Search NASA⌕ Search

SEARCH · Search NASA

Results for “Advanced Scientific Computing”

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 739 records · Page 41
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Advances in 3D transient plasma dynamics and control through MHD and hybrid fluid-kinetic simulations with JOREK

Transient phenomena and their control are of high relevance in magnetic confinement fusion plasmas to guarantee a stable and safe plasma operation. Interpretative simulations can maximize the insights gained from experiments on present machines and predictive simulations can help in the preparation of design, mitigation techniques and operational scenarios for future devices. In this article, we provide an overview of recent advances and novel scientific results obtained with the 3D non-linear hybrid fluid-kinetic code JOREK, covering physics of plasma transients from the core to the scrape-off layer (SOL) both for tokamak and stellarator devices. Substantial progress was made in the physics understanding, model validation with experiments and experiment interpretation, thus, giving confidence for predictions to devices like DTT, ITER and DEMO. The topics addressed comprise a wide range: the edge physics of new operation scenarios and edge localized mode suppression; major disruptions with a focus on runaway electrons and vertical displacement events as well as disruption mitigation by shattered pellet injection; the physics mechanisms and operational limits of the flux pumping regime for sawtooth control; MHD limits of stellarators and work towards incorporating advanced edge/SOL/exhaust dynamics; continuing improvements of the code for more efficient hybrid simulations on conventional and accelerated high performance computing architectures.

disruptions↗
DOE OSTI2025

From PINNs to PIKANs: recent advances in physics-informed machine learning

Physics-Informed Neural Networks (PINNs) have emerged as a key tool in Scientific Machine Learning since their introduction in 2017, enabling the efficient solution of ordinary and partial differential equations using sparse measurements. Over the past few years, significant advancements have been made in the training and optimization of PINNs, covering aspects such as network architectures, adaptive refinement, domain decomposition, and the use of adaptive weights and activation functions. A notable recent development is the Physics-Informed Kolmogorov-Arnold Networks (PIKANS), which leverage a representation model originally proposed by Kolmogorov in 1957, offering a promising alternative to traditional PINNs. In this review, we provide a comprehensive overview of the latest advancements in PINNs, focusing on improvements in network design, feature expansion, optimization techniques, uncertainty quantification, and theoretical insights. We also survey key applications across a range of fields, including biomedicine, fluid and solid mechanics, geophysics, dynamical systems, heat transfer, chemical engineering, and beyond. Lastly, we review computational frameworks and software tools developed by both academia and industry to support PINN research and applications.

Kolmogorov-Arnold networks↗
DOE OSTI2021

Laboratory earthquake forecasting: A machine learning competition

Earthquake prediction, the long-sought holy grail of earthquake science, continues to confound Earth scientists. Could we make advances by crowdsourcing, drawing from the vast knowledge and creativity of the machine learning (ML) community? We used Google’s ML competition platform, Kaggle, to engage the worldwide ML community with a competition to develop and improve data analysis approaches on a forecasting problem that uses laboratory earthquake data. The competitors were tasked with predicting the time remaining before the next earthquake of successive laboratory quake events, based on only a small portion of the laboratory seismic data. The more than 4,500 participating teams created and shared more than 400 computer programs in openly accessible notebooks. Complementing the now well-known features of seismic data that map to fault criticality in the laboratory, the winning teams employed unexpected strategies based on rescaling failure times as a fraction of the seismic cycle and comparing input distribution of training and testing data. In addition to yielding scientific insights into fault processes in the laboratory and their relation with the evolution of the statistical properties of the associated seismic data, the competition serves as a pedagogical tool for teaching ML in geophysics. The approach may provide a model for other competitions in geosciences or other domains of study to help engage the ML community on problems of significance.

