Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Transformer model”

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 1,531 records · Page 85
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Spatially resolved phase transformation mapping in 410 stainless steel during additive manufacturing visualized via real-time infrared data

Here, in this study, we demonstrate that spatially resolved cooling curves derived from real-time infrared (IR) thermography during additive manufacturing (AM) can capture spatial variations in phase transformation temperatures through cooling curve analysis (CCA). Using this approach, we show that during laser hot-wire deposition of 410 stainless steel (410SS), the martensite start temperature (M s ) evolves dynamically throughout the build. The M s temperature is spatially nonuniform, ranging from 185 °C to 348 °C, with the lowest values toward the build center and higher values toward the upper region of the deposit. In the lower portion of the build, no M s inflection is detected via CCA, consistent with transformation occurring earlier during thermal cycling followed by tempering during subsequent thermal cycles. These trends in M s are corroborated by characterizing the microstructure by electron backscatter diffraction (EBSD). Traditionally, M s is assumed to be constant, and a single interpass temperature is applied during both deposition and residual stress modeling. Our results demonstrate that IR-derived cooling curves provide a route to spatially and temporally resolved transformation temperature tracking for dynamic interpass control and improved residual-stress modeling.

Additive manufacturing↗
NASA NTRS2009

Proceedings of the 40th Lunar and Planetary Science Conference

The 40th Lunar and Planetary Science Conference included sessions on: Phoenix: Exploration of the Martian Arctic; Origin and Early Evolution of the Moon; Comet Wild 2: Mineralogy and More; Astrobiology: Meteorites, Microbes, Hydrous Habitats, and Irradiated Ices; Phoenix: Soil, Chemistry, and Habitability; Planetary Differentiation; Presolar Grains: Structures and Origins; SPECIAL SESSION: Venus Atmosphere: Venus Express and Future Missions; Mars Polar Caps: Past and Present; SPECIAL SESSION: Lunar Missions: Results from Kaguya, Chang'e-1, and Chandrayaan-1, Part I; 5 Early Nebula Processes and Models; SPECIAL SESSION: Icy Satellites of Jupiter and Saturn: Cosmic Gymnasts; Mars: Ground Ice and Climate Change; SPECIAL SESSION: Lunar Missions: Results from Kaguya, Chang'e-1, and Chandrayaan-1, Part II; Chondrite Parent-Body Processes; SPECIAL SESSION: Icy Satellites of Jupiter and Saturn: Salubrious Surfaces; SNC Meteorites; Ancient Martian Crust: Primary Mineralogy and Aqueous Alteration; SPECIAL SESSION: Messenger at Mercury: A Global Perspective on the Innermost Planet; CAIs and Chondrules: Records of Early Solar System Processes; Small Bodies: Shapes of Things to Come; Sulfur on Mars: Rocks, Soils, and Cycling Processes; Mercury: Evolution and Tectonics; Venus Geology, Volcanism, Tectonics, and Resurfacing; Asteroid-Meteorite Connections; Impacts I: Models and Experiments; Solar Wind and Genesis: Measurements and Interpretation; Mars: Aqueous Processes; Magmatic Volatiles and Eruptive Conditions of Lunar Basalts; Comparative Planetology; Interstellar Matter: Origins and Relationships; Impacts II: Craters and Ejecta Mars: Tectonics and Dynamics; Mars Analogs I: Geological; Exploring the Diversity of Lunar Lithologies with Sample Analyses and Remote Sensing; Chondrite Accretion and Early History; Science Instruments for the Mars Science Lander; . Martian Gullies: Morphology and Origins; Mars: Dunes, Dust, and Wind; Mars: Volcanism; Early Solar System Chronology; Seek Out and Explore: Upcoming and Future Missions; Mars: Early History and Impact Processes; Mars Analogs II: Chemical and Spectral; Achondrites and their Parent Bodies; and Planning for Future Exploration of the Moon The poster sessions were: Lunar Missions: Results from Kaguya, Chang'e-1, and Chandrayaan-1; LRO and LCROSS; Geophysical Analysis of the Lunar Surface and Interior; Remote Observation and Geologic Mapping of the Lunar Surface; Lunar Spectroscopy; Venus Geology, Geophysics, Mapping, and Sampling; Planetary Differentiation; Bunburra and Buzzard Coulee: Recent Meteorite Falls; Meteorites: Terrestrial History; CAIs and