Search NASA⌕ Search

SEARCH · Search NASA

Results for “Generative AI”

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 217 records · Page 12
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Brochure on the 2024 ASCR Workshop on Energy-Efficient Computing for Science

Large-scale computing has enabled numerous scientific discoveries, including ground-breaking achievements facilitated by the US Department of Energy (DOE) supercomputers and advances in applied mathematics and computer science. While important advances were made in energy efficiency to enable exascale computing, continued efforts are needed to dramatically improve the energy efficiency of the next generation of high-performance computing (HPC) systems and, more broadly, AI data centers. Without substantial improvements in energy efficiency, the energy consumption associated with computing could become a limiting factor for future scientific discovery, national security, and technological advancement.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

LLM-Inference-Bench: Inference Benchmarking of Large Language Models on AI Accelerators

Large Language Models (LLMs) have propelled groundbreaking advancements across several domains and are commonly used for text generation applications. However, the computational demands of these complex models pose significant challenges, requiring efficient hardware acceleration. Benchmarking the performance of LLMs across diverse hardware platforms is crucial to understanding their scalability and throughput characteristics. We introduce LLM-Inference-Bench, a comprehensive benchmarking suite to evaluate the hardware inference performance of LLMs. We thoroughly analyze diverse hardware platforms, including GPUs from Nvidia and AMD and specialized AI accelerators, Intel Habana and SambaNova. Our evaluation includes several LLM inference frameworks and models from LLaMA, Mistral, and Qwen families with 7B and 70B parameters. Our benchmarking results reveal the strengths and limitations of various models, hardware platforms, and inference frameworks. We provide an interactive dashboard to help identify configurations for optimal performance for a given hardware platform.

Chitty-Venkata, Krishna Teja↗
DOE OSTI2026

Discovery of hydrogen storage molecules using large language models and machine learning

Accelerating the discovery of new molecules with targeted properties is a central challenge in molecular design. In this contribution, we present an AI-driven molecular discovery framework that integrates Large Language Models (LLMs) for generative molecular design with Machine Learning (ML)-based screening to identify novel Liquid Organic Hydrogen Carrier (LOHC) candidates. Using the developed framework, LOHC molecules were systematically generated, evaluated, and refined iteratively, combining LLM-guided molecular generation and ML-predicted hydrogenation enthalpies (Δ H ), under physicochemical property constraints such as optimal melting points (MP), desired hydrogen storage capacity (wt% H 2 ), and synthetic accessibility (SA) scores. This approach enabled the discovery of 42 new LOHC candidates in two distinct campaigns, one seeded with experimentally known and another with previously computationally identified LOHCs, respectively. Although we began with different numbers of starting molecules (31 vs . 7 seed molecules), both runs yielded a comparable number of viable candidates, suggesting an influence of chemically intuitive seed molecule selection for success. Selected LOHC molecules, such as 3-methyl pyridine, 1-ethylnapthalene, 1,1-diphenylethane, and benzofuran, were experimentally tested and compared with benchmark LOHCs (toluene and 9-ethylcarbazole) for hydrogenation using a series of commercial supported metal catalysts. The order of conversion into fully hydrogenated products at 200 °C was 3-methyl pyridine (100%) > 9-ethyl carbazole (86.4%) > 2,3-benzofuran (74%) > 1,1-diphenylethane (66.9%) > 1-ethylnapthalene (66.7%) > toluene (57%), further validating the AI-guided molecular design. This study demonstrates promise of LLM-driven molecular design in conjunction with ML-based screening for accelerated discovery and design of molecules.

Harb, Hassan [Argonne National Laboratory (ANL), A↗
DOE OSTI2025

AI‐Driven Defect Engineering for Advanced Thermoelectric Materials

Thermoelectric materials offer a promising pathway to directly convert waste heat to electricity. However, achieving high performance remains challenging due to intrinsic trade-offs between electrical conductivity, the Seebeck coefficient, and thermal conductivity, which are further complicated by the presence of defects. This review explores how artificial intelligence (AI) and machine learning (ML) are transforming thermoelectric materials design. Advanced ML approaches including deep neural networks, graph-based models, and transformer architectures, integrated with high-throughput simulations and growing databases, effectively capture structure-property relationships in a complex multiscale defect space and overcome the “curse of dimensionality”. This review discusses AI-enhanced defect engineering strategies such as composition optimization, entropy and dislocation engineering, and grain boundary design, along with emerging inverse design techniques for generating materials with targeted properties. Finally, it outlines future opportunities in novel physics mechanisms and sustainability, highlighting the critical role of AI in accelerating the discovery of thermoelectric materials.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Exploring the energy landscape of RBMs: reciprocal space insights into bosons, hierarchical learning and symmetry breaking