58 GEOSCIENCES↗
DOE OSTI2022

SMALE: Enhancing Scalability of Machine Learning Algorithms on Extreme-Scale Computing Platforms

Deployment and execution of machine learning tasks on extreme-scale computing platforms face several significant technical challenges: 1) High computing cost incurred by dense networks – The computing workload of deep networks with densely-connected topology increases rapidly with the network size, imposing a non-scalable computing model of extreme-scale computing platforms; 2) Non-optimized workload distribution – Many advanced deep learning algorithms, e.g., sparsification and irregular net-work topology, produce very unbalanced workload distribution on extreme-scale computing platforms. The computation efficiency is greatly hindered by the incurred data and computation redundancies as well as long tails of the node with extensive workload; 3) Constraints in data movement and I/O bottle-neck – Inter-node data movement in extreme-scale computing platforms are associated with high energy and latency costs, and subject to the constraints of I/O bandwidth; and 4) Generalization of algorithm realization and acceleration on computing platforms – The large varieties of machine learning algorithms and structures of extreme-scale computing platforms make the derivation of a generalized algorithm realization and acceleration method very challenging, which, however, is the requirement by domain scientists and interested users. We call the above challenges Smale’s Problems in Machine Learning and Understanding for High-Performance Computing Scientific Discovery. The objective of our three-year research project is to develop a holistic innovation set at structure, assembly, and acceleration layers of machine learning algorithms to address the above challenges in algorithm deployment and execution. Three tasks are particularly performed, including: At the algorithm structure level, we investigate the techniques that can structurally sparsify on the topology of deep networks for computing workload reduction. We also study clustering and pruning techniques that can optimize the workload distributions over the extreme-scale computing platforms; At the algorithm assembly level, we derive a unified learning framework for unsupervised transfer learning and dynamic growing capabilities. Novel training methods are also exploited to enhance the training efficiency of the proposed framework; At the algorithm acceleration level, we will develop a series of techniques that can accelerate the computation of sparse matrix operations, which are one of the core executions in deep learning and optimize memory access of the concerned platforms. Our proposed techniques attack the fundamental problems in machine learning algorithms running on extreme-scale computing platforms by vertically integrating the solutions at three closely entangled layers, paving the long-term scaling path of machine learning applications under DOE context. Three tasks corresponding to the above respective research orientations are performed during the three-year project period with our collaborators at ORNL. The outcome of the proposed project is anticipated to form a holistic solution set of novel algorithms and network topologies, efficient training techniques, and fast acceleration methods to promote the computing scalability of the machine learning applications of particular interest to DOE.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Resolving Lonsdaleite's decade-long controversy: Atomistic insights into a metastable diamond polymorph

Lonsdaleite, a theoretically proposed hexagonal diamond polymorph, has remained at the center of a five-decade scientific controversy since its 1967 identification. While some studies claim it exhibits superior hardness through compression-induced structural changes, others contend it is merely a stacking-faulted cubic diamond. Meteoritic samples and synthetic preparations have yielded conflicting evidence, with even advanced characterisation techniques like XRD and TEM failing to provide definitive proof. In this work, we employ first-principles density functional theory (DFT) and molecular dynamics (MD) simulations to generate unambiguous theoretical fingerprints through XRD, Raman, and SAED patterns that distinguish true Lonsdaleite from cubic diamond and its defective variants. Our atomistic approach quantifies the thermodynamic metastability of Lonsdaleite under realistic pressure-temperature conditions, reveals distinct spectral signatures through simulated Raman and resolves the structural ambiguity through generalised stacking fault energy analysis. By establishing clear criteria for definitive identification, this study provides long-awaited clarity to the Lonsdaleite debate while offering a robust computational framework for characterising metastable carbon phases in meteoritic, synthetic and industrial materials.

DFT↗
DOE OSTI2016

Basic Research Needs Workshop on Synthesis Science for Energy Relevant Technology

