SEARCH · Engineering Papers

Results for “deep transfer learning”

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.

SourceResearch connectionFrom yearThrough year

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

At least 217 records · Page 12

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Reduced-order modeling for efficient cross section library development in high-temperature gas reactor pebble-bed depletion analysis

Accurate modeling of running-in and equilibrium conditions in pebble-bed reactors (PBRs) requires precise microscopic multigroup neutron cross sections. In Griffin, deterministic neutronics calculations rely on multivariate interpolation over large cross section libraries, resulting in significant memory usage and performance bottlenecks. This work, together with a companion paper on Griffin integration, explores reduced-order models (ROMs) to replace interpolation with lightweight surrogates. Several ROM techniques are benchmarked, with deep neural networks (DNNs) demonstrating superior memory efficiency, scalability, and predictive accuracy. A total of 295 DNNs were trained to build a comprehensive isotope library, integrated into Griffin through a custom LibTorch interface for depletion analysis. Initial results demonstrate that DNN-based ROMs drastically reduce memory demands while preserving accuracy, enabling finer tabulations and additional state variables without overhead. In conclusion, the framework also supports online cross section generation and real-time DNN updates through transfer learning, improving fidelity by capturing self-shielding and evolving nuclide compositions during burnup.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2023

Data-driven prediction of α IIb β 3 integrin activation paths using manifold learning and deep generative modeling

The integrin heterodimer is a transmembrane protein critical for driving cellular process and is a therapeutic target in the treatment of multiple diseases linked to its malfunction. Activation of integrin involves conformational transitions between bent and extended states. Some of the conformations that are intermediate between bent and extended states of the heterodimer have been experimentally characterized, but the full activation pathways remain unresolved both experimentally due to their transient nature and computationally due to the challenges in simulating rare barrier crossing events in these large molecular systems. An understanding of the activation pathways can provide new fundamental understanding of the biophysical processes associated with the dynamic interconversions between bent and extended states and unveil new putative therapeutic targets. In this work, we apply nonlinear manifold learning to coarse-grained molecular dynamics simulations of bent, extended, and two intermediate states of αI I b β3 integrin to learn a low-dimensional embedding of the configurational phase space. We then train deep generative models to learn an inverse mapping between the low-dimensional embedding and high-dimensional molecular space and use these models to interpolate the molecular configurations constituting the activation pathways between the experimentally characterized states. Furthermore, this work furnishes plausible predictions of integrin activation pathways and reports a generic and transferable multi-scale technique to predict transition pathways for biomolecular systems.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

Deconvoluting experimental decay energy spectra: The O 26 case

In nuclear reaction experiments, the measured decay energy spectra can give insights into the shell structure of decaying systems. However, extracting the underlying physics from the measurements is challenging due to detector resolution and acceptance effects. The Richardson-Lucy (RL) algorithm, a deblurring method that is commonly used in optics and has proven to be a successful technique for restoring images, was applied to our experimental nuclear physics data. The only inputs to the method are the observed energy spectrum and the detector's response matrix also known as the transfer matrix. We demonstrate that the technique can help access information about the shell structure of particle-unbound systems from the measured decay energy spectrum that is not immediately accessible via traditional approaches such as χ-square fitting. For a similar purpose, we developed a machine learning model that uses a deep neural network (DNN) classifier to identify resonance states from the measured decay energy spectrum. We tested the performance of both methods on simulated data and experimental measurements. Then, we applied both algorithms to the decay energy spectrum of 26 O → 24 O + n + n measured via invariant mass spectroscopy. Here, the resonance states restored using the RL algorithm to deblur the measured decay energy spectrum agree with those found by the DNN classifier. Both deblurring and DNN approaches suggest that the raw decay energy spectrum of 26 O exhibits three peaks at approximately 0.15 MeV, 1.50 MeV, and 5.00 MeV, with half-widths of 0.29 MeV, 0.80 MeV, and 1.85 MeV, respectively.