Deep generative models have become ubiquitous due to their ability to learn and sample from complex distributions. Despite the proliferation of various frameworks, the relationships among these models remain largely unexplored, a gap that hinders the development of a unified theory of AI learning. In this work, we address two central challenges: clarifying the connections between different deep generative models and deepening our understanding of their learning mechanisms. We focus on Restricted Boltzmann Machines (RBMs), a class of generative models known for their universal approximation capabilities for discrete distributions. By introducing a reciprocal space formulation for RBMs, we reveal a connection between these models, diffusion processes, and systems of coupled bosons. Our analysis shows that at initialization, the RBM operates at a saddle point, where the local curvature is determined by the singular values of the weight matrix, whose distribution follows the Marc̆enko-Pastur law and exhibits rotational symmetry. During training, this rotational symmetry is broken due to hierarchical learning, where different degrees of freedom progressively capture features at multiple levels of abstraction. This leads to a symmetry breaking in the energy landscape, reminiscent of Landau’s theory. This symmetry breaking in the energy landscape is characterized by the singular values and the weight matrix eigenvector matrix. We derive the corresponding free energy in a mean-field approximation. We show that in the limit of infinite size RBM, the reciprocal variables are Gaussian distributed. Our findings indicate that in this regime, there will be some modes for which the diffusion process will not converge to the Boltzmann distribution. To illustrate our results, we trained replicas of RBMs with different hidden layer sizes using the MNIST dataset. Our findings not only bridge the gap between disparate generative frameworks but also shed light on the fundamental processes underpinning learning in deep generative models.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

AI-enabled Lorentz microscopy for quantitative imaging of nanoscale magnetic spin textures

The manipulation and control of nanoscale magnetic spin textures are of rising interest as they are potential foundational units in next-generation computing paradigms. Achieving this requires a quantitative understanding of the spin texture behavior under external stimuli using in situ experiments. Lorentz transmission electron microscopy (LTEM) enables real-space imaging of spin textures at the nanoscale, but quantitative characterization of in situ data is extremely challenging. Here, we present an AI-enabled phase-retrieval method based on integrating a generative deep image prior with an image formation forward model for LTEM. Our approach uses a single out-of-focus image for phase retrieval and achieves significantly higher accuracy and robustness to noise compared to existing methods. Furthermore, our method is capable of isolating sample heterogeneities from magnetic contrast, as shown by application to simulated and experimental data. This approach allows quantitative phase reconstruction of in situ data and can also enable near real-time quantitative magnetic imaging.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Cryo2StructData: A Large Labeled Cryo-EM Density Map Dataset for AI-based Modeling of Protein Structures

The advent of single-particle cryo-electron microscopy (cryo-EM) has brought forth a new era of structural biology, enabling the routine determination of large biological molecules and their complexes at atomic resolution. The high-resolution structures of biological macromolecules and their complexes significantly expedite biomedical research and drug discovery. However, automatically and accurately building atomic models from high-resolution cryo-EM density maps is still time-consuming and challenging when template-based models are unavailable. Artificial intelligence (AI) methods such as deep learning trained on limited amount of labeled cryo-EM density maps generate inaccurate atomic models. To address this issue, we created a dataset called Cryo2StructData consisting of 7,600 preprocessed cryo-EM density maps whose voxels are labelled according to their corresponding known atomic structures for training and testing AI methods to build atomic models from cryo-EM density maps. Cryo2StructData is larger than existing, publicly available datasets for training AI methods to build atomic protein structures from cryo-EM density maps. We trained and tested deep learning models on Cryo2StructData to validate its quality showing that it is ready for being used to train and test AI methods for building atomic models.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Congestion Management Solutions for Enhanced Distribution System Operations with Aggregated Distribution Grid Resources Providing Grid Services and Market Participation

