SPR standards in the mid-1980s and have tall cylindrical shapes. The history of the caverns and their shapes are simulated in a three-dimensional geomechanics model of the site that predicts deformations, strains, and stresses. Future leaching scenarios corresponding to oil drawdowns using fresh water are also simulated by increasing the volume of the caverns. Cavern pressures are varied in the model to capture operational practices in the field. The results of the finite element model are interpreted to provide information on the current and future status of subsidence, well integrity, and cavern stability. The most significant results in this report are relevant to Cavern 6. The cavern is shaped like a bowl with a large ceiling span and is in close proximity to Cavern 9. The analyses predict tensile stresses at the edge of the ceiling during repressurization of Cavern 6 following workover conditions. During a workover the cavern is at low pressure to service a well. The wellhead pressures are atmospheric. When the workover is complete, the cavern is repressurized. The resulting elastic stresses are sufficient to cause tension around the edge of the large ceiling span. With time, these stresses relax to a compressive state because of salt creep. However, the potential for salt fracture and propagation exists, particularly towards Cavern 9. With only 200 feet of salt between the caverns, the operational consequences must be examined if the two caverns become connected. A critical time may be during a workover of Cavern 9 in part because of the operational vulnerabilities, but also because dilatant damage is predicted under the ledge that forms the lower lobe in the cavern. The remaining caverns have no significant issues regarding cavern stability and may be safely enlarged during subsequent oil drawdowns. Predicted well strains and subsidence are significant and consequently future remedial actions may be necessary. These predicted well strains certainly suggest appropriate monitoring through a well-logging program. Subsidence is currently being monitored.

02 PETROLEUM↗

Alternative High-Performance Ceramic Waste Forms

This final report (M5NU-12-NY-AU # 0202-0410) summarizes the results of the project titled “Alternative High-Performance Ceramic Waste Forms,” funded in FY12 by the Nuclear Energy University Program (NEUP Project # 12-3809) being led by Alfred University in collaboration with Savannah River National Laboratory (SRNL). The overall focus of the project is to advance fundamental understanding of crystalline ceramic waste forms and to demonstrate their viability as alternative waste forms to borosilicate glasses. We processed single- and multiphase hollandite waste forms based on simulated waste streams compositions provided by SRNL based on the advanced fuel cycle initiative (AFCI) aqueous separation process developed in the Fuel Cycle Research and Development (FCR&D). For multiphase simulated waste forms, oxide and carbonate precursors were mixed together via ball milling with deionized water using zirconia media in a polyethylene jar for 2 h. The slurry was dried overnight and then separated from the media. The blended powders were then subjected to melting or spark plasma sintering (SPS) processes. Microstructural evolution and phase assemblages of these samples were studied using x-ray diffraction (XRD), scanning electron microscopy (SEM), energy dispersion analysis of x-rays (EDAX), wavelength dispersive spectrometry (WDS), transmission electron spectroscopy (TEM), selective area x-ray diffraction (SAXD), and electron backscatter diffraction (EBSD). These results showed that the processing methods have significant effect on the microstructure and thus the performance of these waste forms. The Ce substitution into zirconolite and pyrochlore materials was investigated using a combination of experimental (in situ XRD and x-ray absorption near edge structure (XANES)) and modeling techniques to study these single phases independently. In zirconolite materials, a transition from the 2M to the 4M polymorph was observed with increasing Ce content. The resulting powders were consolidated via SPS. Ce was reduced to the trivalent oxidation state and the zirconolite was converted into undesirable perovskite. The zirconolite polymorphs found in the synthesized powders were recovered after a post-SPS heat treatment in air. These results demonstrated the potential of processing in controlling the phase assemblage in these waste forms. Hollandites with Cr 3+ trivalent cations were identified as potential hosts for Cs immobilization and are being investigated for Cs retention properties. Series of compositions Ba 1.15-x Cs 2x Cr 2.3 Ti 5.7 O 16 , with increasing Cs loadings, were prepared by sol-gel process and characterized for structural parameters. Structural characterization was performed by a combination of powder XRD and neutron powder diffraction. Phase pure hollandite adapting monoclinic symmetry (I2/m) was observed for 0 ≤ x ≤ 0.55. These results were used to develop a new structural model to interpret Cs immobilization in these hollandites. Performance of these waste forms were evaluated for chemical durability and radiation resistance. Product consistency testing (PCT) and vapor hydration testing (VHT) were used for testing of chemical durability. Radiation resistance was tested using He + ions to simulate $\alpha$ particles and heavy ions such as Au 3+ to simulate a recoil. These results showed that these waste forms were chemically durable. The waste forms also amorphized to various degrees on exposure to simulated radiation.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Analysis of Cavern and Well Stability at the Bryan Mound SPR Site Using the M–D Salt Creep Model