Chondrules: Records of Early Solar System Processes; Volatile and Organic Compounds in Chondrites; Crashing Chondrites: Impact, Shock, and Melting; Ureilite Studies; Petrology and Mineralogy of the SNC Meteorites; Martian Meteorites; Phoenix Landing Site: Perchlorate and Other Tasty Treats; Mars Polar Atmospheres and Climate Modeling; Mars Polar Investigations; Mars Near-Surface Ice; Mars: A Volatile-Rich Planet; Mars: Geochemistry and Alteration Processes; Martian Phyllosilicates: Identification, Formation, and Alteration; Astrobiology; Instrument Concepts, Systems, and Probes for Investigating Rocks and Regolith; Seeing is Believing: UV, VIS, IR, X- and Gamma-Ray Camera and Spectrometer Instruments; Up Close and Personal: In Situ Analysis with Laser-Induced Breakdown Spectroscopy and Mass Spectrometry; Jupiter and Inscrutable Io; Tantalizing Titan; Enigmatic Enceladus and Intriguing Iapetus; Icy Satellites: Cryptic Craters; Icy Satellites: Gelid Geology/Geophysics; Icy Satellites: Cool Chemistry and Spectacular Spectroscopy; Asteroids and Comets; Comet Wild 2: Mineralogy and More; Hypervelocity Impacts: Stardust Models, LDEF, and ISPE; Presolar Grains; Early Nebular Processes: Models and Isotopes; Solar Wind and Genesis: Measurements and Interpretation; Education and Public Outreach; Mercury; Pursuing Lunar Exploration; Sources and Eruptionf Lunar Basalts; Chemical and Physical Properties of the Lunar Regolith; Lunar Dust and Transient Surface Phenomena; Lunar Databases and Data Restoration; Meteoritic Samples of the Moon; Chondrites, Their Clasts, and Alteration; Achondrites: Primitive and Not So Primitive; Iron Meteorites; Meteorite Methodology; Antarctic Micrometeorites; HEDs and Vesta; Dust Formation and Transformation; Interstellar Organic Matter; Early Solar System Chronology; Comparative Planetology; Impacts I: Models and Experiments; Impacts II: Craters and Ejecta; Mars: Volcanism; Mars: Tectonics and Dynamics; Martian Stratigraphy: Understanding the Geologic History of Mars Through the Sedimentary Rock Record; Mars: Valleys and Valley Networks; Mars: Aqueous Processes in Valles Marineris and the Southern Highlands; Mars: Aqueous Geomorphology; Martian Gullies: Morphology and Origins; Mars: Dunes, Dust, and Wind; Mars: Remote Sensing; Mars: Geologic Mapping, Photogrammetry, and Cratering; Martian Mineralogy: Constraints from Missions and Laboratory Investigations; Mars Analogs: Chemical and Physical; Mars Analogs: Sulfates and Sulfides; Missions: Approaches, Architectures, Analogs, and Actualities; Not Just Skin Deep: Electron Microscopy, Heat Flow, Radar, and Seismology Instruments and Planetary Data Systems, Techniques, and Interpretation.

Source record↗
DOE OSTI2023

Using Ultrasound Image Augmentation and Ensemble Predictions to Prevent Machine-Learning Model Overfitting

Deep learning predictive models have the potential to simplify and automate medical imaging diagnostics by lowering the skill threshold for image interpretation. However, this requires predictive models that are generalized to handle subject variability as seen clinically. Here, we highlight methods to improve test accuracy of an image classifier model for shrapnel identification using tissue phantom image sets. Using a previously developed image classifier neural network—termed ShrapML—blind test accuracy was less than 70% and was variable depending on the training/test data setup, as determined by a leave one subject out (LOSO) holdout methodology. Introduction of affine transformations for image augmentation or MixUp methodologies to generate additional training sets improved model performance and overall accuracy improved to 75%. Further improvements were made by aggregating predictions across five LOSO holdouts. This was done by bagging confidences or predictions from all LOSOs or the top-3 LOSO confidence models for each image prediction. Top-3 LOSO confidence bagging performed best, with test accuracy improved to greater than 85% accuracy for two different blind tissue phantoms. This was confirmed by gradient-weighted class activation mapping to highlight that the image classifier was tracking shrapnel in the image sets. Overall, data augmentation and ensemble prediction approaches were suitable for creating more generalized predictive models for ultrasound image analysis, a critical step for real-time diagnostic deployment.