Microgrids and other aggregations of distribution grid resources (DGRs) are poised to actively participate in electricity markets and provide essential grid services in the coming years. In fact, DGRs already play such a role through behind-the-meter (BTM) demand response programs and small-scale BTM dispatchable generation initiatives. At the same time, the rapid growth of artificial intelligence (AI) and cryptocurrency datacenters imposes significant, often unpredictable, demands on the power distribution system. Aggregated DGRs can serve as flexible resources that help mitigate these pressures by using available transmission and distribution capacity more efficiently, supporting resource adequacy and other reliability services, and providing bridge strategies while long-term transmission infrastructure is being developed. The impacts this activity will have on distribution networks are not fully understood and could present significant challenges for distribution utilities due to capacity constraints and the need for congestion management. Technical issues include reverse power flow, variability and possible degradation of equipment integrity, voltage violations, and customer power quality concerns. These issues will likely intensify as electricity market operators across the United States implement Federal Energy Regulatory Commission Order 2222 over the next few years.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2025

FAIR Surrogate Benchmarks Supporting AI and Simulation Research (Final Report)

Computational Science is being revolutionized by integrating AI and simulation and, in particular, by deep learning surrogate models that can replace all or part of traditional large‐scale HPC computations. Such surrogates can achieve remarkable performance improvements, as much as several orders of magnitude, and save both compute time and energy. The Surrogate Benchmark Initiative (SBI) project creates a community repository and FAIR (Findable, Accessible, Interoperable, and Reusable) data ecosystem for HPC application surrogate benchmarks. The SBI team comes from Argonne National Laboratory (ANL), Indiana University (IU), Rutgers University, the University of Tennessee, Knoxville (UTK), and the University of Virginia (UVA). SBI repositories include data, code, and all relevant collateral artifacts that the science and engineering community need to use and reuse these data sets and surrogates. SBI repositories generate active research from both the participants in SBI and the broad community of AI and domain scientists. This project develops surrogates that use several different neural nets to learn and quickly infer the results of simulations and data systems and captures them as surrogate benchmarks with a rich set of metadata covering: Data; Model; Metrics specification; Machine specification; and Science, Speed, and Power Results. We research FAIR metadata for these benchmarks. We develop application surrogate examples as benchmarks across many fields (ANL, UTK, IU, UVA). We also study non-Surrogate benchmarks that have many common features and similar issues as regards FAIRness. We work with MLCommons (UVA, UTK), which is a major machine learning benchmarking activity where we get metadata ontologies, software, and benchmarks, Benchmarks have datasets, models, and metadata and they need a technical framework developed by UTK and Rutgers and deployed by UVA. We study features of Surrogates including performance, training set size, and uncertainty quantification (Rutgers, UVA and IU).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Bifunctional Electrocatalysts with High-Entropy Alloys: Bridging Hydrogen Evolution and Oxygen Reduction

High-entropy alloys (HEAs) have emerged as a promising class of bifunctional electrocatalysts capable of simultaneously driving the hydrogen evolution reaction (HER) and the oxygen reduction reaction (ORR) with high activity and durability. Their near-equiatomic multicomponent compositions give rise to unique physicochemical characteristics, including lattice distortion, sluggish diffusion, high-entropy stabilization, and pronounced electronic heterogeneity, that collectively generate diverse and synergistic active sites inaccessible in conventional alloys. This review summarizes recent progress in HEA-based bifunctional electrocatalysis, with a focus on the fundamental mechanisms governing HER and ORR activity, stability, and selectivity. We discuss advances in synthesis strategies, ranging from confined growth and step-alloying to scalable continuous-flow methods, that enable precise control over composition, size, and surface structure. Complementary computational and data-driven approaches, including density functional theory, machine-learning-assisted screening, and descriptor development, are highlighted as essential tools for navigating the vast HEA design space and establishing structure−property relationships. Particular attention is paid to adsorption-energy distributions, multisite cooperativity, and environmental effects under realistic electrochemical conditions. Finally, we outline current challenges and future opportunities for integrating mechanistic understanding with AI-guided, closed-loop design frameworks to accelerate the discovery of next-generation HEA bifunctional electrocatalysts for sustainable energy conversion.

