SEARCH · Engineering Papers

Results for “Local time stepping”

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 271 records · Page 15

All sources NASA NTRS DOE OSTI

DOE OSTI2023

FPGA Acceleration of GCN in Light of the Symmetry of Graph Adjacency Matrix

Graph Convolutional Neural Networks (GCNs) are widely used to process large-scale graph data. Different from deep neural networks (DNNs), GCNs are sparse, irregular, and unstructured, posing unique challenges to hardware acceleration with regular processing elements (PEs). In particular, the adjacency matrix of a GCN is extremely sparse, leading to frequent but irregular memory access, low spatial/temporal data locality and poor data reuse. Furthermore, a realistic graph usually consists of unstructured data (e.g., unbalanced distributions), creating significantly different processing times and imbalanced workload for each node in GCN acceleration. To overcome these challenges, we propose an end-to-end hardware-software co-design to accelerate GCNs on resource-constrained FPGAs with the features including: (1) A custom dataflow that leverages symmetry along the diagonal of the adjacency matrix to accelerate feature aggregation for undirected graphs. We utilize either the upper or the lower triangular matrix of the adjacency matrix to perform aggregation in GCN to improve data reuse. (2) Unified compute cores for both aggregation and transform phases, with full support to the symmetry-based dataflow. These cores can be dynamically reconfigured to the systolic mode for transformation or as individual accumulators for aggregation in GCN processing. (3) Preprocessing of the graph in software to rearrange the edges and features to match the custom dataflow. This step improves the regularity in memory access and data reuse in the aggregation phase. Moreover, we quantize the GCN precision from FP32 to INT8 to reduce the memory footprint without losing the inference accuracy. We implement our accelerator design in Intel Stratix10 MX FPGA board with HBM2, and demonstrate 1.3x-110.5x improvement in end-to-end GCN latency as compared to the state-of the-art FPGA implementations, on the graph datasets of Cora, Pubmed, Citeseer and Reddit.

Nair, Gopikrishnan R.↗

DOE OSTI2025

A Computational Framework to design 3D stiffness gradient acoustic metamaterials for impedance matching

Acoustic waves play a crucial role in various applications, including medical imaging, non-destructive testing, and sonar systems. One of the significant challenges in these applications is impedance matching, which is essential for minimizing reflections and maximizing the transfer of acoustic energy between different media. Acoustic metamaterials offer a promising solution to this challenge. In addition to impedance control, gradient stiffness can enhance structural efficiency and enable spatial control of wave propagation, making it a valuable feature in acoustic metamaterial design. In this pa- per, we present our developed computational method to design 3D stiffness gradient acoustic metamaterials for impedance matching. The key steps in our approach include generating initial designs using a periodic covariance function to provide unit cells that are both periodic on the boundaries and randomly formed inside the unit cell. Furthermore, we integrated manufacturing constraints into the design process, ensuring that the structures are interconnected for fabrication. We propose two computational optimization algorithms: GenUnit, based on a non-dominated sorting genetic algorithm (NSGA-II), and MLMatch, which leverages differentiable machine learning. The two approaches are not separate contributions but complementary com- ponents of a unified framework. GenUnit requires no training data and directly interfaces with physics-based simulations, making it highly accurate but slower for large-scale exploration. In contrast, MLMatch is data-hungry during training but, once trained, enables near-instantaneous inference and broad design-space coverage. Together, they form a hybrid strategy: ML- Match rapidly explores the global design space, and GenUnit provides local refinement with high-fidelity accuracy. This balance between training cost, inference time, and precision is the motivation for including both methods in the same study. We applied this dual-algorithm framework to generate two metallic-based metamaterial designs that match the acoustic impedance of water while exhibiting a controlled gradient in stiffness (from stiff to soft). The stiffness gradient is particularly advantageous in applications where one side of the structure must interface with soft or sensitive surfaces, such as human tissue or delicate components. Here, this work paves the way for improved materials in various acoustic applications, particularly in ultrasound devices, by providing better impedance.

