Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “validation dataset”

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 1,027 records · Page 57
All sourcesNASA NTRSDOE OSTI
NASA NTRS2019

Validation of CERES Flight Model 5 In-Orbit Calibrations Using Lunar Observations

Scientific studies require radiation fluxes over the Earth to be accurate within 1% for shortwave fluxes and 0.5% for outgoing longwave fluxes. The validation of in-orbit instrument performance requires both stability in calibration source and also calibration corrections to compensate for instrument changes. The Moon offers an external source whose signal variance is predictable and non-degrading. CERES detectors register the signal output from the entire face of the Moon. Lunar observations performed by CERES Flight Models (FM) 1 through 4 have been successful in assisting validation of radiances to the required accuracy. CERES Flight Model 5 (FM-5) is on Suomi/NPP spacecraft, orbiting since October 2011. This paper uses lunar measurements to validate detector output of FM-5. These measurements are adjusted to remove orbital effects due to variations in distance between Moon and Sun, distance between the satellite and the Moon and lunar phase angle. These effects create a total variation in lunar irradiance of 20% in the total channel and 8% in the shortwave channel. The change in orientation of the Moon as seen by the detector is called libration and causes variations of about 1% of the irradiance. A consistent dataset spanning at least 2 years in length is required to remove variations due to libration. The major uncertainties remaining in the measurements are assumed to be due to changes of the spectral responses of the channels due to degradation of optical surfaces in orbit. The results demonstrate that lunar observations can be used to validate FM-5 measurements.

Clouds and Earth Radiant Energy System↗
DOE OSTI2026

Validation of Hermes-3 turbulence simulations against the TCV-X21 diverted L-mode reference case

Electrostatic flux-driven turbulence simulations with the Hermes-3 code are performed in TCV L-mode conditions in forward and reversed toroidal field configurations, and compared to the TCV-X21 reference dataset (Oliveira et al 2022 Nucl. Fusion 62 096001) qualitatively and with a quantitative methodology. Using only the magnetic equilibrium, total power across the separatrix (120 kW) and total particle flux to the targets (3 x 10 21 s −1 ) as inputs, the simulations produce time-averaged plasma profiles in good agreement with experiment. Shifts in the target peak location when the toroidal field direction is reversed are reproduced in simulation, including the experimentally observed splitting of the outer strike point into two density peaks. The overall normalized discrepancy between simulation and observation is better than any previously reported in the reversed field configuration, and matches the best previously reported in forward field configuration. Differences between simulation and experiment include density profiles inside the separatrix and at the inner target in forward (favorable $\bigtriangledown B$) field configuration. These differences in target temperature in forward field configuration lead to differences in the balance of current to the inner and outer divertor in the private flux region. The cause of these differences is most likely the lack of neutral gas in these simulations, indicating that even in low recycling regimes neutral gas plays an important role in determining edge plasma profiles. These conclusions are consistent with findings in Oliveira et al (2022 https://github.com/SPCData/TCV-X21).

Physics - Plasma physics↗
DOE OSTI2024

Artificial intelligence to unlock real-world evidence in clinical oncology: A primer on recent advances

Purpose: Real world evidence is crucial to understanding the diffusion of new oncologic therapies, monitoring cancer outcomes, and detecting unexpected toxicities. In practice, real world evidence is challenging to collect rapidly and comprehensively, often requiring expensive and time-consuming manual case-finding and annotation of clinical text. In this Review, we summarise recent developments in the use of artificial intelligence to collect and analyze real world evidence in oncology. Methods: We performed a narrative review of the major current trends and recent literature in artificial intelligence applications in oncology. Results: Artificial intelligence (AI) approaches are increasingly used to efficiently phenotype patients and tumors at large scale. These tools also may provide novel biological insights and improve risk prediction through multimodal integration of radiographic, pathological, and genomic datasets. Custom language processing pipelines and large language models hold great promise for clinical prediction and phenotyping. Conclusions: Despite rapid advances, continued progress in computation, generalizability, interpretability, and reliability as well as prospective validation are needed to integrate AI approaches into routine clinical care and real-time monitoring of novel therapies.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2025