Spectrometers & spectroscopic techniques↗

NASA NTRSDate not supplied

Biosentinel: The First Deep Space Biology CubeSat Mission- Mission Summary and Lessons Learned

Launched on Artemis-1, BioSentinel carries a biology experiment into deep space for the first time in 50 years. A 6U CubeSat form factor was utilized for the spacecraft which included technologies newly developed or adapted for operations beyond Earth orbit. The spacecraft carries onboard budding yeast, Saccharomyces cerevisiae, as an analog to human cells to test the biological response to deep space radiation. This was the maiden deep-space voyage for many of the subsystems, and the first time to evaluate their performance in flight operation. Flying a CubeSat beyond LEO comes with unique challenges with respect to trajectory uncertainty and mission operations planning. The nominal plan was a lunar fly-by, followed by an insertion into Heliocentric orbit. However, some possible scenarios included lunar eclipses that could have severely impacted the power budget during that phase of the mission, while others could have resulted in a “Retrograde” hyperbola at swing-by resulting in the spacecraft traveling inward toward Earth or even towards a collision with the lunar surface. The commissioning phase of the mission was successful and completed a week ahead of schedule. It did not come without its exciting moments and challenges. First contact with the spacecraft uncovered that the vehicle was unexpectedly tumbling after deployment, a situation that needed to be corrected urgently. The mission operations team executed a contingency plan to stabilize the spacecraft, with just moments to spare before the battery ran out of power. The BioSensor payload onboard the spacecraft is a complex instrument that includes microfluidics, fluid systems, sensor control electronics, as well at the living yeast cells. BioSentinel also included a TimePix radiation sensor implemented by JSC’s RadWorks group. Dose and Linear Energy Transfer (LET) data is compared directly to the rate of DSB-and repair events measured by the S. cerevisiae cells. BioSentinel mature nanosatellite technologies included: deep space communications and navigation, autonomous attitude control and momentum management, and micro-propulsion systems, to provide an adaptable nanosatellite platform for deep space uses. This paper discusses the performance of the BioSentinel spacecraft through the mission phase, and includes lessons learned from challenges and anomalies. BioSentinel had many successes and will be a pathfinder for future deep space CubeSats and biology missions.

BioSentinel↗

NASA NTRS2024

3D Cloud Masking Across A Broad Swath Using Multi-Angle Polarimetry and Deep Learning

Understanding the 3-dimensional structure of clouds is of crucial importance to modeling our changing climate. Active sensors, such as radar and lidar, provide accurate vertical cloud profiles, but are mostly restricted to along-track sampling. Passive sensors can capture a wide swath, but struggle to see beneath cloud tops. In essence, both types of products are restricted to two dimensions: as a cross-section in the active case, and an image in the passive case. However, multi-angle sensor configurations contain implicit information about 3D structure, due to parallax and atmospheric path differences. Extracting that implicit information can be challenging, requiring computationally expensive radiative transfer techniques. Machine learning, as an alternative, may be able to capture some of the complexity of a full 3D radiative transfer solution with significantly less computational expense. In this work, we make three contributions towards understanding 3D cloud structure from multi-angle polarimetry. First, we introduce a large-scale, publicly available dataset that fuses existing cloud products into a format more amenable to machine learning. This dataset treats multi-angle polarimetry as an input, and radar-based vertical cloud profiles as an output. Second, we describe and evaluate strong baseline machine learning models based that predict these profiles from the passive imagery. Notably, these models are trained only on center-swath labels, but can predict cloud profiles over the entire passive imagery swath. Third, we leverage the information-theoretic nature of machine learning to draw conclusions about the relative utility of various sensor configurations, including spectral channels, viewing angles, and polarimetry. These findings have implications for Earth-observing missions such as NASA’s Plankton, Aerosol, Cloud-ocean Ecosystem (PACE) and Atmosphere Observing System (AOS) missions, as well as in informing future applications of computer vision to atmospheric remote sensing.