Metamaterial↗

DOE OSTI2020

A method for mapping submicron-scale crystallographic order/disorder applied to human tooth enamel

Tooth enamel, the outermost layer of human teeth, is a complex, hierarchically structured biocomposite. The details of this structure are important in multiple human health contexts, from understanding the progression of dental caries (tooth decay) to understanding the process of amelogenesis and related developmental defects. Enamel is composed primarily of long, nanoscale crystallites of hydroxyapatite that are bundled by the thousands to form micron-scale rods. Studies with transmission electron microscopy show the relationships between small groups of crystallites and X-ray diffraction characterize averages over many rods, but the direct measurement of variations in local crystallographic structure across and between enamel rods has been missing. Here, we describe a synchrotron X-ray-based experimental approach and a novel analysis method developed to address this gap in knowledge. A ~500-nm-wide beam of monochromatic X-rays in conjunction with a sample section only 1 μm in thickness enables 2D diffraction patterns to be collected from small well-separated volumes within the enamel microstructure but still probes enough crystallites (~300 per pattern) to extract population-level statistics on crystallographic features like lattice parameter, crystallite size, and orientation distributions. Furthermore, the development of a quantitative metric to characterize relative order and disorder based on the azimuthal autocorrelation of diffracted intensity enables these crystallographic measurements to be correlated with their location within the enamel microstructure (e.g., between rod and interrod regions). These methods represent a step forward in the characterization of human enamel and will elucidate the variation of the crystallographic structure across and between enamel rods for the first time.

crystallographic order/disorder↗

DOE OSTI2026

Substrate-Directed Underlayer Growth of Bilayer MoS 2 Revealed by Mo Isotope Labeling

Direct control over the vertical formation sequence and stacking registry in van der Waals (vdW) bilayers is essential for device performance and moiré engineering yet difficult to resolve unambiguously with conventional probes. Here, we use Mo isotope labeling in a two-step chemical vapor deposition process to synthesize bilayer MoS 2 and trace its vertical formation on common substrates. By combining site-selective laser thinning, Raman spectroscopy, time-of-flight secondary ion mass spectrometry, and atomic-resolution scanning transimission electron microscopy (STEM), we find a clear substrate dependence: on SiO 2 /Si, the second layer nucleates and grows beneath the first (underlayer), whereas on sapphire, it forms on top (overlayer). Density functional theory indicates that a larger equilibrium interfacial separation and weaker MoS 2 –substrate interactions on amorphous SiO 2 permit confined interfacial diffusion and underlayer nucleation, whereas stronger interactions and smaller separations on sapphire favor overlayer growth. On SiO 2 , confined epitaxy templates commensurate 2H, 3R, and mixed bilayers, as confirmed by second harmonic generation spectroscopy and STEM. During underlayer coalescence, embedded mirror-twin grain boundaries stitch atomically sharp 2H|3R junctions via alternating 4|8 ring motifs. Molecular-dynamics simulations reveal that these alternating 4|8 motifs accommodate interlayer vdW coupling and locally modulate the stacking registry. These results provide mechanistic insight into confined epitaxial growth and establish isotope labeling as a powerful probe of two-dimensional materials synthesis.

MoS2↗

DOE OSTI2025

Magnetically and optically active edges in phosphorene nanoribbons

Abstract Nanoribbons, nanometre-wide strips of a two-dimensional material, are a unique system in condensed matter. They combine the exotic electronic structures of low-dimensional materials with an enhanced number of exposed edges, where phenomena including ultralong spin coherence times 1,2 , quantum confinement 3 and topologically protected states 4,5 can emerge. An exciting prospect for this material concept is the potential for both a tunable semiconducting electronic structure and magnetism along the nanoribbon edge, a key property for spin-based electronics such as (low-energy) non-volatile transistors 6 . Here we report the magnetic and semiconducting properties of phosphorene nanoribbons (PNRs). We demonstrate that at room temperature, films of PNRs show macroscopic magnetic properties arising from their edge, with internal fields of roughly 240 to 850 mT. In solution, a giant magnetic anisotropy enables the alignment of PNRs at sub-1-T fields. By leveraging this alignment effect, we discover that on photoexcitation, energy is rapidly funnelled to a state that is localized to the magnetic edge and coupled to a symmetry-forbidden edge phonon mode. Our results establish PNRs as a fascinating system for studying the interplay between magnetism and semiconducting ground states at room temperature and provide a stepping-stone towards using low-dimensional nanomaterials in quantum electronics.