Development of an interatomic potential for the Ta–Li system

A new interatomic potential for the Ta–Li system is introduced to facilitate the study of phase stability, mechanical properties and non-equilibrium dynamics after Li implantation in Ta. Here, this potential is based on a generalization of the embedded atom method (GEAM) and includes contributions from embedding energy, explicit two- and three-body interactions, and nonlocal many-body interaction terms. The parameters of the potential are optimized using energies and atomic forces for a wide range of configurations obtained from ab initio density functional theory (DFT) calculations. The potential is rigorously validated across a range of physical properties, including elastic constants, equations of state, phonon dispersion curves, point defect properties, and melting temperatures for different compositions. Although our potential is trained on a small dataset, its accuracy is comparable to that of available machine learning potentials for Li and Ta. Our simulations show that at temperatures below 500 K, Li atoms in Ta–Li alloys form clusters separated by Ta-rich domains, and we find no evidence of ordered phase formation. For Li concentrations below a few percent, Li atoms preferentially segregate to surfaces and grain boundaries. However, in alloys containing more than ~10% Li, the accumulation of Li in symmetric-tilt grain boundaries can lead to one of the following effects: formation of amorphous-like regions, changes in grain boundary structural units, or lateral movement of the grain boundary.

GEAM potential↗
DOE OSTI2022

When physics-informed data analytics outperforms black-box machine learning: A case study in thickness control for additive manufacturing

Aerosol jet printing (AJP) has emerged as a promising noncontact additive manufacturing method for high-resolution printing for a wide range of material systems. A key challenge limiting the broader adoption of AJP in the material science community is the lack of methods to precisely control thickness. Herein, we develop a model-based design of experiment (MBDoE) framework that integrates physics-informed models, nonlinear regression, and information criteria to postulate, select and calibrate the best model to describe and optimize the AJP manufacturing process. Starting with already available data from system commissioning (e.g., prior single variable sensitivity analysis), four candidate physics-informed models are postulated and trained. MBDoE identifies a single additional optimal experiment to validate these predictive models with quantified uncertainties, which are then used to determine the best experimental conditions to control printed film thickness. As a comparative benchmark, the analysis is repeated using the same dataset with nonparametric Gaussian process regression (GPR) model that does not incorporate physical information. Using MBDoE principles, we find that only five experiments are necessary to calibrate the nonlinear physics-informed parametric model, and with said limited data, this model outperforms the black-box machine learning GPR model. This key result underscores an emerging trend in the data science community: incorporating physical information into predictive models often drastically reduces the data requirements. Leveraging MBDoE further increased the data efficiency. By design, the proposed data science framework is general in nature and can be easily extended to other experimental and additive manufacturing systems beyond AJP.

Aerosol jet printing↗
DOE OSTI2025

Next generation Arctic vegetation maps: Aboveground plant biomass and woody dominance mapped at 30 m resolution across the tundra biome

The Arctic is warming faster than anywhere else on Earth, placing tundra ecosystems at the forefront of global climate change. Plant biomass is a fundamental ecosystem attribute that is sensitive to changes in climate, closely tied to ecological function, and crucial for constraining ecosystem carbon dynamics. However, the amount, functional composition, and distribution of plant biomass are only coarsely quantified across the Arctic. Therefore, we developed the first moderate resolution (30 m) maps of live aboveground plant biomass (g m −2 ) and woody plant dominance (%) for the Arctic tundra biome, including the mountainous Oro Arctic. We modeled biomass for the year 2020 using a new synthesis dataset of field biomass harvest measurements, Landsat satellite seasonal synthetic composites, ancillary geospatial data, and machine learning models. Additionally, we quantified pixel-wise uncertainty in biomass predictions using Monte Carlo simulations and validated the models using a robust, spatially blocked and nested cross-validation procedure. Observed plant and woody plant biomass values ranged from 0 to ∼6000 g m −2 (mean ≈ 350 g m −2 ), while predicted values ranged from 0 to ∼4000 g m −2 (mean ≈ 275 g m −2 ), resulting in model validation root-mean-squared-error (RMSE) ≈ 400 g m −2 and R 2 ≈ 0.6. Our maps not only capture large-scale patterns of plant biomass and woody plant dominance across the Arctic that are linked to climatic variation (e.g., thawing degree days), but also illustrate how fine-scale patterns are shaped by local surface hydrology, topography, and past disturbance. By providing data on plant biomass across Arctic tundra ecosystems at the highest resolution to date, our maps can significantly advance research and inform decision-making on topics ranging from Arctic vegetation monitoring and wildlife conservation to carbon accounting and land surface modeling.