The technology that lies at our fingertips becomes more powerful each day. Smartphones connect us instantly to family, friends, and co-workers around the globe; give us access to a limitless stream of information; control the heating in our homes; and serve as our cameras, calculators, flashlights, music players, boarding passes and, on occasion, our phones. Cars are ever more fuel-efficient, safer, semi-autonomous, and have more computing power than the systems that guided humankind to the moon. LED lighting and solar panels are becoming commonplace, replacing less efficient technologies and expanding the energy options available worldwide. Novel polymers and nanoparticles are playing a crucial role in enhanced oil recovery. None of these advances would have been possible without the discovery and development of, and ability to create, new materials and chemical processes. Now imagine what our world would be like if we could accelerate those discoveries a thousandfold. What if the only limit to synthesizing new forms of matter were the imagination? We could build complex assemblies of atoms and molecules with architectures and capabilities far exceeding those of materials found in nature—for example, develop catalysts that turn garbage into fuels, design solar cells to power our homes directly from sunlight, make batteries with the energy density of gasoline, and create one- and two-dimensional solids that transport charge hundreds of times faster than silicon or allow us to build quantum bits based on the spins of electrons or photons to realize the promise of “beyond Moore’s law” computing. Advances in synthesis science are required to bring about this future—we not only must know how to design new molecules and materials with desired functions and properties through theory and computational techniques; we also must be able to make the materials we envision. New approaches to discovering as yet unimagined matter require a sea change in the way we think about the science of synthesis. Chemical and materials sciences have traditionally focused on understanding structure–function relationships with the goal of predicting where the atoms should be placed to achieve a targeted property or process. Much less effort has been directed toward a predictive science of synthesis—understanding how to get the atoms where they need to go to achieve the desired structure. This report, which is the result of the Basic Energy Sciences Workshop on Basic Research Needs for Synthesis Science for Energy Technologies, lays out the scientific challenges and opportunities in synthesis science. The workshop was attended by more than 100 leading national and international scientific experts. Its five topical and two crosscutting panels identified four priority research directions (PRDs) for realizing the vision of predictive, science-directed synthesis.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2017

Considerations on the Use of Custom Accelerators for Big Data Analytics

Accelerators, including Graphic Processing Units (GPUs) for gen- eral purpose computation, many-core designs with wide vector units (e.g., Intel Phi), have become a common component of many high performance clusters. The appearance of more stable and reliable tools tools that can automatically convert code written in high-level specifications with annotations (such as C or C++) to hardware de- scription languages (High-Level Synthesis - HLS), is also setting the stage for a broader use of reconfigurable devices (e.g., Field Pro- grammable Gate Arrays - FPGAs) in high performance system for the implementation of custom accelerators, helped by the fact that new processors include advanced cache-coherent interconnects for these components. In this chapter, we briefly survey the status of the use of accelerators in high performance systems targeted at big data analytics applications. We argue that, although the progress in the use of accelerators for this class of applications has been sig- nificant, differently from scientific simulations there still are gaps to close. This is particularly true for the ”irregular” behaviors exhibited by no-SQL, graph databases. We focus our attention on the limits of HLS tools for data analytics and graph methods, and discuss a new architectural template that better fits the requirement of this class of applications. We validate the new architectural templates by mod- ifying the Graph Engine for Multithreaded System (GEMS) frame- work to support accelerators generated with such a methodology, and testing with queries coming from the Lehigh University Benchmark (LUBM). The architectural template enables better supporting the task and memory level parallelism present in graph methods by sup- porting a new control model and a enhanced memory interface. We show that out solution allows generating parallel accelerators, pro- viding speed ups with respect to conventional HLS flows. We finally draw conclusions and present a perspective on the use of reconfig- urable devices and Design Automation tools for data analytics.

Data Analytics↗
DOE OSTI2018

Analysis Report for Exascale Storage Requirements for Scientific Data.

Over the next 10 years, the Department of Energy will be transitioning from Petascale to Exascale Computing resulting in data storage, networking, and infrastructure requirements to increase by three orders of magnitude. The technologies and best practices used today are the result of a relatively slow evolution of ancestral technologies developed in the 1950s and 1960s. These include magnetic tape, magnetic disk, networking, databases, file systems, and operating systems. These technologies will continue to evolve over the next 10 to 15 years on a reasonably predictable path. Experience with the challenges involved in transitioning these fundamental technologies from Terascale to Petascale computing systems has raised questions about how these will scale another 3 or 4 orders of magnitude to meet the requirements imposed by Exascale computing systems. This report is focused on the most concerning scaling issues with data storage systems as they relate to High Performance Computing- and presents options for a path forward. Given the ability to store exponentially increasing amounts of data, far more advanced concepts and use of metadata will be critical to managing data in Exascale computing systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Earth and Space Science Informatics Perspectives on Integrated, Coordinated, Open, Networked (ICON) Science