Science & Technology - Other Topics↗

DOE OSTI2025

Macro-level mechanical interlocking: A rapid joining approach for additively manufactured compression molded composite panels

Composite joining typically involves multiple steps, such as drilling and surface treatment, as part of the manufacturing process, which leads to low throughput and long cycle times. In the present study, we demonstrated a macro-level mechanical interlocking (MI) based, rapid joining technique to assemble additively manufactured compression molded (AMCM) panels, enabling the production of parts larger than the mold dimensions. Composite panels made of 20 wt% short carbon fiber reinforced acrylonitrile butadiene styrene (CF/ABS) were joined using MI features of various geometries, namely tree (TR), dovetail (Dov), rectangle 2 (Rect2), and rectangle 1 (Rect1), and their in-plane strength was evaluated. The resultant strength of the tested MI joints reached up to 74 % of the baseline tensile strength (i.e., the ‘no joint’ case). Observations from optical and scanning electron microscopy revealed inadequate polymer diffusion between the adherends, indicating that the joint strength was primarily derived from mechanical interlocking. Additionally, the fracture surfaces exhibited stress-whitening marks, which were characterized using differential scanning calorimetry (DSC). The increase in melting enthalpy suggested local stretching of polymer chains due to MI. Finite element analysis (FEA) indicated that the Rect1 MI feature, which generated the lowest stress concentration, outperformed the others in terms of joint strength, achieving 42 MPa. As a demonstration of the MI joining method, a battery box tray measuring 108 cm × 34 cm using a mold with an effective dimension of 36 cm × 34 cm successfully manufactured, resulting in a part with an area three times larger than the mold. In conclusion, this study presents a promising approach to improving composite joining techniques while minimizing production complexities.

In-plane joining↗

DOE OSTI2024

One-Step Synthesis and Operando Electrochemical Impedance Spectroscopic Characterization of Heterostructured MoP–Mo 2 N Electrocatalysts for Stable Hydrogen Evolution Reaction

Here this study presents a novel synthesis of self-standing MoP and Mo 2 N heterostructured electrocatalysts with enhanced stability and catalytic performance. Facilitated by the controlled phase and interfacial microstructure, the seamless structures of these catalysts minimize internal resistivity and prevent local corrosion, contributing to increased stability. The chemical synthesis proceeds with etching step to activate the surface, followed by phosphor-nitriding in a chemical vapor deposition chamber to produce MoP-Mo 2 N@Mo heterostructured electrocatalysts. X-ray diffraction analyses confirm the presence of MoP, Mo 2 N, and Mo phases in the electrocatalyst. Morphology studies using scanning electron microscopy characterizes the hierarchical growth of structures, indicating successful formation of the heterostructure. X-ray photoelectron spectroscopy (XPS) analyses of the as-synthesized and post-catalytic activity samples reveal the chemical shift in terms of binding energy (BE) of Mo 3d XPS peak, especially after catalytic activity. The XPS BE shifts attributed to changes in oxidation state, electron transfer, and surface reconstruction during catalysis. Electrochemical evaluation of the catalysts demonstrates the superior performance of the MoP-Mo 2 N@Mo heterostructured catalyst in hydrogen evolution reactions (HER), with lower overpotentials and enhanced Tafel slopes. Stability tests reveal changes in double layer capacitance over time, suggesting surface reconstruction and increased active surface area during catalysis. Operando electrochemical impedance spectroscopy (EIS) further elucidates the dynamic changes in resistance and charge transfer during HER. Overall, comprehensive understanding of the synthesis, characterization, and electrochemical behavior of the developed MoP-Mo 2 N@Mo heterostructured electrocatalyst as presented in this paper highlights their potential utilization in sustainable energy applications.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2024

Integrative Multi-PTM Proteomics Reveals Dynamic Global, Redox, Phosphorylation, and Acetylation Regulation in Cytokine-treated Pancreatic Beta Cells