Climate change↗
DOE OSTI2023

Noninvasive liquid level sensing with laser generated ultrasonic waves

This article proposes a noninvasive liquid level sensing technique using laser-generated ultrasound waves for nuclear power plant applications. Liquid level sensors play an important role of managing the coolant system safely and stably in the plant structure. Current sensing techniques are mostly intrusive, performing inside the fluidic structure, which is disadvantageous in terms of the regular maintenance of the plant system. Furthermore, typical intrusive sensors do not perform stably under varying environmental conditions such as temperature and radiation. In this study, sensing units are attached to the outer surface of a liquid vessel to capture guided ultrasound waves in a nonintrusive manner. The signal intensity of the guided wave dissipates when the signal interacts with the internal liquid media. The sensing mechanism is mathematically expressed as an index value to correlate the liquid level with the sensor signal. For the acoustic wave generation, laser-generated ultrasound was adopted instead of using typical contact type transducers. Following the simulation validation of the proposed concept, the performance of the developed sensor was confirmed through experimental results under elevated liquid temperature conditions. Finally, the nonlinear multivariable regression exhibited the best-fit to the datasets measured under the variable liquid level and temperature conditions.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2025

MPEX AI Digital Twins

All magnetically confined plasma fusion power plant concepts (Tokamak, Spherical Tokamak, Stellarator, Mirror, ...) must exhaust the heat and plasma from the core confinement region to the material walls. The primary channel for this exhaust is through a plasma divertor which directs plasma along open magnetic field lines to a material target. The Material Plasma Exposure eXperiment (MPEX) illustrated in Figure 1, is a high-power, steady-state linear plasma device designed to produce the plasma material interaction (PMI) conditions of the divertor of future magnetic confinement fusion power plants: energy flux 20MW/m 2 , ion fluence 1031/m 2 , pulse duration 106 sec. These goals of plasma exposure in MPEX are well beyond those achieved in magnetic fusion experimental devices. Successfully achieving these high power steady state conditions for long pulses requires operational control of the heating and particle sources and the plasma flux to the walls and target. The MPEX AI Hot Spot Controller, proposed in this project, will help achieve the operational milestones of MPEX. The MPEX device will begin commissioning at the end of FY26. A smaller proto-MPEX was operated for 14,666 plasma discharges and will resume operation in September of 2025 as proto-MPEX-lite, with reduced capability, to test a new window for the Helicon plasma source. The proto-MPEX data has undergone surrogate modeling with machine learning methods (R. Archibald, 2022 IEEE International Conference on Big Data). This proto-MPEX data will be used to begin development of the AI digital twins described in this white paper. The scientific mission of MPEX is to qualify materials of different composition for use in the high energy and plasma flux conditions of a fusion power plant. The materials exposed in MPEX will in some cases be exposed to high neutron fluxes at other ORNL facilities to measure the changes to their PMI properties. The targets exposed in MPEX will be transported under vacuum to a Surface Analysis Station (SAS). The SAS will be equipped with the following diagnostics: Focused Ion Beam (FIB) for trench milling, 100-400 angstrom resolution scanning electron microscope (SEM), surface mapping x-ray spectrometer, high resolution camera, and a future upgrade to a laser induced breakdown spectroscopy quadruple mass spectrometer (LIBS-QMS). The MPEX experiments will generate diverse pre- and post-exposure measurement data of detailed material properties down to the crystal grain level in 3D for post-exposure assessment of PMI damage (e.g. cracking, melting, erosion and redeposition of the material). Physics models for the PMI, and how the material composition and manufacturing impact its performance under high energy plasma exposure, need to be validated with MPEX data to guide the selection of new candidate materials. Our vision for the MPEX AI Digital Twins project is to supply experimental and physics model simulation data to train Artificial Intelligence (AI) models