Abstract This article is composed of three independent commentaries about the state of Integrated, Coordinated, Open, Networked (ICON) principles (Goldman, et al., 2021b, https://doi.org/10.1029/2021EO153180 ) in Earth and Space Science Informatics (ESSI) and includes discussion on the opportunities and challenges of adopting them. Each commentary focuses on a different topic: (Section 2) Global collaboration, cyberinfrastructure, and data sharing; (Section 3) Machine learning for multiscale modeling; (Section 4) Aerial and satellite remote sensing for advancing Earth system model development by integrating field and ancillary data. ESSI addresses data management practices, computation and analysis, and hardware and software infrastructure. Our role in ICON science therefore involves collaborative work to assess, design, implement, and promote practices and tools that enable effective data management, discovery, integration, and reuse for interdisciplinary work in Earth and space science disciplines. Networks of diverse people with expertise across Earth, space, and data science disciplines are essential for efficient and ethical exchanges of findable, accessible, interoperable, and reusable (FAIR) research products and practices. Our challenge is then to coordinate the development of standards, curation practices, and tools that enable integrating and reusing multiple data types, software, multi‐scale models, and machine learning approaches across disciplines in a way that is as open and/or FAIR as ethically possible. This is a major endeavor that could greatly increase the pace and potential of interdisciplinary scientific discovery.

58 GEOSCIENCES↗
DOE OSTI2024

Bayesian Conavigation: Dynamic Designing of the Material Digital Twins via Active Learning

Scientific advancement is universally based on the dynamic interplay between theoretical insights, modeling, and experimental discoveries. However, this feedback loop is often slow, including delayed community interactions and the gradual integration of experimental data into theoretical frameworks. This challenge is particularly exacerbated in domains dealing with high-dimensional object spaces, such as molecules and complex microstructures. Hence, the integration of theory within automated and autonomous experimental setups, or theory in the loop-automated experiment, is emerging as a crucial objective for accelerating scientific research. The critical aspect is to use not only theory but also on-the-fly theory updates during the experiment. Furthermore, we introduce a method for integrating theory into the loop through Bayesian conavigation of theoretical model space and experimentation. Our approach leverages the concurrent development of surrogate models for both simulation and experimental domains at the rates determined by latencies and costs of experiments and computation, alongside the adjustment of control parameters within theoretical models to minimize epistemic uncertainty over the experimental object spaces. This methodology facilitates the creation of digital twins of material structures, encompassing both the surrogate model of behavior that includes the correlative part and the theoretical model itself. While being demonstrated here within the context of functional responses in ferroelectric materials, our approach holds promise for broader applications, such as the exploration of optical properties in nanoclusters, microstructure-dependent properties in complex materials, and properties of molecular systems.

Microscopy↗
DOE OSTI2024

AIR Framework for Physics-Inspired Artificial Intelligence in High Energy Physics (Final Report)

The FAIR4HEP project was a collaboration between Argonne National Laboratory, University of Illinois Urbana-Champaign, Massachusetts Institute of Technology, University of Minnesota, and the University of California San Diego funded by the US Department of Energy, Office of Science, Office of Advanced Scientific Research (ASCR) from 2020 to 2023. The primary focus of the FAIR4HEP project was to advance our understanding of the relationship between data and artificial intelligence (AI) models by exploring relationships among them through the development of findable, accessible, interoperable, and reusable (FAIR) frameworks. Using high-energy physics (HEP) as the science driver, this project developed a FAIR framework to advance our understanding of AI, provide new insights to apply AI techniques and provide an environment where novel approaches to AI can be explored. This final report summarizes the accomplishments of the MIT group.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Institute for Catalysis in Energy Processes (ICEP) (Final Report)