60 APPLIED LIFE SCIENCES↗
NASA NTRSDate not supplied

Foundation AI Models for Science

Foundation Models (FM) are AI models that are designed to replace a task or an application specific model. These FM can be applied to many different downstream applications. These FM are trained using self supervised techniques and can be built on any type of sequence data. The use of self supervised learning removes the hurdle for developing a large labeled dataset for training. Most FM use transformer architecture utilizes the notion of self attention which allows the network to model the influence of distant data points to each other both in space and time. The FM models exhibit emergent properties that are induced from the data. FM can be an important tool for science. The scale of these models results in better performance for different downstream applications and these applications show better accuracy over models built from scratch. FM drastically reduces the cost of entry to build different downstream applications both in time and effort. FM for selected science datasets such as optical satellite data, can accelerate applications ranging from data quality monitoring, feature detection and prediction. FM can make it easier to infuse AI into scientific research by removing the training data bottleneck and increasing the use of science data.

Manil Maskey↗
DOE OSTI2025

A structured framework for predicting sustainable aviation fuel properties using liquid-phase FTIR and machine learning

Sustainable aviation fuels have the potential to improve efficiency, reduce emissions, and enhance energy security. To help identify viable sustainable aviation fuels and accelerate research, machine learning models have been developed to predict relevant physicochemical properties. However, many models have limited applicability, leverage data from complex analytical techniques with confined spectral ranges, or use feature decomposition methods that offer limited interpretability. Using liquid-phase Fourier Transform Infrared (FTIR) spectra, this study presents a structured method for creating accurate and interpretable property prediction models for neat molecules, aviation fuels, and blends. Liquid FTIR spectra can be collected quickly and consistently, offering high reliability, sensitivity, and component specificity using less than 2 ml of sample. The method first decomposes FTIR spectra into fundamental building blocks using non-negative matrix factorization (NMF) to enable scientific analysis of FTIR spectra attributes and fuel properties. The NMF features are then used to create five ensemble models for predicting final boiling point, flash point, freezing point, density at 15°C, and kinematic viscosity at -20°C. All models were trained using experimental property data from neat molecules, aviation fuels, and blends. The models accurately predict key properties across a broad range of neat molecules and representative fuels and blends, while enabling interpretation of relationships between compositional elements, such as functional groups or chemical classes, and their resulting properties. This demonstrates strong potential to support sustainable aviation fuel research and development. The models and data are available on an interactive web tool.

Fourier transform infrared spectroscopy↗
NASA NTRSDate not supplied

Photon Harvesting and Rocky Planet Biosignatures with 3D Climate Models

On the Earth solar photons are collected, sifted, and harvested by life at a scale that has transformed and defined the planet and continues to do so. That transformation is considered to be a key biosignature to look for in the rocky exoplanet population of our galaxy. However, the availability of photons on planetary surfaces depends on orbital and spin configuration as well as shadowing and scattering by clouds and other atmospheric phenomena. We are using a state-of-the-art 3D climate model (ROCKE-3D, Way et al. 2017) to study rocky planet surface irradiances and the implications for biosignatures and future missions such as the Habitable Worlds Observatory.

exoplanets↗
DOE OSTI2023

Blackout Diffusion: Generative Diffusion Models in Discrete-State Spaces

Typical generative diffusion models rely on a Gaussian diffusion process for training the backward transformations, which can then be used to generate samples from Gaussian noise. However, real world data often takes place in discrete-state spaces, including many scientific applications. Here, we develop a theoretical formulation for arbitrary discrete-state Markov processes in the forward diffusion process using exact (as opposed to variational) analysis. We relate the theory to the existing continuous-state Gaussian diffusion as well as other approaches to discrete diffusion, and identify the corresponding reverse-time stochastic process and score function in the continuous-time setting, and the reverse-time mapping in the discrete-time setting. As an example of this framework, we introduce “Blackout Diffusion”, which learns to produce samples from an empty image instead of from noise. Numerical experiments on the CIFAR-10, Binarized MNIST, and CelebA datasets confirm the feasibility of our approach. Generalizing from specific (Gaussian) forward processes to discrete-state processes without a variational approximation sheds light on how to interpret diffusion models, which we discuss.