for data processing, analysis, operational control, PMI and materials simulation to maximize the scientific output of the MPEX device. Ultimately, an AI digital twin of MPEX material assessment metrics for tested and synthetic material types with simulated PMI will be trained by the AI Modeling Teams on the experimental and physics simulation data submitted to the American Science Cloud by this project. A purely empirical search for the best material is inefficient given the finite number of samples that can be tested on MPEX. In order to expand the material properties database for training the MPEX Material Assessment AI Digital Twin, and to gain physics understanding of the PMI processes, physics models of the material properties and PMI processes are required. The physics simulations provide detailed simulation data, like impact angles for plasma ions, sputtering yields, transport of the ionized sputtered target material in the plasma, and redeposition locations. This simulation data expands the measurement data for deeper physics understanding. The experimental data is essential to validate the PMI and material structure simulation models. The validated models can then be used to generate new simulation data of MPEX material assessments for synthetic material compositions that have not been exposed in MPEX. These predictive simulations, plus the whole experimental dataset, will be used to train the MPEX Material Assessment AI Digital Twin allowing a rapid generative AI search for new materials with reduced PMI damage by interpolating the domain of the training set. These new optimum materials can be simulated with the physics codes and/or tested in MPEX. The ability of AI neural networks to interpolate multi-dimensional parameter spaces and generate virtual data is exploited for a more efficient search for optimum materials. The advent of the Transformational AI Models Consortium (TAIMC) is an opportunity to engage with state of the art private and public AI developers to achieve the goals of the AI digital twins and AI accelerated physics models proposed in this project. Our partners at ORNL from the Advance Scientific Computing Research (ASCR) organization will collaborate in accelerating the integrated plasma material interaction simulation framework. This simulation framework will provide a platform for generating simulation data across a range of physical fidelities, including hybrid methods that produce multi-fidelity results. This data will be leveraged for AI model development, both for generation of surrogates and the automation of simulation campaigns. A part of the research below will include collaborative efforts with the TAIMC to (i) adapt data storage approaches to ensure AI-readiness, (ii) provide a protypical exemplar to inform and exercise constructed workflows, and (iii) generate and share data, using the TAIMC unified AI data standard, for foundational models that will be trained from multiple sources across the DOE complex. We will also collaborate with the TAIMC, as well as the planned AI modeling teams, to develop approaches for reducing the cost of data generation. These include tailored multi-fidelity approaches as well as fine-tuning strategies to augment general, large-scale foundational models.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRS2000

Benchmark Active Controls Technology (BACT) Wing CFD Results

The Benchmark Active Controls Technology (BACT) wing test (see chapter 8E) provides data for the validation of aerodynamic, aeroelastic, and active aeroelastic control simulation codes. These data provide a rich database for development and validation of computational aeroelastic and aeroservoelastic methods. In this vein, high-level viscous CFD analyses of the BACT wing have been performed for a subset of the test conditions available in the dataset. The computations presented in this section investigate the aerodynamic characteristics of the rigid clean wing configuration as well as simulations of the wing with a static and oscillating aileron and spoiler deflection. Two computational aeroelasticity codes extensively used at NASA Langley Research Center are implemented in this simulation. They are the ENS3DAE and CFL3DAE computational aeroelasticity programs. Both of these methods solve the three-dimensional compressible Navier-Stokes equations for both rigid and flexible vehicles, but they use significantly different approaches to the solution 6f the aerodynamic equations of motion. Detailed descriptions of both methods are presented in the following section.