The Institute for Catalysis in Energy Processes (ICEP) was a multi-PI program located at the Northwestern University Center for Catalysis and Surface Science from 2006-2024. Over several renewal cycles and several organizing themes, ICEP addressed fundamental questions in catalysis science. In turn, the scientific questions addressed were directly relevant to the efficient use of the nation’s resources to produce fuels and commodity chemicals, harness alternate energy sources for chemical reactions such as light or electricity, reduce emissions and waste, and minimize the impact of our use of plastics. Selective oxidation, (oxidative) dehydrogenation, deNOx, CO 2 reduction, hydrogen release, polymer decomposition / depolymerization, and many other reactions of intense interest to the US Department of Energy were studied in the center. Major ICEP strengths were in catalyst synthesis, in measurements that elucidated catalyst properties, in reaction mechanisms and kinetics, and in predictive theory and modeling. ICEP researchers were often the inventors or developers of materials and methods that were brought to bear on the Institute’s catalytic systems. Key characterization tools advanced through this project included resonance Raman and related techniques, sum frequency generation, and X-ray standing wave spectroscopy. Center members relied extensively on computational tools such as density functional theory and microkinetic modeling to fully investigate catalyst structures and catalytic mechanisms. ICEP researchers were early developers of atomic layer deposition (ALD) for catalyst synthesis and were early pioneers of metal organic frameworks for catalysis. ICEP innovations are found in areas of intense research such as single-atom catalysts, catalytic deconstruction of plastics, and catalytic MOFs, to name a few. Research areas initiated by ICEP indirectly led to DOE EFRCs, startup companies, and other achievements. Over its 18 years of existence, ICEP involved two different PIs (Stair and Notestein) and 24 senior investigators across several academic disciplines at Northwestern University. There were tight collaborations with Argonne National Laboratory, including specialized equipment and experiments permanently located there. The center supported ~400 person-years of effort by graduate students and postdocs, either directly through DOE support or indirectly by leveraging independent support provided to the trainees, e.g. through the NSF GRFP or internal fellowships. The project resulted in 328 manuscripts and numerous conference presentations.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Roadmap on methods and software for electronic structure based simulations in chemistry and materials

This Roadmap article provides a succinct, comprehensive overview of the state of electronic structure methods and software for molecular and materials simulations. Seventeen distinct sections collect insights by 51 leading scientists in the field. Each contribution addresses the status of a particular area, as well as current challenges and anticipated future advances, with a particular eye towards software related aspects and providing key references for further reading. Foundational sections cover density functional theory and its implementation in real-world simulation frameworks, Green's function based many-body perturbation theory, wave-function based and stochastic electronic structure approaches, relativistic effects and semiempirical electronic structure theory approaches. Subsequent sections cover nuclear quantum effects, real-time propagation of the electronic structure, challenges for computational spectroscopy simulations, and exploration of complex potential energy surfaces. The final sections summarize practical aspects, including computational workflows for complex simulation tasks, the impact of current and future high-performance computing architectures, software engineering practices, education and training to maintain and broaden the community, as well as the status of and needs for electronic structure based modeling from the vantage point of industry environments. Overall, the field of electronic structure software and method development continues to unlock immense opportunities for future scientific discovery, based on the growing ability of computations to reveal complex phenomena, processes and properties that are determined by the make-up of matter at the atomic scale, with high precision.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Solar Spectrum Conversion for an Algae Bioreactor (CRADA Final Report)

This project focused on developing advanced optical coatings to improve solar energy utilization. The research aimed to create lanthanide-doped upconversion nanoparticles (UCNPs) capable of capturing unused near-infrared (NIR) light from the sun and converting it into visible light (blue and red photons) that can be used for photosynthesis. The primary goal was to identify, synthesize, and integrate highly efficient UCNPs into a transparent thin-film device. Through a comprehensive workflow involving computer simulations, high-throughput robotic synthesis, and detailed optical characterization, the project successfully developed a high-performance material. The key technical achievement was the creation of a core-shell UCNP (NaYF₄:20%Yb³⁺, 2%Er³⁺ coated with a 10 nm NaYF₄ shell) that demonstrated a quantum yield of 3.2% for converting 980 nm NIR light into visible light. Transparent thin films fabricated from these nanoparticles showed excellent optical properties, confirming their potential for practical applications. This research adds to the scientific understanding of energy transfer in lanthanide materials and demonstrates a technically effective method for creating efficient light-converting coatings. The primary benefit to the public lies in the potential for these coatings to enhance the efficiency of solar-driven processes, such as boosting the growth of algae in photobioreactors for biofuel production.