clouds↗

NASA NTRSDate not supplied

BioSentinel: Forging the path for Deep Space CubeSat Missions

The BioSentinel mission was launched in 2022 aboard the SLS launch vehicle as part of the Artemis-I campaign and continues mission operations into 2024. The 6U CubeSat has been characterizing deep space radiation at large distances from Earth. This presentation gives a status of the mission performance to date, as well as some of the lessons learned from project. BioSentinel has achieved unprecedented performance as an SLS secondary payload due to preparation, planning, and a robust design. Pre-launch antenna and interface testing with both DSN and ESA confirmed command and data pathways and allowed for operational flexibility in the critical early hours post-deployment. Mission Operations simulations prior to launch identified potential risks and trained operators to respond in flight, preparing the team to react quickly and successfully to detumble the spacecraft and enter a power-positive state. The spacecraft would not have survived without the inclusion of the trailblazing 3D-printed composite cold gas propulsion system. The non-standard tank geometry enabled efficient use of the limited space available in the CubeSat form factor as well as the capability to detumble the spacecraft and manage momentum for extended mission durations, while providing sufficient margin to execute potential delta-V maneuvers. Following the conclusion of the primary science mission, the Linear Energy Transfer (LET) Spectrometer has continued to collect solar and galactic radiation data from its unique location in heliocentric orbit. The free space dataset offered by the BioSentinel LET is a valuable source of data for both model validation and future mission planning. As the spacecraft travels farther from Earth it is poised to provide longitudinally distributed measurements of solar particle events during solar maximum. NASA Ames led development of the BioSentinel spacecraft to operate for long durations in deep space. The novel subsystems and COTS components that comprise the BioSentinel bus can serve as a template for future deep space missions, while the lessons the team has learned from well over a year of continuous operations will enable improved performance in the generation of deep space CubeSat missions.

BioSentinel↗

DOE OSTI2023

AI-Accelerated Design of Targeted Covalent Inhibitors for SARS-CoV-2

Direct-acting antivirals for the treatment of the COVID-19 pandemic caused by the SARS-CoV-2 virus are needed to complement vaccination efforts. Given the ongoing emergence of new variants, automated experimentation, and active learning based fast workflows for antiviral lead discovery remain critical to our ability to address the pandemic’s evolution in a timely manner. While several such pipelines have been introduced to discover candidates with noncovalent interactions with the main protease (M pro ), here we developed a closed-loop artificial intelligence pipeline to design electrophilic warhead-based covalent candidates. Here, this work introduces a deep learning-assisted automated computational workflow to introduce linkers and an electrophilic “warhead” to design covalent candidates and incorporates cutting-edge experimental techniques for validation. Using this process, promising candidates in the library were screened, and several potential hits were identified and tested experimentally using native mass spectrometry and fluorescence resonance energy transfer (FRET)-based screening assays. We identified four chloroacetamide-based covalent inhibitors of M pro with micromolar affinities (K I of 5.27 μM) using our pipeline. Experimentally resolved binding modes for each compound were determined using room-temperature X-ray crystallography, which is consistent with the predicted poses. The induced conformational changes based on molecular dynamics simulations further suggest that the dynamics may be an important factor to further improve selectivity, thereby effectively lowering KI and reducing toxicity. These results demonstrate the utility of our modular and data-driven approach for potent and selective covalent inhibitor discovery and provide a platform to apply it to other emerging targets.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2025

Virtual Reality for Shoot/No-Shoot Decision Training in Law Enforcement: A Literature Review and Research Agenda

Virtual reality (VR) can materially improve “shoot / no-shoot” (SNS) training by giving officers realistic, repeatable practice making high-stakes decisions under pressure. Traditional tools—live-fire ranges and video simulators—build basics, but they cannot adapt to each officer in real time or fully mirror the complexity of the field. VR closes that gap by creating immersive scenarios that are safer, more flexible, easier to scale across units, and able to capture objective performance data. SNS decisions are not just about marksmanship; they rely on perception, judgment, memory, and the ability to hold fire when a threat is uncertain. Effective training therefore needs realism, decision complexity, and branching outcomes that reflect the true consequences of choices. These elements strengthen recognition of hostile intent while reducing false positives and building the self-control required in ambiguous situations. VR brings specific advantages: dynamic environments, full-body interaction, and the ability to measure performance with precision—enabling targeted feedback and better transfer of learning to the street. At the same time, responsible deployment must address scenario quality (credible environments and behaviors), lawful decision models, and user wellbeing (appropriate stress levels, comfort, and safety). Sandia’s VIPER Lab is positioned to lead this work. The team combines human-performance science, AI/ML, and VR/AR development with a deep equipment bench (e.g., omnidirectional treadmill, eye-tracking, haptics, multiple HMDs). This ecosystem supports building and validating next-generation SNS training that is immersive, measurable, and trustworthy. Bottom line: Investment in VR-enabled SNS training that blends evidence-based design with careful validation and legal safeguards is expected to pay off in safer, more consistent decision-making and improved community trust, delivered through training that is practical to deploy at scale.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

DOE OSTI2026

What do physics-informed DeepONets learn? Understanding and improving training for scientific computing applications

Physics-informed deep operator networks (DeepONets) have emerged as a promising approach toward numerically approximating the solution of partial differential equations (PDEs). In this work, we aim to develop further understanding of what is being learned by physics-informed DeepONets by assessing the universality of the extracted basis functions and demonstrating their potential toward model reduction with spectral methods. Results provide clarity about measuring the performance of a physics-informed DeepONet through the decays of singular values and expansion coefficients. In addition, we propose a transfer learning approach for improving training for physics-informed DeepONets between parameters of the same PDE as well as across different, but related, PDEs where these models struggle to train well. This approach results in significant error reduction and learned basis functions that are more effective in representing the solution of a PDE.