Schuster, David M.↗
NASA NTRS2011

A Comparison of Multivariate and Pre-Processing Methods for Quantitative Laser-Induced Breakdown Spectroscopy of Geologic Samples

The ChemCam instrument selected for the Curiosity rover is capable of remote laser-induced breakdown spectroscopy (LIBS).[1] We used a remote LIBS instrument similar to ChemCam to analyze 197 geologic slab samples and 32 pressed-powder geostandards. The slab samples are well-characterized and have been used to validate the calibration of previous instruments on Mars missions, including CRISM [2], OMEGA [3], the MER Pancam [4], Mini-TES [5], and Moessbauer [6] instruments and the Phoenix SSI [7]. The resulting dataset was used to compare multivariate methods for quantitative LIBS and to determine the effect of grain size on calculations. Three multivariate methods - partial least squares (PLS), multilayer perceptron artificial neural networks (MLP ANNs) and cascade correlation (CC) ANNs - were used to generate models and extract the quantitative composition of unknown samples. PLS can be used to predict one element (PLS1) or multiple elements (PLS2) at a time, as can the neural network methods. Although MLP and CC ANNs were successful in some cases, PLS generally produced the most accurate and precise results.

Anderson, R. B.↗
DOE OSTI2024

Reactor Containment Passive Safety Analysis: Steam Condensation in Presence of Non-condensable Gas Scaled Experiment and Modeling

This study presents steam condensation scaled experiments and semi-empirical models in presence of nitrogen (N)—a noncondensable gas (NCG), simulating air in the reactor containment—to support water-cooled small modular reactors (SMRs) passive containment cooling system (PCCS) design and analysis. Previous experimental studies on PCCS are focused on fixed and smaller tube (mostly 2-in.) geometries and specific test condition variations, bringing challenges with geometric scaling and mismatching with SMR prototypic design. To address these challenges, this study presents steam condensation test dataset obtained from three scaled test sections of 1-, 2-, and 4-in.-diameter steam condensers with an annular/jacket cooling of 2-, 3-, and 6 in.-diameter tubes, respectively. Test data were collected for steam ranges from 58 to 63 kg/hr., and NCG flow of 4.4 to 13.3 kg/hr. Annular cooling water flow was varied to obtain required testing conditions of saturated steam inlet and fully condensed outlet. Axial temperature test data of bulk cooling water, steam and condensate were collected by thermocouples for three test sections and various steam-NCG mixing/testing conditions. A standard data reduction method was adopted—utilizing iterative and nodalized mass and heat transfer calculation—to estimate axial local heat fluxes, heat transfer coefficients (HTCs), condensation rates, film thickness, and Nusselt number. Based on the obtained dataset semi-empirical model results—a ratio of experimental and Nusselt’s theoretical HTC are presented. Such results and findings are supportive of developing scaled-up testing facility, to enable model validations and accelerate next generation of reactors development and deployment

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2024

Presentation: Reactor Containment Passive Safety Analysis: Steam Condensation in Presence of Non-condensable Gas Scaled Experiment and Modeling

This study presents steam condensation scaled experiments and semi-empirical models in presence of nitrogen--a noncondensable gas (NCG), simulating air in the reactor containment--to support water-cooled small modular reactors (SMRs) passive containment cooling system (PCCS) design and analysis. Previous experimental studies on PCCS are focused on fixed and smaller tube (mostly 2-in.) geometries and specific test condition variations, bringing challenges with geometric scaling and mismatching with SMR prototypic design. To address these challenges, this study presents steam condensation test dataset obtained from three scaled test sections of 1-, 2-, and 4-in.-diameter steam condensers with an annular/jacket cooling of 2-, 3-, and 6 in.-diameter tubes, respectively. Test data were collected for steam ranges from 58 to 63 kg/hr., and NCG flow of 4.4 to 13.3 kg/hr. Annular cooling water flow was varied to obtain required testing conditions of saturated steam inlet and fully condensed outlet. Axial temperature test data of bulk cooling water, steam and condensate were collected by thermocouples for three test sections and various steam-NCG mixing/testing conditions. A standard data reduction method was adopted--utilizing iterative and nodalized mass and heat transfer calculation to estimate axial local heat fluxes, heat transfer coefficients (HTCs), condensation rates, film thickness, and Nusselt number. Based on the obtained dataset semi-empirical model results--a ratio of experimental and Nusselt's theoretical HTC are presented. Such results and findings are supportive of developing scaled-up testing facility, to enable model validations and accelerate next generation of reactors development and deployment.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
NASA NTRS2001