Santos, Javier E.↗
NASA NTRS1990

Radar polarimetry for geoscience applications

The present volume on radar polarimetry for geoscience applications discusses wave properties and polarization, scattering matrix representation for simple targets, scattering models for point and distributed targets, polarimetric scatterometer systems and measurements, polarimetric radar system design, and polarimetric SAR applications. Attention is given to plane waves in a lossless homogeneous medium-wave polarization, polarization synthesis and response, and coordinate system transformations. Topics addressed include high- and low-frequency scattering, rough-surface scattering models, radiative transfer theory and deficiencies thereof, solutions for the radiative transfer equation, and a radiative transfer model for a forest canopy. Also discussed are network analyzer-based polarimetric scatterometers, calibration of polarimetric scatterometers, synthesized polarization response of distributed targets, and measurement of the propagation parameters of a forest canopy.

Ulaby, Fawwaz T.↗
DOE OSTI2021

Model of ramp compression of diamond from ab initio simulations

Ramp compression experiments characterize high-pressure states of matter at temperatures well below those present in shock compression. However, because temperature is typically not directly measured during ramp compression, it is uncertain how much heating occurs under these shock-free conditions. Here, we performed a series of ab initio simulations on carbon in order to match the density-stress measurements of Smith et al. [Smith et al., Nature (London) 511, 330 (2014)]. We considered isotropically as well as uniaxially compressed solid carbon in the diamond and BC8 phases, with and without defects, as well as liquid carbon. Our idealized model ascribes heating during ramp compression to an initially uniaxially compressed cell transforming isochorically into an isotropically (hydrostatic equivalent) compressed state having lower internal energy, hence higher temperature so as to conserve energy. Multiple such heating events can occur during a single ramp experiment, leading to higher temperatures than with isentropic compression. Comparison with experiments shows that heating alone does not explain the equation of state measurements on diamond, instead implying that a significant uniaxial stress component remains present at high compression. The temperature predictions of our ramp compression model remain to be verified with future laboratory measurements.

58 GEOSCIENCES↗
DOE OSTI2024

Resolving crystallographic geometrically necessary dislocations in three dimensions in a hexagonal close packed titanium alloy

Abstract Geometrically necessary dislocation (GND) content is measured from mm 3 -scaled Ti7Al three-dimensional (3D) microstructural data using a theory extended for hexagonal close packed crystals, which accounts for basal, prismatic and pyramidal ⟨ c + a ⟩ type dislocation content. The Ti7Al samples have been mechanically pre-strained to two different strain levels, and will then be strained along the same axis in uniaxial tension during simulation. Both inter- and intragranular GNDs across the microstructures have been characterized, with a large contribution of pyramidal ⟨ c + a ⟩ GNDs, consistent with the relative slip activity involved in pre-straining. The spatially resolved crystallographic GND distributions within the 3D microstructures are used to instantiate a microstructure model for forward modeling deformation simulations by a dislocation density hardening elasto-viscoplastic fast Fourier transform framework. Coarsening the voxel resolution during the initial microstructure construction procedure is shown to strongly impact both the magnitude and spatial distribution of the GNDs and in turn the forward deformation response of the pre-strained material. This study indicates that the voxel resolution desired when transferring from measured to model microstructures need not only be proportionally scaled with the microstructure but also sufficiently fine to capture the subgranular orientation gradients that may already be present in the material.

Materials Science↗
DOE OSTI2023

A Machine Learning Model for Predicting Composition of Catalytic Coprocessing Products from Molecular Beam Mass Spectra

Demand for the development of an automated and integrated refining process for biofuels has increased in recent years due to the lack of generalized process inspection tools. In bio-oil upgrading processes, all process variables are maintained based on the offline specification of intermediates and products. A lack of real-time product specifications in batch-wise monitoring can cause process failure and wasted resources. Therefore, there is a need for a fast and accurate intermediates/product specification tool that can be used for real-time specification to reduce waste and mitigate the risk of process failure. Here, to address this gap, we developed a machine learning (ML) model for predicting speciated bio-oil composition, including paraffin, iso-paraffins, olefins, naphthene, and aromatics. The model is trained using the mass spectra from upgraded products collected in the vapor phase before condensation and predicts the composition of the condensed product. Training ML models using raw mass spectra is challenging due to numerous overlapped peaks originating from different parent compounds. With this in mind, we propose a protocol that (i) transforms raw mass spectra to chemistry-inspired predefined features and (ii) trains decision tree-based models using these features. Our results show that the random forest model was robust against overfitting and had the highest accuracy compared to other models. Moreover, a stochastic ablation method determined the eight most significant features while maximizing the accuracy. Our protocol facilitates real-time compositional analysis of upgraded bio-oils and thus real-time process monitoring. Additionally, this protocol enables the rational design of efficient catalysts and the determination of optimal process conditions.