Studying regulation of protein function at a systems level necessitates an understanding of the interplay among diverse post-translational modifications (PTMs). A variety of proteomics sample processing workflows are currently used to study specific PTMs but rarely characterize multiple types of PTMs from the same sample inputs. Method incompatibilities and laborious sample preparation steps complicate large-scale physiological investigations and can lead to variations in results. The single-pot, solid-phase-enhanced sample preparation (SP3) method for sample cleanup is compatible with different lysis buffers and amenable to automation, making it attractive for high-throughput multi-PTM profiling. Herein, we describe an integrative SP3 workflow for multiplexed quantification of protein abundance, cysteine thiol oxidation, phosphorylation, and acetylation. The broad applicability of this approach is demonstrated using cell and tissue samples, and its utility for studying interacting regulatory networks is highlighted in a time-course experiment of cytokine-treated ß-cells. We observed a swift response in global regulation of protein abundances consistent with rapid activation of JAK-STAT and NF-?B signaling pathways. Regulators of these pathways as well as proteins involved in their target processes displayed multi-PTM dynamics indicative of a complex cellular response stages: acute, adaptation, and chronic (prolonged stress). PARP14, a negative regulator of JAK-STAT, had multiple co-localized PTMs that may be involved in intraprotein regulatory crosstalk. Our workflow provides a high-throughput platform that can profile multi-PTMomes from the same sample set, which is valuable in unraveling the functional roles of PTMs and their co-regulation.

proteomics, PTM, automation, SP3, cysteine thiol o↗

DOE OSTI2022

High-Temperature 2D Optical Relaxation Visualizes Enhanced Oxygen Exchange Kinetics at Metal-Mixed Conducting Oxide Interfaces

Solid-state heterointerfaces are of interest for emergent local behavior that is distinct from either bulk parent compound. One technologically relevant example is the case of mixed ionic/electronic conductor (MIEC)–metal interfaces, which play an important role in electrochemistry. Metal–MIEC composite electrodes can demonstrate improved catalytic activity vs single-phase MIECs, improving fuel cell efficiency. Similarly, MIEC surface reaction kinetics are often evaluated using techniques that place metal current collectors in contact with the surface under evaluation, potentially altering the response vs the native surface. Here, techniques enabling direct and local in situ observation of the behavior at and around such heterointerfaces are needed. Here, we develop a spatially resolved optical transmission relaxation (2D-OTR) method providing continuous evaluation of local, high-temperature, controlled atmosphere defect kinetics across a ~1 cm2 sample area simultaneously in a contact-free manner. We apply it to observe the spatial variance of oxygen incorporation and evolution rates at ~525–620 °C, in response to step changes in oxygen partial pressure, on MIEC SrTi 0.65 Fe 0.35 O 3–x films as a function of distance from porous Pt and Au layers. Using this model geometry, we find significant enhancements in kinetics adjacent to the metals that decay over a few millimeter distance. To extract kinetic parameters, we fit the short-term optical data (initial portion of relaxations) with an exponential decay function appropriate for surface-exchange-limited kinetics, yielding apparent surface exchange coefficients (k chem ) with spatial resolution, decreasing with distance from the metal. To understand the kinetic processes governing the complete (long-term) optical relaxations, we performed COMSOL simulations, which demonstrated that a combination of laterally varying k chem and in-plane diffusion controls the observed kinetics over the full time range. Further support for spatially varying kchem comes from demonstrations of changing surface and bulk chemistry vs distance from the metal–MIEC interface, by X-ray photoelectron and optical absorption spectroscopies, respectively. Although microporous Pt and Au are not excellent electrodes in isolation, both metals exert a synergistic effect on the oxygen surface exchange rate in the presence of the mixed conducting film.

36 MATERIALS SCIENCE↗

DOE OSTI2024

Impact of the Exciter and Governor Parameters on Forced Oscillations

In recent years, the frequency of forced oscillation events due to control system malfunctions or improper parameter settings has increased. Tuning the parameters of exciters and governor models is crucial for maintaining power system stability. Traditional simulation studies typically involve small transient disturbances or step changes to find optimal parameter sets, but existing optimization algorithms often fall short in fine-tuning for forced oscillations. Identifying the sensitive parameters within these control models is essential for ensuring stability during large, sustained disturbances. This study focuses on identifying these critical exciter and governor model parameters by analyzing their influence on sustained forced oscillations. Using Kundur’s two-area system, we analyze common exciter models such as SCRX, ESST1A, and AC7B, along with governor models like GAST, HYGOV, and GGOV1, utilizing PSS®E software version 34. Sustained forced oscillations are injected at generator-1 of area-1, with individual parameter changes dynamically simulated. By considering a local oscillation frequency of 1.4 Hz and an inter-area oscillation mode of 0.25 Hz, we analyze the impact of each parameter change on the magnitude and frequency of forced oscillations as well as on active and reactive power outputs. This novel approach highlights the most influential parameters of each tested model—such as exciter, governor, and turbine gains, as well as time constant parameters—on the impact of forced oscillations. Based on our findings, the sensitive parameters of each tested model are ranked. These would provide valuable insights for industry operators to fine-tune control settings during oscillation events, ultimately enhancing system stability.