Synthesis of Multispectral Bands from Hyperspectral Data: Validation Based on Images Acquired by AVIRIS, Hyperion, ALI, and ETM+

Multispectral data requirements for Earth science applications are not always studied rigorously studied before a new remote sensing system is designed. A study of the spatial resolution, spectral bandpasses, and radiometric sensitivity requirements of real-world applications would focus the design onto providing maximum benefits to the end-user community. To support systematic studies of multispectral data requirements, the Applications Research Toolbox (ART) has been developed at NASA's Stennis Space Center. The ART software allows users to create and assess simulated datasets while varying a wide range of system parameters. The simulations are based on data acquired by existing multispectral and hyperspectral instruments. The produced datasets can be further evaluated for specific end-user applications. Spectral synthesis of multispectral images from hyperspectral data is a key part of the ART software. In this process, hyperspectral image cubes are transformed into multispectral imagery without changes in spatial sampling and resolution. The transformation algorithm takes into account spectral responses of both the synthesized, broad, multispectral bands and the utilized, narrow, hyperspectral bands. To validate the spectral synthesis algorithm, simulated multispectral images are compared with images collected near-coincidentally by the Landsat 7 ETM+ and the EO-1 ALI instruments. Hyperspectral images acquired with the airborne AVIRIS instrument and with the Hyperion instrument onboard the EO-1 satellite were used as input data to the presented simulations.

Blonksi, Slawomir↗
DOE OSTI2021

Considerations regarding the Use of Computer Vision Machine Learning in Safety-Related or Risk-Significant Applications in Nuclear Power Plants

With the advancements made to date in the field of artificial intelligence (AI), significant potential exists to utilize AI capabilities for nuclear power plant (NPP) applications. AI can replicate human decision making and it is usually faster and more accurate than humans. For implementations that impact critical NPP applications (e.g., safety-related or non-safety systems that potentially affect overall plant risk), a deeper safety analysis of the AI methods is necessary. AI applied to NPP operations could resemble the use of digital I&C (DI&C) because such applications involve digital computer hardware and custom-designed software that input plant data, execute complex software algorithms, and output the results to a system or licensed human operator to potentially provoke an action. For AI methods to be compliant with current safety requirements for DI&C, AI compatibility must be evaluated, and AI-related gaps may exist that prevent the prompt deployment of AI in NPPs. This effort aims to evaluate how example AI technologies align with the DI&C safety framework, and discusses how they could be analyzed, modeled, tested, and validated in a manner similar to typical DI&C technologies. Because AI is a broad field that encompasses areas such as machine learning (ML), natural language processing, and computer vision, this research focused on a subset of methods categorized as the computer vision ML (CVML) methods. This report explores two CVML use cases, gauge reading and fire watch, considered relevant to the DI&C standards, as they could play a safety-critical role. For the gauge reading use case, a CVML-enabled technology that can read gauges at oblique angles is utilized. For the fire watch use case, a CVML-enabled technology is utilized that migrates fire watch from a manual (human) approach to automated fire detection. These use cases are mainly intended to give context to the CVML system discussion. This effort assumes the worst-case scenario, with the CVML system being used to replace a safety-related or risk-significant system, thus requiring evaluation. Evaluating CVML against most of the relevant safety requirements for DI&C yielded several CVML-specific considerations due to the uniqueness of its characteristics in comparison with typical DI&C systems. For example, CVML models often employ commonly used (open-source) datasets, and it is not always possible to determine the level of overlap among open-source datasets. Therefore, the independence of the developed CVML models when demonstrating diversity is questionable, therefore creating vulnerability to common cause failure (CCF). The design verification process is also impacted since the data overlap could result in overestimation of the software validation and verification (V&V) performance results. Section 2 of this report evaluates a list of the identified CVML-specific characteristics and discusses the resulting considerations and potential solutions in the context of each referenced requirement. A summation is provided in Section 3. This report is not to be used as a guideline. It was developed to identify and consider issues in the implementation of ML technologies used to augment activities that may have a bearing on plant operation. The report draws parallels to the use of DI&C technologies, for which many standards are available to guide their use in nuclear plant operation. It considers the technologies and some of the potential implications of their use in safety-related applications but is not intended to address regulatory or licensing related issues.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2023