09 BIOMASS FUELS↗
DOE OSTI2021

A Modular System for Increasing Predictiveness for Extreme Climate Predictions

We know that climate change is poised to reshape our world, but we lack clear enough predictions about precisely how. The preponderance of these changes is associated with human activity, specifically the emission of CO 2 and other greenhouse gases. Problematically, projections of climate change continue to be marred by unacceptably large uncertainties which hamper informed decision-making and cost society a chance to adapt proactively and effectively. These uncertainties stem from deficiencies in predictions of future greenhouse gas emissions, but also from inaccuracies in the representation of the physical models used to predict the climate response to such emissions. The uncertainties in projections associated with the inaccurate representation of climate physics, chemistry and biology are similar to those that plagued the first global climate models developed fifty years ago, despite more than a factor 10 8 increase in computer performance. Our transformational question is then, how can the accuracy of climate projections be dramatically improved by applying recent advances in the computational and data sciences to train the models with the wealth of data being constantly collected about the ongoing changes in the climate system?

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Accelerating manufacturing for biomass conversion via integrated process and bench digitalization: a perspective

We present a perspective for accelerating biomass manufacturing via digitalization. We summarize the challenges for manufacturing and identify areas where digitalization can help. A profound potential in using lignocellulosic biomass and renewable feedstocks, in general, is to produce new molecules and products with unmatched properties that have no analog in traditional refineries. Discovering such performance-advantaged molecules and the paths and processes to make them rapidly and systematically can transform manufacturing practices. Furthermore, we discuss retrosynthetic approaches, text mining, natural language processing, and modern machine learning methods to enable digitalization. Laboratory and multiscale computation automation via active learning are crucial to complement existing literature and expedite discovery and valuable data collection without a human in the loop. Such data can help process simulation and optimization select the most promising processes and molecules according to economic, environmental, and societal metrics. We propose the close integration between bench and process scale models and data to exploit the low dimensionality of the data and transform the manufacturing for renewable feedstocks.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS1978

Continuum modeling of three-dimensional truss-like space structures

A mathematical and computational analysis capability has been developed for calculating the effective mechanical properties of three-dimensional periodic truss-like structures. Two models are studied in detail. The first, called the octetruss model, is a three-dimensional extension of a two-dimensional model, and the second is a cubic model. Symmetry considerations are employed as a first step to show that the specific octetruss model has four independent constants and that the cubic model has two. The actual values of these constants are determined by averaging the contributions of each rod element to the overall structure stiffness. The individual rod member contribution to the overall stiffness is obtained by a three-dimensional coordinate transformation. The analysis shows that the effective three-dimensional elastic properties of both models are relatively close to each other.

Nayfeh, A. H.↗
DOE OSTI2023

Neural network surrogate models for equations of state

Equation of state (EOS) data provide necessary information for accurate multiphysics modeling, which is necessary for fields such as inertial confinement fusion. Here, we suggest a neural network surrogate model of energy and entropy and use thermodynamic relationships to derive other necessary thermodynamic EOS quantities. We incorporate phase information into the model by training a phase classifier and using phase-specific regression models, which improves the modal prediction accuracy. Our model predicts energy values to 1% relative error and entropy to 3.5% relative error in a log-transformed space. Although sound speed predictions require further improvement, the derived pressure values are accurate within 10% relative error. Our results suggest that neural network models can effectively model EOS for inertial confinement fusion simulation applications.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRS1993

Image registration for luminescent paint applications

The use of pressure sensitive luminescent paints is a viable technique for the measurement of surface pressure on wind tunnel models. This technique requires data reduction of images obtained under known as well as test conditions and spatial transformation of the images. A general transform which registers images to subpixel accuracy is presented and the general characteristics of transforms for image registration and their derivation are discussed. Image resection and its applications are described. The mapping of pressure data to the three dimensional model surface for small wind tunnel models to a spatial accuracy of 0.5 percent of the model length is demonstrated.

Bell, James H.↗