42 ENGINEERING↗

DOE OSTI2023

Development and implementation of an in situ high-resolution isotopic microscope for measuring metabolic interactions in soil mesocosms (Final Report)

Many of our planet’s ecosystems rely on the activities of soil microbial communities. These microbes have significant and wide-ranging effects: they metabolize carbon and other nutrients, interact with plants and fungi, and perform other processes important to soil health. However, our ability to directly observe the enzymatic and metabolic activities of microbes within soil is currently limited, not only by the complexity of soil microbial communities themselves, but also by the lack of experimental tools to study them and their molecular interactions in situ. These challenges hinder our understanding of the life-sustaining processes of biomass decomposition and the manner in which it contributes to the movement of freed carbon within soil ecosystems. Here we aimed to develop a novel ultrahigh-resolution isotopic microscope that combines complementary imaging modalities to gain insights into metabolic cycling in soil. This final report covers both the portion of this work that was initially completed at the University of North Carolina at Chapel Hill from 2018 to 2019 (as award DE-SC0019012) and then from 2020 to 2022 (2023 in NCE) at the University of Massachusetts Chan Medical School. This goal was to build an integrated platform consisting of fluorescence microscopy, Raman microspectroscopy, and nanospray desorption electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry (nanoDESI-FTICR-MS) to directly investigate microbial activities and molecular transformations occurring in soil by exploiting the use of both fluorescent labels and stable isotope probing. Specifically, our instrument was conceived as integrating: (A) fluorescence detection to localize soil microbes, identify bacteria taking up polysaccharides, and monitor gene expression of enzymes involved in decomposition; (B) Raman microspectroscopy to determine which microbes incorporate decomposition products into their biomass, as well as which decomposers are metabolically active; and (C) nanoDESI-FTICR-MS imaging to spatially probe, in real-time, the metabolites in the surrounding area, which we expect will reveal the distributions of the products of the enzymatic breakdown of polysaccharides, as well as specialized metabolites acting as cell-cell signals between decomposers. To enable these measurements, we exploited fluorescence-based probes to map the microbes that are enzymatically active (per A) and utilized stable isotope-labeled substrates to visualize both the microbial and molecular fate of decomposed biomass (B and C, respectively). To accomplish our goal, we combined the expertise of a multidisciplinary group of scientists to pursue the construction of this microscope and to investigate scientific questions that would be facilitated by such a capability. We aimed to employ this spatially informative, high-resolution isotopic microscope to visualize the critical steps of biomass degradation and the molecular fate of other environmentally relevant substrates within soil mesocosms. Accordingly, this technology will enhance our understanding of the microbial and metabolic interactions occurring within soil communities that are relevant to carbon degradation and other soil processes. The instrument has been developed and housed at EMSL, where it will be available to the entire EMSL User Base, enabling a variety of related DOE-relevant systems to be interrogated in the future by diverse scientific research groups.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2020