Boundary Layer Controls on the Shallow-to-Deep Cumulus Transition (Final Technical Report)

This project advanced our understanding of the processes governing cumulus cloud formation and provides an improved observational basis for validating earth system models. To be specific, the project used laser- and radar-remote sensors to examine the physical properties of updrafts that rise from earth’s surface and initiate clouds deeper in the atmosphere. These updrafts comprise “thermals” and “plumes” and occur at small spatial and temporal scales (e.g., 10s of minutes, 100s of meters). These small scales preclude explicit representation in most earth system and climate models, and thus necessitate “sub-grid-scale” parameterization of updraft processes. The innovation of this project was to directly measure the size, shape, strength and water vapor content of these updrafts with Doppler and Raman lidars, respectively, and to link these updraft properties to cloud processes using vertically pointed weather radars. The resulting data sets comprise 100s of thousands of updrafts and thousands of clouds, which far exceeds previous efforts, and thereby provides a robust statistical and physical representation of these processes. From these large datasets the project produced a sequence of scientific analyses that: (1) Elucidate how variations in the turbulent structure of the convective boundary layer control shallow cumulus convection, (2) Quantify the upward transport of water vapor to cloud base via thermals and plumes, (3) Validate large-eddy simulations of updrafts and shallow convective clouds, (4) Demonstrate a size-to-strength relationship between updraft width and updraft speed, and (5) Demonstrate how updrafts interact with the stability at the top of the convective boundary layer to modulate the depth and vigor of convective clouds. These results have been disseminated via several published journal articles, academic theses, and conference presentations. Collectively these results contribute to the Atmospheric System Research (ASR) program’s goal to “improve understanding of the key cloud, aerosol, precipitation, and radiation processes that affect the Earth’s radiative balance and hydrological cycle, particularly processes that limit the predictive ability of regional and global earth system models”.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRSDate not supplied

Towards an Aviation Large Language Model by Fine-tuning and Evaluating Transformers

In the aviation domain, there are many applications for machine learning and artificial intelligence tools that utilize natural language. For example, there is a desire to know the commonalities in written safety reports such as voluntary post incidents reports or aerial wildfire operations reports to better understand the risks present. Another use-case is the possibility of extracting airspace procedures and constraints currently written in documents such as Letters of Agreement. These applications can benefit from the use of state-of-the-art natural language processing techniques when adapted to the language/phraseology specific to the aviation domain. This paper evaluates the viability of adaptation of NLP tools to the aviation domain by fine-tuning transformer based models using aviation data sets. In 2018, a novel language model based on neural units (also called transformers) was created and became known as “Bidirectional Encoder Representations from Transformers” or BERT. This architecture combined with large amounts of English training data and innovative semi-supervised training tasks set the standard for what would later emerge as Large Language Models. The performance of these models was further improved by hyperparameter tuning and refinement of the semi-supervised training task and resulted in “Robustly Optimized BERT Pre-training Approach through hyperparameter tuning” or RoBERTa models. These pre-trained Large Language Models proved to be useful for a wide variety of natural language processing tasks such as text classification and question answering through a process called fine-tuning. The transformer architecture with pre-trained weights served as the basis with the last few layers replaced with layers fine-tuned to perform a new task e.g., a layer that provides a label for the entire input text. This process of fine-tuning can also be used to adapt the models to new domains; e.g., BioBERT started with the pre-trained BERT model and was completed by additional fine-tuning and training on biomedical documents. Transformer-based architectures can also be used to create rich representations of text called embeddings which can serve as the input to other machine learning models. This allows simpler algorithms such as logistic regression to use context-rich representations of the text while still remaining quick to train and evaluate. In the world of aviation, there is a growing demand for natural language processing and understanding but the domain presents unique challenges. Due to the technical content (and specialized language) of most aviation documents, fine-tuning pre-trained Large Language Models to specific tasks has not met the benchmark on natural language processing tasks set by simpler models trained from scratch on the data. To address this deficiency, this paper evaluates the improvements from fine-tuning a Large Language Model on a large set of aviation documents using the original semi-supervised training tasks before performing specific natural language tasks. In fine-tuning, a domain-specific dataset is used on the original training task but with the pre-trained Large Language Model instead of starting from a random initialization. This approach allows the model to be adapted to the specific domain language without discarding the information gained from training on general English data. This paper utilized two major dataset types to train and assess the RoBERTa fine-tuning performance. The first are 7,057 Letters of Agreement which are Federal Aviation Administration (FAA) documents that formalize airspace operations across the national airspace system. They contain many examples of ‘aviation English’ using domain specific terminology and phrasing which serves as a representative basis to perform the semi-supervised fine-tuning. The second type is the 494 document classification labels to be used for evaluation. This down-stream evaluation aims to show the performance of the fine-tuned model, better understand how much data is needed for an effective fine-tuning, and how fine-tuning can be adapted for different applications in-the domain. After semi-supervised training, evaluation begins by encoding the documents for classification using the fine-tuned RoBERTa model. Then a logistic regression classifier is trained to label the document type and compared against our ground truth labels. This currently leads to a 82.8% accuracy on 10-fold cross validation showing improvement over baseline RoBERTa which achieved 81.0%. We plan to measure the improvements on additional tasks and it is expected that these improvements will lead to more robust models that can tackle the natural language processing challenges present in aviation datasets.