14 SOLAR ENERGY↗
DOE OSTI2020

Adaptive simulations enable computational design of electron beam processing of nanomaterials with supersonic micro-jet precursor

Focused Electron Beam Induced Processing (FEBIP) is a powerful tool for the “direct-write” of nanomaterials with the possibility of atomistic control on suspended 2D material substrates. FEBIP capabilities have been significantly expanded by using a localized jet-based delivery of precursors, and especially when a thermally energized supersonic micro-jet enhances the delivery of mass flux along with controlling the far-from-equilibrium thermodynamic state of adsorbed adatoms. The possibilities of growing nanomaterials with “dialed-in” composition with ultra-high growth rates and aspect ratios up to 100:1 have been demonstrated using the supersonic micro-jet-FEBIP. Bringing this scientific discovery to the level of maturity required for practical applications in additive nanomanufacturing requires simulation tools that are capable of capturing the complex flow physics of micro-jet-substrate interactions that bridge a wide range of flow regimes from the high-density gas micro-jet expanding into the vacuum environment of FEBIP. To address this significant computational challenge, a new approach has been developed and described in this work for a multiscale adaptive DSMC (Direct Simulation Monte Carlo) algorithm to predict the micro-jet gas dynamics in an FEBIP environment, spanning the full range of flow regimes from low Knudsen (Kn) number O(0.01) continuum flow to high Kn of O(10) for the molecular flow within a unified computational framework. Here, the fundamental principles of the adaptive DSMC algorithm are described, its viability as a DSMC technique is demonstrated, and computational improvements for the benchmark cases are discussed in the context of advancing 3D nanofabrication with FEBIP. Ultimately, combining the first principle simulations via adaptive DSMC with the complementary experimental data will enable the creation of the powerful CAD tools for in silico design and optimal operation of micro-jet-FEBIP.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Cooperative Research and Development Agreement With Georgetown University Report: National Institutes of Health, National Center for Advancing Translational Sciences Clinical and Translational Science Award

Oak Ridge National Laboratory (ORNL) is participating with Georgetown University (GU) as a subrecipient in response to the National Institutes of Health (NIH), National Center for Advancing Translational Sciences (NCATS) Clinical and Translational Science Award (CTSA) (U54 Clinical Trial Optional) funding opportunity announcement. This Cooperative Research and Development Agreement is put in to place to facilitate the development and implementation of clinical interventions that demonstrably improve human health is currently a complex, recursive, and inefficient process that leads to delays of years or decades before discoveries in biomedical research result in health benefits for patients and communities. NCATS conducts and supports research in the science of translation, to discover the mechanistic and operational principles of the intervention development and dissemination process, thereby providing the scientific foundation for improvements in translational efficiency that will accelerate the realization of interventions that improve human health. Under NCATS’ leadership, the CTSA Program supports a national network of medical research institutions called hubs. GU is the lead institution in one of the NIH hubs that was created as a result of a previous NIH CTSA. The missions of the GU have historically included the advancement of health through research in the clinical and biomedical sciences, the education of future leaders in medical and nursing practice and academia, and the provision of compassionate and scientifically competent patient care and service to the Washington, DC community and the nation. GU is the lead institution for the Georgetown-Howard Universities Center for Clinical and Translational Science (GHUCCTS), a multi-institutional partnership of medical research institutions forged from a desire to promote clinical research and translational science. Through multiple collaborations among these institutions, GHUCCTS is transforming clinical research and translational science in order to bring new scientific advances to health care. Oak Ridge National Laboratory is the Department of Energy's (DOE) largest science and energy laboratory. Managed since April 2000 by a partnership of the University of Tennessee and Battelle, ORNL was established in 1943 as a part of the secret Manhattan Project to pioneer a method for producing and separating plutonium. During the 1950s and 1960s, ORNL became an international center for the study of nuclear energy and related research in the physical and life sciences. With the creation of DOE in the 1970s, ORNL's mission broadened to include a variety of energy technologies and strategies. Today the laboratory supports the nation with a peacetime science and technology mission that is just as important as, but very different from, its role during the Manhattan Project. ORNL is home to the world's premier center for high performance supercomputing to enable scientific discovery. ORNL has extensive expertise in various areas of computer science that are uniquely situated to support GU. Additionally, ORNL’s leading computational user facilities present a unique opportunity to leverage the largest scale machines for open science in support of the stated mission of the NCATS CTSA. ORNL's partnership with GU will offer unparalleled opportunity in data analytics, deep-learning, artificial intelligence, and urban dynamics.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2017