Solar Resource and Infrastructure Assessment for the Town of Wendell

This report is a solar resource and infrastructure assessment for the town of Wendell, Massachusetts. The assessment was funded through the National Renewable Energy Laboratory, Solar Energy Innovation Network (NREL SEIN) Solar in Rural Communities Program, as part of a project to develop a Community-Informed Proactive Solar Siting and Financing Model. As a first step, the project lead organization, UMass Clean Energy Extension prepared an assessment of existing infrastructure, resources, and potential solar development opportunities in participating municipalities, including Wendell. This assessment was designed to describe relevant bylaws and infrastructure within the town, identify the types of solar facilities that could be developed, and quantify the total space available for each type of facility. In this report, we reviewed existing electricity grid infrastructure, and the potential to interconnect additional solar facilities. At the present time, both distribution lines providing electricity to Wendell are over-saturated with authorized and proposed solar projects, and cannot accommodate additional solar projects to interconnect to the grid. It appears that National Grid is planning an upgrade to the Wendell Depot substation, which might then allow in-process projects are to proceed, and potentially free up additional capacity for new, large projects. Meanwhile, most three-phase lines could likely accommodate additional small-to-medium scale projects (under 200 kW), and most single-phase lines could likely accommodate additional projects under 50 kW in size. This description represents the local grid infrastructure as it is - planning for future scenarios of development could include recommendations for areas of grid infrastructure improvement to allow siting of distributed generation in locations preferred by the community. Future scenarios may also include the addition energy storage and other "non-wires alternatives."

14 SOLAR ENERGY↗

DOE OSTI2025

Machine Learning for Anomaly Detection in Neural Network Security and SRF Cavities

This dissertation explores the development and deployment of machine learning approaches to address critical challenges in anomaly detection across two distinct domains: neural network security in federated learning settings and cavity behavior analysis in particle accelerator operations at Jefferson Lab in Newport News, Virginia. Anomaly detection identifies deviations from expected patterns, safeguarding systems in cybersecurity, industry, and research against malicious activities and failures. This dissertation demonstrates how our machine learning approaches enhance detection accuracy and efficiency in both neural network security and industrial applications. First, we investigate vulnerabilities in deep neural networks deployed in federated learning. Although federated learning preserves user privacy by training models locally, it remains vulnerable to backdoor attacks, in which malicious participants embed hidden triggers that induce targeted misbehavior. We propose a self-supervised contrastive learning framework to detect and mitigate such backdoor attacks. In our experiments, this method achieves higher detection accuracy and lower false positive rates than existing defenses, while operating without access to local model updates or original training data and thus preserving the privacy guarantees of the federated setting. Second, we address the operational reliability of superconducting radio-frequency (SRF) cavities at the Continuous Electron Beam Accelerator Facility (CEBAF). Our research leverages an unsupervised learning approach, combined with Principal Component Analysis (PCA) and k-means clustering, to identify anomalous behaviors in SRF cavities. Our method detects subtle anomalous behavior by analyzing SRF signal data. This knowledge allows for the early detection and resolution of potential faults, significantly improving the efficiency and reliability of operations. Third, we extend these insights to time-series anomaly detection more broadly. We design a contrastive-learning based model tailored to increasingly dynamic environments and academic research. This model improves detection accuracy in settings that require real-time monitoring and predictive maintenance. Our research underscores the broader applicability and impact of advanced machine learning techniques in anomaly detection. By extracting meaningful patterns from complex data, machine learning can significantly enhance security in distributed neural networks and improve the efficiency of particle accelerator operations. This dissertation serves as a stepping stone for future investigations into the vast possibilities of anomaly detection, inspiring further exploration and development of machine learning techniques in this field.

Ferguson, Hal [Old Dominion University]↗

DOE OSTI2025

On the numerical sensitivity of cellular automata grain structure predictions to large thermal gradients and cooling rates

Cellular automata (CA) models of as-solidified grain structure, originally developed and applied to casting, have become a common means of predicting grain structure resulting from Additive Manufacturing (AM) processes. The majority of these models are based on the decentered octahedron approach, which attempts to correct for the effect of grid anisotropy on the prediction of competitive solidification of dendritic grains. However, AM solidification occurs under cooling rates ($\dot{T}$) and thermal gradients (G) that are orders of magnitude larger than those encountered in casting, and no systematic investigation on the effect of the CA model cell size (Δx) and time step (Δt) on AM microstructure predictions has been performed. Here, in this study, such an investigation is first performed via simulation of individual grains of various crystallographic orientations with a fixed, unidirectional G, showing that CA prediction of the steady-state undercooling matched the expected values based on the interfacial response function at small G and deviated from the expected values at large G. Simulation of competitive growth of multiple grains showed a weakening of the predicted texture as G and Δx became large. Simulation of solidification under AM conditions, where G and $\dot{T}$ vary spatially across the melt pools, showed that not only does grain selection weaken and deviate from expectations at large Δx, but grains with crystallographic $\langle$100$\rangle$ aligned with the grid directions are more adversely affected by the temperature field discontinuities than grains with other crystallographic orientations. Despite the fact that the exact grain competition results depended on Δt, the overall texture development was notably less sensitive to Δt than Δx, provided that a reasonable value of Δt is selected based on the ratio of Δx to the maximum local solidification velocity in the simulation domain. Finally, from the directional solidification and AM simulation results, an analysis of computational cost compared to simulation resolution is performed based on an equation derived to quantify the relatively inaccuracy in grain selection based on the model and temperature field inputs. From this analysis, it is concluded that there is a need for algorithmic improvements to improve CA grain competition accuracy for large G processing conditions as sufficiently small Δx to resolve the necessary competition is intractable for many AM processing conditions.