Deep operator networks↗

DOE OSTI2020

Improving the accessibility and transferability of machine learning algorithms for identification of animals in camera trap images: MLWIC2

Motion-activated wildlife cameras (or “camera traps”) are frequently used to remotely and noninvasively observe animals. The vast number of images collected from camera trap projects has prompted some biologists to employ machine learning algorithms to automatically recognize species in these images, or at least filter-out images that do not contain animals. These approaches are often limited by model transferability, as a model trained to recognize species from one location might not work as well for the same species in different locations. Furthermore, these methods often require advanced computational skills, making them inaccessible to many biologists. We used 3 million camera trap images from 18 studies in 10 states across the United States of America to train two deep neural networks, one that recognizes 58 species, the “species model,” and one that determines if an image is empty or if it contains an animal, the “empty-animal model.” Our species model and empty-animal model had accuracies of 96.8% and 97.3%, respectively. Furthermore, the models performed well on some out-of-sample datasets, as the species model had 91% accuracy on species from Canada (accuracy range 36%–91% across all out-of-sample datasets) and the empty-animal model achieved an accuracy of 91%–94% on out-of-sample datasets from different continents. Our software addresses some of the limitations of using machine learning to classify images from camera traps. By including many species from several locations, our species model is potentially applicable to many camera trap studies in North America. We also found that our empty-animal model can facilitate removal of images without animals globally. We provide the trained models in an R package (MLWIC2: Machine Learning for Wildlife Image Classification in R), which contains Shiny Applications that allow scientists with minimal programming experience to use trained models and train new models in six neural network architectures with varying depths.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2021

Deep reinforcement learning for dynamic control of fuel injection timing in multi-pulse compression ignition engines

Conventional compression-ignition (CI) engines have long offered high thermal efficiencies and torque across a wide range of loads, but often require extensive exhaust gas treatment that decreases efficiency to meet ever-increasing emissions regulations. One strategy to decrease emissions is to split the fuel injection into a series of smaller injections. In this paper, we explore a new way of discovering optimal control strategies for the next generation of CI engines using deep reinforcement learning (DRL). We outline a DRL procedure to maximize the weighted reward of engine work while minimizing end-of-cycle NO x emissions. Through the procedure outlined in this paper, we show that the DRL agent is able to reduce NO x emissions threefold while only decreasing network by 2%. We demonstrate the use of transfer learning (TL) across hierarchies of physical models to accelerate the learning process, making this approach feasible for a range of control problems within this space. This paper presents a framework and demonstration for using DRL to design control systems in technology areas such as multi-pulse engine control where a hierarchy of models combined with multi-objective rewards are used for optimal operation.

33 ADVANCED PROPULSION SYSTEMS↗

NASA NTRSDate not supplied

Eleven Countries, an Integrated Spacecraft: the Story of International Collaboration that Built the Orion Spacecraft and Powered the Success of the Artemis I Mission

The quest to return humans to the Moon in the next step towards humanity's exploration of space is more alive than ever. After a great deal of achievements, failures, and lessons learned, the Artemis I mission set o to the Moon on November 16, 2022, with the goal of testing a new rocket, the Space Launch System, and a new spacecraft, Orion: designed, assembled, and tested across two continents, and 11 countries. Behind this mission, decades of experience with the International Space Station, Autonomous Transfer Vehicle operations, and many other program collaborations built the know-how on how to succeed together in the toughest environment | deep space. The Artemis I mission proved to be an incredible success, meeting 161 total mission objectives, including 21 developed during the flight based on outperforming spacecraft. It was also a case-study in international collaboration, given that ESA, NASA, and industry partners Airbus and Lockheed Martin for the first time had to design, build, test, and fly a fully integrated human-rated spacecraft, with most critical functions dependent and interconnected across U.S. and European systems. The U.S.-built Orion Crew Module and Crew Module Adapter and European-built European Service Module (ESM) shared critical interfaces and commodities, from propulsion, avionics, active/passive thermal, electrical power generation, storage and distribution to the software that managed it all. In this paper, we will describe relevant aspects of the integrated spacecraft design, providing context for the challenges that the team faced in all phases required to get Orion ready to fly, and provide a direct account of how the joint team formed, trained, and supported the operations of the successful Artemis I mission. We will also explore the evolution of the partnerships, given that these allow a multi-national e ort to sustain the program production, share costs, leverage a broader base of engineering expertise, and build more diverse capabilities over the long haul to support the Artemis goals and objectives. Lastly, we will cover critical lessons learned and how the Orion Program has implemented these in preparation of the next Artemis missions to repeat the success of Artemis I. The purpose of this paper is to document knowledge we gained and lessons we learned through the development of an integrated Orion spacecraft, since it is imperative we build on this now, at the dawn of the Artemis Program, an international endeavor to push human space exploration.