Alloys↗
DOE OSTI2026

ChatMPI: LLM-Driven MPI Code Generation for HPC Workloads

The Message Passing Interface (MPI) standard plays a crucial role in enabling scientific applications for parallel computing and is an essential component in high-performance computing (HPC). However, implementing MPI code manually—especially applying a proper domain decomposition and communication pattern—is a challenging and error-prone task. We present ChatMPI, an AI assistant for MPI parallelization of sequential C codes. In our analysis, we focus on testing six essential HPC workloads, which are based on Basic Linear Algebra Subprograms levels 1, 2, and 3 as well as sparse, stencil, and iterative operations. We analyze the process of creating ChatMPI by using the ChatHPC library. This lightweight large language model (LLM)–based infrastructure enables HPC experts to efficiently create and supervise trustworthy AI capabilities for critical HPC software tasks. We study the data required for training (fine-tuning) ChatMPI to generate parallel codes that not only use MPI syntax correctly but also apply HPC techniques to reduce memory communication and maximize performance by using proper work decomposition. With a relatively small training dataset composed of a few dozen prompts and fewer than 15 minutes of fine-tuning on one node equipped with two NVIDIA H100 GPUs, ChatMPI elevates trustworthiness for MPI code generation of current LLMs (e.g., Code Llama, ChatGPT-4o and ChatGPT 5). Additionally, we evaluate the performance of the MPI codes generated by ChatMPI in comparison with the ones generated by ChatGPT-4o and ChatGPT-5. The codes generated by ChatMPI provide up to a 4 × boost in performance by using better problem decomposition, communication patterns, and HPC techniques (e.g., communication avoiding).

Valero Lara, Pedro [ORNL] (ORCID:0000000214794310)↗
DOE OSTI2026

PowerModelsGAT-AI: Physics-Informed Graph Attention for Multi-System Power Flow With Continual Learning

Solving the alternating current power flow equations in real time is essential for secure grid operation, yet classical Newton–Raphson solvers can be slow under stressed conditions. Existing graph neural networks for power flow are typically trained on a single system and often degrade on different systems. We present PowerModelsGAT-AI, a physics-informed graph attention network that predicts bus voltages and generator injections. The model uses bus-type-aware masking to handle different bus types and balances multiple loss terms, including a power-mismatch penalty, using learned weights. We evaluate the model on 14 benchmark systems (4 to 6,470 buses) and train a unified model on 13 of these under contingency conditions with up to two branch outages, achieving an average normalized mean absolute error of 0.89% for voltage magnitudes and R 2 >0.99 for voltage angles. We also show continual learning: when adapting a base model to a new 1,354-bus system, standard fine-tuning causes severe forgetting with error increases exceeding 1000% on base systems, while our experience replay and elastic weight consolidation strategy keeps error increases below 2% and in some cases improves base-system performance. Interpretability analysis shows that learned attention weights correlate with physical branch parameters (susceptance: r=0.38 ; thermal limits: r=0.22 ), and feature importance analysis supports that the model captures established power flow relationships.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2025

Data Integration and Visualization for Enhanced Resilience and Sustainability in Hydropower (DIVERS-H)

U.S. hydropower plants face potential threats from shrinking water supply, rising demands, and warmer stream temperatures from various causes. Power plant owners, operators, and regulators require new tools to take advantage of and interpret the diverse range of scientific data being produced by both observational methods (for example, satellite, radar, stream gauges) and computer modeling methods that evaluate and predict how earth's dynamic systems (atmosphere, oceans, land surface, and sea ice) are changing and interacting. Combining datasets such as these with AI-based analyses introduces a novel decision support system to help users anticipate and address potential impacts on power generation stations. This new technology has been named DIVERS-H for "Data Integration and Visualization for Enhanced Resilience and Sustainability in Hydropower." In Phase I, technical feasibility was established with the development and demonstration of all the new technologies that are required. Most notably, DIVERS-H will use new artificial intelligence (AI) methods to capture the complex dynamics of water availability, demand, and environmental changes. In addition, new data management software was developed, and a prototype user interface was implemented as the precursor to a full scale decision support system. With technical research complete, the project focus now shifts to development of a commercial software product to provide users with actionable insight into water availability and the risk/resilience of critical systems at their locations of interest. Although DIVER-H was originally conceived as a tool for hydroelectric power applications, the same underlying technology can be readily applied to other water-consuming systems including coal, natural gas, oil, and nuclear power plants.