36 MATERIALS SCIENCE↗

DOE OSTI2025

HPC Campaign Management: Remote data access with user-defined error bound using ADIOS and ZFP

Remote access to large-scale scientific datasets, like those generated by combustion simulations or other high-performance computing (HPC) applications, presents a significant challenge. Downloading entire datasets is often impractical due to their size and the bandwidth limitations of typical networks. To address this challenge, we propose a novel approach that enables efficient remote access to large datasets distributed across multiple facilities. Our method enables technologies to download only the data values of a select variable, in a select region of interest, to a user-defined accuracy. For this purpose, we extended the ADIOS IO library to provide read functions with user-defined accuracy, a remote data server that understands multidimensional selections of specific variables, steps and accuracy from an ADIOS dataset, and which uses lossy compression on the remote site to reduce the data to be transferred back to the client. In addition, our extension of the ADIOS library collects metadata from multiple datasets in small files called Campaign Archives, which can be shared among project participants on any HPC, cloud or laptop, and which can easily facilitate the discovery of content and pointers to the data location as well as remote access to the data by local tools as if data was local. This feature called Campaign Management, enables a group of scientists to manage related datasets stored in multiple files, across multiple facilities as if it was in a single file/database. We demonstrate the effectiveness of our approach using a 1.5 TB dataset from the S3D combustion simulation on Frontier at the Oak Ridge Leadership Facility. Even a single variable from this dataset, at 64 GB, is too large to be processed on a standard laptop. We show two different reading patterns for 2D plots and 3D visualization, with careful settings that a scientist studying combustion data would do and show that running the same Python scripts on Frontier directly takes comparable time than running them on the local laptop with remote access to the data on Frontier.

Podhorszki, Norbert [ORNL] (ORCID:000000019647542X↗

DOE OSTI2026

Ultrafast Formation of Jahn–Teller Polarons Revealed by State-Selective Excitation in Correlated Spinel Co 3 O 4

Jahn–Teller polarons are quasiparticles that stem from symmetry breaking and strong local electron–phonon coupling. They originate from an excess charge carrier being dressed by a local lattice distortion, caused by the Jahn–Teller effect, and they critically impact electrical, structural, and magnetic properties in transition metal oxides. The observation of the microscopic steps involved in their formation is essential for enabling control over material properties through the targeted activation of local, site-specific modifications with light pulses. While Jahn–Teller distortion associated with polaron formation was predicted to contribute significantly to changes in electronic band gap and optical properties in Co 3 O 4 , its experimental observation remains elusive, requiring signatures of local symmetry reduction. In this work, we demonstrate Jahn–Teller polaron formation in spinel Co 3 O 4 . By exciting electronic transitions at 3.10 eV and 1.55 eV, we target either the Oh Cobalt(III) or the Td Cobalt(II) ions, and drive the subsequent coherent responses of the system through two different pathways. For the former, we demonstrate that ligand-to-metal charge transfer leads to Jahn–Teller polaron formation, which is linked to the deformation potential and magnetoelastic coupling. For the latter, we identify the coherent excitation of a T 2g phonon mode launched by on-site d-d electronic transitions. Key to our observations is the ability to target site-specific electronic excitations in spinel Co 3 O 4 using ultrafast optical pulses, while monitoring the ensuing low-energy collective modes through the coherent time-domain response of the material. Our approach, which combines the comparative analysis of experimental fingerprints with the support from density functional theory calculations, is broadly applicable to systems in which Jahn–Teller polaron physics has been theoretically predicted but remains experimentally unverified, and it underscores the potential of electronic-state targeting as a route to selectively excite and probe quasiparticle dynamics in solids.