Deep Space Exploration↗

DOE OSTI2021

DeepShadows: Separating low surface brightness galaxies from artifacts using deep learning

Searches for low-surface-brightness galaxies (LSBGs) in galaxy surveys are plagued by the presence of a large number of artifacts (e.g., objects blended in the diffuse light from stars and galaxies, Galactic cirrus, star-forming regions in the arms of spiral galaxies, etc.) that have to be rejected through time consuming visual inspection. In future surveys, which are expected to collect hundreds of petabytes of data and detect billions of objects, such an approach will not be feasible. We investigate the use of convolutional neural networks (CNNs) for the problem of separating LSBGs from artifacts in survey images. We take advantage of the fact that we have available a large number of labeled LSBGs and artifacts from the Dark Energy Survey, that we use to train, validate, and test a CNN model. That model, which we call DeepShadows , achieves a test accuracy of 92.0%, a significant improvement relative to feature-based machine learning models. We also study the ability to use transfer learning to adapt this model to classify objects from the deeper Hyper-Suprime-Cam survey, and we show that after the model is retrained on a very small sample from the new survey, it can reach an accuracy of 87.6%. Finally, these results demonstrate that CNNs offer a very promising path in the quest to study the low-surface-brightness universe.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2023

Machine learning nonequilibrium electron forces for spin dynamics of itinerant magnets

Abstract We present a generalized potential theory for conservative as well as nonconservative forces for the Landau-Lifshitz magnetization dynamics. Importantly, this formulation makes possible an elegant generalization of the Behler-Parrinello machine learning (ML) approach, which is a cornerstone of ML-based quantum molecular dynamics methods, to the modeling of force fields in adiabatic spin dynamics of out-of-equilibrium itinerant magnetic systems. We demonstrate our approach by developing a deep-learning neural network that successfully learns the electron-mediated exchange fields in a driven s-d model computed from the nonequilibrium Green’s function method. We show that dynamical simulations with forces predicted from the neural network accurately reproduce the voltage-driven domain-wall propagation. Our work also lays the foundation for ML modeling of spin transfer torques and opens a avenue for ML-based multi-scale modeling of nonequilibrium dynamical phenomena in itinerant magnets and spintronics.

36 MATERIALS SCIENCE↗

NASA NTRSDate not supplied

Biological Research and Space Health Enabled by Machine Learning to Support Deep Space Missions

A key science goal of the NASA “Moon to Mars” campaign is to understand how biology responds to the Lunar, Martian, and deep space environments in order to advance fundamental knowledge, reduce risk, and support safe, productive human space missions. Through the powerful emerging computer science approaches of artificial intelligence (AI) and machine learning (ML), a paradigm shift has begun in biomedical science and engineered astronaut health systems, to enable Earth-independence and autonomy of mission operations. We present a decadal view of AI/ML architecture to support deep space mission goals, developed in concert with leaders in the field. We describe current AI/ML methods to support 1) fundamental biology, 2) in situ analytics, 3) high performance computing hardware, 4) automated science, 5) self-driving labs, 6) remote data management, 7) integrated real-time mission biomonitoring, and 8) a Precision Space Health system. Cutting-edge AI/ML approaches that can be integrated to support these domains include active learning, explainable AI, adaptive learning, causal inference, knowledge graphs, federated learning, transfer learning, and large language models. Finally, we present results from several current ML projects that are underway in the field to address key challenges of small sample n, high feature count, heterogeneity, and sparse data. These include 1) connecting omics data to phenotypic data using an ensemble model to infer causality of spaceflight rodent liver health disruption, 2) usage of explainable ML to interrogate the muscular underpinnings of spaceflight muscle atrophy, 3) ML models analyzing and determining directed acyclic graphs of human space health risk leveraging rodent bone datasets, 4) usage of large pre-trained models connecting biomedical knowledgebases with small spaceflight datasets to understand gene-to-gene interaction networks, and 5) a suite of benchmarked open science datasets (spaceflight mouse liver; radiation DNA damage) enabling programmers to identify the best ML algorithms to answer space biological science questions.