Chaudhary, Aashish [Kitware, Inc., Clifton Park, N↗
DOE OSTI2024

Identifying Outliers in AI-based Image Compression

Image compression using artificial intelligence (AI) is becoming increasingly prevalent across various fields, including scientific research. Scientific instruments can generate hundreds of images per second, and effectively compressing these images with high compression ratios is crucial for facilitating scientific discoveries. However, automatically detecting outlier cases, where compression may not have succeeded or where interesting scientific phenomena are present, poses a significant challenge. To address this, we have developed a methodology based on unsupervised machine learning techniques for detecting outlier compressed images. This methodology utilizes metrics such as peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), structural texture similarity index measure (STSIM), and deep image and structural texture similarity index (DISTS). We have evaluated our methodology on several unlabeled datasets, including microscopy and x-ray images, and have successfully identified multiple outlier images using our proposed approach. Furthermore, our approach has enabled us to identify image semantics that are valuable for post-experiment analysis by scientists.

Data Analysis↗
DOE OSTI2026

Control Room of the Future Testbed Workshop – After-Action Report

The U.S. Department of Energy’s Office of Electricity is supporting a one-year, multi-laboratory effort to define the needs and requirements for a Control Room of the Future testbed, or CROFT. The effort responds to increasing grid complexity driven by large new loads, dynamic generation resources, and the growing adoption of advanced technologies and tools, including artificial intelligence (AI) and machine learning (ML). To support safe, secure, and effective grid modernization, CROFT will focus on how emerging technologies and tools can be rigorously evaluated in realistic operational settings, with attention to human-machine interaction, cognitive load, and workforce readiness. The project team includes Argonne National Laboratory, Idaho National Laboratory, National Laboratory of the Rockies, and Pacific Northwest National Laboratory. As part of the scoping effort, the team conducted two industry-focused workshops: one at DTECH on February 5, 2026, informed by prior industry interviews, and a second on May 4, 2026, adjacent to IEEE T&D. These engagements brought together utilities, vendors, consultants, national laboratories, academia, and government stakeholders to identify and prioritize use cases, barriers, validation needs, data-sharing constraints, and near- and longer-term requirements. This feedback will directly inform CROFT’s architecture and research focus areas, ensuring the testbed is grounded in real-world operational needs and designed to evaluate emerging technologies and tools in realistic control-room environments.

artificial intelligence↗
DOE OSTI2026

Ecosystems for Scientific Computing in the Age of AI

Scientific computing is at an inflection point. Artificial intelligence (AI) is reshaping how scientific software is developed, how teams collaborate, how projects are governed, and how the next generation is trained. Drawing on insights from a 2025 workshop report, this article argues that the future of discovery will depend on agile, robust ecosystems built through socio-technical co-design—the intentional integration of technical and human systems. This perspective is essential for ensuring that future scientific computing remains trustworthy, sustainable, and scalable. It combines advances in AI, high-performance computing, and software with new models for cross-disciplinary collaboration, education, and workforce development. Key recommendations include building modular, trustworthy AI-enabled software ecosystems; enabling teams to integrate AI into scientific workflows while preserving human creativity, integrity, and rigor; and developing adaptive training pathways that keep pace with rapid technological change. By sharing these perspectives, we hope to stimulate broader community dialogue and encourage coordinated action.

AI↗
DOE OSTI2026

Genesis Data Card Schema, Template and Supporting Tools

Genesis Data Cards provide a standardized template and schema for documenting scientific datasets in support of discovery, access, interoperability, reusability, governed use, and AI usability. This release of the Genesis Data Card repository includes a versioned Markdown template, a LinkML schema with generated Pydantic and JSON artifacts, schema documentation, and example completed data cards. Validation tooling is provided to ensure that completed data cards conform to the schema prior to submission. Accompanying documentation for the structured metadata is provided as a Field Reference Guide. The schema and accompanying template provided in this repository address the call for actionable context that enables humans and AI systems to find, access, interpret, cite, and reuse data, and, when appropriate, integrate it into AI and machine learning workflows. The data card is intended to serve as a common metadata artifact intended to support standardized, cross-program dataset documentation across Department of Energy (DOE)-aligned efforts, including but not limited to Genesis Mission-related implementations, the Office of Science, National Nuclear Security Administration (NNSA), and Advanced Simulation and Computing (ASC) data governance and stewardship initiatives.

data card↗