Lattices↗

DOE OSTI2023

Statistical Learning for Nonlinear Model Reduction from Local Simulations of Stochastic and Particle- and Agent-Based Systems

Stochastic physical systems across the sciences that have very high-dimensional state spaces, with a large number of fast degrees of freedom that force direct simulators to proceed by integration steps that are orders of magnitude smaller than events of interests (e.g., particle collisions). Examples range from molecular motion to dynamics of large populations of cells. A grand challenge in the simulation and understanding of such systems is the systematic construction of accurate, interpretable, reduced models, enabling faster simulations, revealing fundamental properties of the dynamics, and predicting phenomena of interest that the original simulator could not reached with sufficient accuracy or within a given computational budget. In this projected we developed novel statistical estimation/machine learning techniques for analyzing and building empirical reduced models for important families of high-dimensional stochastic systems, in particular: - we developed techniques for estimating interaction kernels in interacting particle- and agent-based systems, which are ubiquitous in Physics, Biology and many other sciences, given observed trajectories of the system; - we developed techniques for nonlinear model reduction for high-dimensional stochastic systems that have a small number of unknown, nonlinear slow variables, and a large number of fast modes, that are possibly of large magnitude, given observed short trajectories of the system in the form of bursts of trajectories from different initial conditions; - we developed novel techniques for estimating linear dynamical systems on graphs when both the dynamics and the underlying graph are unknown, and we have a sparse set of space-time observations; - we considered the problem of estimating an unknown nonlinear observation function of a standard process (e.g. Brownian motion), so that we can recognized if an observed dynamics is "just" a nonlinear version of a known dynamics; we also developed benchmarks for learning algorithms aimed at learning and classifying diffusion processes.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

Adaptive Cybersecurity for Distributed Energy Resources (AdCyDER): Online Reinforcement Learning with Stackelberg-Optimized Defenses — Pipeline Architecture, Evaluation Methodology, and Findings from a Synthetic-Data Evaluation

This report documents the design and evaluation of an integrated online-learning pipeline developed within the AdCyDER project for Distributed Energy Resource (DER) cybersecurity. The pipeline couples a Reinforcement Learning (RL) attack classifier — which produces an attack-type probability distribution — with a Stackelberg game-theoretic (GT) defense selector that consumes those distributions alongside SME-encoded priors over (defense, attack) effectiveness pairings and perdefense costs to choose grid-health-preserving defenses. The objective is not attack classification per se but production of distributions that drive effective defense selection through the Stackelberg layer, learned from delayed grid-health feedback rather than labeled attack data. AdCyDER as a whole is broader than the work presented here; this report covers the specific RL/GT loop integration and its evaluation. We present the integrated pipeline (SCADA telemetry with Fronius inverter physics, Suricata IDS, time-windowed aggregation, per-facility LSTM classifier, Stackelberg optimizer, OpenC2 actuators), an experimental campaign of 28 eight-hour iterations across three baseline modes, and a pipeline-ordered diagnostic protocol. The protocol identifies two distinct failure modes within the loop: paired supervised ceilings on the same features establish that the deployed online RL classifier (macro F1 ≈ 0.07) sits at least 4.7× below a same-architecture supervised LSTM (≈ 0.34) and 10–11× below a linear feature-signal ceiling (≈ 0.70–0.79 depending on per-facility isolation), localizing the dominant failure to the training procedure; and the reward signal driving online updates carries weak directional coupling with classifier correctness in the methodology-expected direction (multi-lens convergent: top-decile P(true) records produce more frequent state changes and slightly larger improvements, top-vs-bot Cohen’s 𝑑 ≈ −0.19), but at effect magnitudes too small to drive gradient-based learning at the campaign sample size. The original learning hypothesis is not supported by the data. The primary contributions are the diagnostic methodology — proposed as a transferable falsification protocol for online RL/GT defense pipelines learning from delayed environmental reward — and the open, reproducible experimental infrastructure. We outline reward reformulation as the highest-priority aspirational next step given the underpowered-but-aligned Q6 reading, with hardware-in-the-loop evaluation as the broadest scope-expansion option.

Blakely, Benjamin [Argonne National Laboratory (AN↗