This report presents computational analyses that simulate the structural response of caverns at the Strategic Petroleum Reserve Bryan Mound site. The cavern field comprises 20 caverns. Five caverns (1, 2, 4, and 5; 3 was later plugged and abandoned) were acquired from industry and have unusual shapes and a history dating back to 1946. The other 16 caverns (101-116) were leached according to SPR standards in the mid-1980s and have tall cylindrical shapes. The history of the caverns and their shapes are simulated in a 3-D geomechanics model of the site that predicts deformations, strains, and stresses. Historical wellhead pressures are used to calculate cavern pressures up through July 2016. Because of the extent of heterogeneous creep behavior observed throughout the Bryan Mound site, a set of cavern-specific creep coefficients was developed to produce better matches with measured cavern closure and surface subsidence. For this new implementation of the model, there are two significant advances: the use of the multimechanism deformation (M-D) salt creep model to evaluate both steady-state and transient salt creep; and the creation of finite element mesh geometries for the caverns that nearly exactly match the geometries obtained through sonar measurements. The results of the finite element model are interpreted to provide information on the current and future status of subsidence, well integrity, cavern stability, and drawdown availability.

58 GEOSCIENCES↗

Feature Selection and Inferential Procedures for Video Data [Slides]

With the rise of electronic and high-dimensional data, new and innovative feature detection and statistical methods are required to perform accurate and meaningful statistical analysis of these datasets that provide unique statistical challenges. In the area of feature detection, much of the recent feature detection research in the computer vision community has focused on deep learning methods, which require large amounts of labeled training data. However, in many application areas, training data is very limited and often difficult to obtain. We develop methods for fast, unsupervised, precise feature detection for video data based on optical flows, edge detection, and clustering methods. We also use pretrained neural networks and interpretable linear models to extract features using very limited training data. In the area of statistics, while high-dimensional data analysis has been a main focus of recent statistical methodological research, much focus has been on populations of high-dimensional vectors, rather than populations of high-dimensional tensors, which are three-dimensional arrays that can be used to model dependent images, such as images taken of the same person or ripped video frames. Our feature detection method is a non-model-based method that fusses information from dense optical flow, raw image pixels, and frame differences to generate detections. Our hypothesis testing methods are based on the assumption that dependent images are concatenated into a tensor that follows a tensor normal distribution, and from this assumption, we derive likelihood-ratio, score, and regression-based tests for one- and multiple-sample testing problems. Our methods will be illustrated on simulated and real datasets. We conclude this report with comments on the relationship between feature detection and hypothesis testing methods.

97 MATHEMATICS AND COMPUTING↗

From Stars to Compact Objects: The Initial-Final Mass Relation

One of the key phases of stellar evolution that remains poorly understood is stellar death. We lack a predictive model for how a star of a given mass explodes and what kind of remnant it leaves behind (i.e. the initial-final mass relation, IFMR). Progress has been limited due to the difficulty in finding and weighing black holes and neutron stars in large numbers. Technological advances that allow for sub-milliarcsecond astrometry in crowded fields have opened a new window for finding black holes and neutron stars: astrometric gravitational lensing. Finding and weighing a sample of compact objects with astrometric microlensing will allow us to place some of the first constraints on the present-day mass function of isolated black holes and neutron stars, their multiplicity, and their kick velocities. All of these are fundamental inputs into understanding the death phase of stellar evolution, improving supernovae models, and interpreting LIGO detections in an astrophysical context. To achieve these goals, we require large area surveys, such as the WFIRST exoplanet microlensing survey, to photometrically identify long-duration (>120 day), un-blended microlensing events as candidate compact objects. We also require high-precision astrometric follow-up monitoring using extremely large telescopes, equipped with adaptive optics, such as TMT and GMT.

79 ASTRONOMY AND ASTROPHYSICS↗