Heat deposition analysis for the High Flux Isotope Reactor’s HEU and LEU core models

The High Flux Isotope Reactor at Oak Ridge National Laboratory is an 85 MW th pressurized light-water-cooled and -moderated flux-trap type research reactor. The reactor is used to conduct numerous experiments, advancing various scientific and engineering disciplines. As part of an ongoing program sponsored by the US Department of Energy National Nuclear Security Administration Office of Material Management and Minimization, studies are being performed to assess the feasibility of converting the reactor’s highly enriched uranium fuel to low-enriched uranium fuel. To support this conversion project, reference models with representative experiment target loading and explicit fuel plate representation were developed and benchmarked for both fuels to (1) allow for consistent comparison between designs for both fuel types and (2) assess the potential impact of low-enriched uranium conversion. These high-fidelity models were used to conduct heat deposition analyses at the beginning and end of the reactor cycle and are presented herein. This article (1) discusses the High Flux Isotope Reactor models developed to facilitate detailed heat deposition analyses of the reactor’s highly enriched and low-enriched uranium cores, (2) examines the computational approach for performing heat deposition analysis, which includes a discussion on the methodology for calculating the amount of energy released per fission, heating rates, power and volumetric heating rates, and (3) provides results calculated throughout various regions of the highly enriched and low-enriched uranium core at the beginning and end of the reactor cycle. These are the first detailed high-fidelity heat deposition analyses for the High Flux Isotope Reactor’s highly enriched and low-enriched core models with explicit fuel plate representation. Lastly, these analyses are used to compare heat distributions obtained for both fuel designs at the beginning and end of the reactor cycle, and they are essential for enabling comprehensive thermal hydraulics and safety analyses that require detailed estimates of the heat source within all of the reactor’s fuel element regions.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2025

Scalable training of trustworthy and energy-efficient predictive graph foundation models for atomistic materials modeling: a case study with HydraGNN

We present our work on developing and training scalable, trustworthy, and energy-efficient predictive graph foundation models (GFMs) using HydraGNN, a multi-headed graph convolutional neural network architecture. HydraGNN expands the boundaries of graph neural network (GNN) computations in both training scale and data diversity. It abstracts over message passing algorithms, allowing both reproduction of and comparison across algorithmic innovations that define nearest-neighbor convolution in GNNs. This work discusses a series of optimizations that have allowed scaling up the GFMs training to tens of thousands of GPUs on datasets consisting of hundreds of millions of graphs. Our GFMs use multitask learning (MTL) to simultaneously learn graph-level and node-level properties of atomistic structures, such as energy and atomic forces. Using over 154 million atomistic structures for training, we illustrate the performance of our approach along with the lessons learned on two state-of-the-art US Department of Energy (US-DOE) supercomputers, namely the Perlmutter petascale system at the National Energy Research Scientific Computing Center and the Frontier exascale system at Oak Ridge Leadership Computing Facility. The HydraGNN architecture enables the GFM to achieve near-linear strong scaling performance using more than 2000 GPUs on Perlmutter and 16,000 GPUs on Frontier.

97 MATHEMATICS AND COMPUTING↗