space biology↗

NASA NTRSDate not supplied

Biological Research and Space Health Enabled by Machine Learning to Support Deep Space Missions

space biology↗

DOE OSTI2024

Dataset for "Evaluating Deep Learning Approaches for Predictions in Unmonitored Basins with Continental-scale Stream Temperature Models" Willard et al. (2024)

This data release provides all data and code used in the paper " "Evaluating Deep Learning Approaches for Predictions in Unmonitored Basins with Continental-scale Stream Temperature Models" Willard et al. (2024)" to model stream temperature, evaluate, and assess results. The associated manuscript explores current open questions in prediction in ungauged and unmonitored basins concerning top-down versus bottom-up approaches, tradeoffs between data available and input requirements, and the appropriate representation of catchment attributes as inputs to deep learning models. Modeling was done primarily with long short-term memory (LSTM) models, and stream site coverage spans 1362 locations across the conterminous United States. The data is organized into these items items:Code repository and data for the paper " "Evaluating Deep Learning Approaches for Predictions in Unmonitored Basins with Continental-scale Stream Temperature Models" Willard et al. (2024)".Code: stream_temp_ml_regionalization.zip contains the code repositoryData to run the code: - data_dir.zip -- contains all files that should be moved to the "DATA_DIR" variable defined in the "set_env_vars.sh" script in the code repository- metadata_dir.zip -- contains all files that should be moved to the "METADATA_DIR" variable defined in the "set_env_vars.sh" script in the code repository- error_analysis_attribute_and_groundwater_dir.zip - workflows for the extended error analysis by stream attribute and groundwater influenceData produced by the code and used in the paper:- outputs_dir.zip - contains model output and results (outputs_dir/results), model weights (outputs_dir/models), and all other outputs used for the paper including feature importances.To cite this code, please use the following BibTeX or MLA entries:bibtex:@misc{willard2024streamdata, author = {Jared Willard and Fabio Ciulla and Helen Weierbach and Vipin Kumar and Charuleka Varadharajan}, title = {Dataset for "Evaluating Deep Learning Approaches for Predictions in Unmonitored Basins with Continental-scale Stream Temperature Models"}, year = {2024}, doi = {10.15485/2448016}, publisher = {ESS-DIVE Repository}, url = {https://doi.org/10.15485/2448016}}MLA: Willard, Jared, et al. Dataset for "Evaluating Deep Learning Approaches for Predictions in Unmonitored Basins with Continental-scale Stream Temperature Models". 2024. ESS-DIVE Repository, doi:10.15485/2448016.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2025

Micrometer: Micromechanics transformer for predicting full field mechanical responses of heterogeneous materials

Predicting mechanical responses of heterogeneous materials across scales remains a significant challenge. Traditional computational methods often struggle with complex and multiscale nature of these materials, limiting their effectiveness in real-world applications. Here, in this paper, we introduce Micrometer, a vision transformer based deep learning model designed to predict full field mechanical responses of heterogeneous materials, bridging the gap between computer vision and solid mechanics problems. We show that Micrometer, trained on a large-scale high-resolution dataset of 2D fiber-reinforced composites, can achieve state-of-the-art performance in predicting microscale strain fields across a wide range of material properties and loading conditions. Our model demonstrates accuracy and computational efficiency in applications such as computational homogenization and multiscale modeling, reducing computational time by up to two orders of magnitude compared to conventional numerical solvers while maintaining less than 1 % errors in predicting macroscale stress fields. Furthermore, we showcase Micrometer’s adaptability through transfer learning experiments on new materials with limited data, highlighting its potential to tackle diverse scenarios in computational solid mechanics. These results represent a significant step towards AI-driven innovation in materials science, addressing the limitations of traditional numerical methods and paving the way for more efficient simulations of heterogeneous materials across various industrial applications.

Composite materials↗