ATM↗
NASA NTRS2012

Using Apollo Sites and Soils to Compositionally Ground Truth Diviner Lunar Radiometer Observations

Apollo landing sites and returned soils afford us a unique opportunity to "ground truth" Diviner Lunar Radiometer compositional observations, which are the first global, high resolution , thermal infrared measurements of an airless body. The Moon is the most accessible member of the most abundant class of solar system objects, which includes Mercury, asteroids, and icy satellites. And the Apollo samples returned from the Moon are the only extraterrestrial samples with known spatial context. Here we compare Diviner observations of Apollo landing sites and compositional and spectral laboratory measurements of returned Apollo soils. Diviner, onboard NASA's Lunar Reconnaissance Orbiter, has three spectral channels near 8 micron that were designed to characterize the mid-infrared emissivity maximum known as the Christiansen feature (CF), a well-studied indicator of silicate mineralogy. It has been observed that thermal infrared spectra measured in simulated lunar environment (SLE) are significantly altered from spectra measured under terrestrial or martian conditions, with enhanced CF contrast and shifted CF position relative to other spectral features. Therefore only thermal emission experiments conducted in SLE are directly comparable to Diviner data. With known compositions, Apollo landing sites and soils are important calibration points for the Diviner dataset, which includes all six Apollo sites at approximately 200 m spatial resolution. Differences in measured CFs caused by composition and space weathering are apparent in Diviner data. Analyses of Diviner observations and SLE measurements for a range of Apollo soils show good agreement, while comparisons to thermal reflectance measurements under ambient conditions do not agree well, which underscores the need for SLE measurements and validates our measurement technique. Diviner observations of Apollo landing sites are also correlated with geochemical measurements of Apollo soils from the Lunar Sample Compendium. In particular, the correlations between CF and FeO and AI203 are very strong, owing to the dependence on the feldspar-mafic ratio. Our analyses suggest that Diviner data may offer an independent measure of soil iron content from the existing optical and gamma-ray spectrometer datasets.

Greenhagen, Benjamin T.↗
DOE OSTI2021

Identifying build orientation of 3D-printed materials using convolutional neural networks

The advent of additive manufacturing (AM) processes brought with it intense research into various materials and manufacturing processes. At the same time, the need for validation of material properties, as well as study and forecasting of aging, has arisen. Modern imaging techniques, like X-ray computed tomography (XCT), are a convenient vehicle for such studies; however, the large datasets they produce require novel analysis techniques to efficiently extract critical information. Here, we present our work on developing a 3D extension of the ResNet architecture to distinguish between two build orientations of tensile bars produced by AM. Using only information from XCT, our method achieves a 99.3% correct classification at a misidentification of 1%.

36 MATERIALS SCIENCE↗