SEARCH · Search NASA

Results for “Hierarchical Clustering”

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 253 records · Page 14

All sources NASA NTRS DOE OSTI

DOE OSTI2024

Investigating the influence of particle distribution on force and torque statistics using hierarchical machine learning

An accurate representation of hydrodynamic force and torque experienced by every particle in a distribution can be obtained from particle resolved (PR) simulations. These unique quantities are influenced by the deterministic position of surrounding particles. However, systems simulated with this methodology are typically limited to particles due to the involved computational cost. This resource requirement is a major bottleneck in analyzing the effect of variations in particle distribution. Here, this article attempts to address this bottleneck by availing relatively inexpensive deep learning models. The surrogate models that we employ in this article use a physics‐based hierarchical framework and symmetry‐preserving neural networks to achieve robustness with limited training data. This article first performs additional generalizability tests on PR data of distinct distributions that are not involved in the training process. The models are then deployed on several different particle distributions. Impact of clustering and structure on the observed statistics are investigated.

42 ENGINEERING↗

DOE OSTI2022

High performance FPGA embedded system for machine learning based tracking and trigger in sPhenix and EIC

We present a comprehensive end-to-end pipeline to classify triggers versus background events in this paper. This pipeline makes online decisions to select signal data and enables the intelligent trigger system for efficient data collection in the Data Acquisition System (DAQ) of the upcoming sPHENIX and future EIC (Electron-Ion Collider) experiments. Starting from the coordinates of pixel hits that are lightened by passing particles in the detector, the pipeline applies three-stage of event processing (hits clustering, track reconstruction, and trigger detection) and labels all processed events with the binary tag of trigger versus background events. The pipeline consists of deterministic algorithms such as clustering pixels to reduce event size, tracking reconstruction to predict candidate edges, and advanced graph neural network-based models for recognizing the entire jet pattern. In particular, we apply the message-passing graph neural network to predict links between hits and reconstruct tracks and a hierarchical pooling algorithm (DiffPool) to make the graph-level trigger detection. We obtain an impressive performance (≥70% accuracy) for trigger detection with only 3200 neuron weights in the end-to-end pipeline. We deploy the end-to-end pipeline into a field-programmable gate array (FPGA) and accelerate the three stages with speedup factors of 1152, 280, and 21, respectively.

Instruments & Instrumentation↗

NASA NTRS1993

A search for quasar protoclusters at z greater than 4

In the CDM and many other hierarchical scenarios for the origins of large scale structure, the existence of luminous quasars at very high redshifts (z greater than 3 or 4) is difficult to understand, unless such objects form at the very highest peaks of the density field. One then might expect a strong clustering of quasars at large redshifts. This is a generic prediction for practically any reasonable primordial density fluctuation spectrum. For CDM, Efstathiou & Rees (1988) predicted that quasars at z greater than 4 should be clustered as strongly as the bright galaxies at z approx. than 0. Cole & Kaiser (1989) suggest that z greater than 4 quasars might represent greater than or approximately = 4(sigma) peaks of the density field and thus, should be clustered more strongly than galaxies at z approximately = 0. We are performing the following experiment: a search for quasars, AGN, or other discrete objects, e.g., starforming galaxies, near known, z greater than 4 quasars. In other words, use the early quasars as markers of possible protoclusters. This is a fairly basic test of our understanding of the formation of galaxies, large-scale structure, and the origin of the first quasars themselves.

Smith, J. D.↗

NASA NTRS2015

Advances in Parallelization for Large Scale Oct-Tree Mesh Generation

Despite great advancements in the parallelization of numerical simulation codes over the last 20 years, it is still common to perform grid generation in serial. Generating large scale grids in serial often requires using special "grid generation" compute machines that can have more than ten times the memory of average machines. While some parallel mesh generation techniques have been proposed, generating very large meshes for LES or aeroacoustic simulations is still a challenging problem. An automated method for the parallel generation of very large scale off-body hierarchical meshes is presented here. This work enables large scale parallel generation of off-body meshes by using a novel combination of parallel grid generation techniques and a hybrid "top down" and "bottom up" oct-tree method. Meshes are generated using hardware commonly found in parallel compute clusters. The capability to generate very large meshes is demonstrated by the generation of off-body meshes surrounding complex aerospace geometries. Results are shown including a one billion cell mesh generated around a Predator Unmanned Aerial Vehicle geometry, which was generated on 64 processors in under 45 minutes.

O'Connell, Matthew↗

DOE OSTI2021

Enhancing the inherent catalytic activity and stability of TiO 2 supported Pt single-atoms at CeO x –TiO 2 interfaces

Single-atoms (SAs) with atomically coordinated reaction centers are considered the next generation of catalysts that can exhibit exceptional catalytic efficiency. However, the general concern about thermodynamic vulnerabilities of SAs questions their practical value. Moreover, whether the inherent catalytic nature of SAs is superior compared with that of larger nanoparticles is still under debate. Here, we address two controversies by a comparative study using two catalysts: Pt/TiO 2 and Pt/CeO x –TiO 2 . Based on a hierarchical study of density functional theory, time-resolved catalysis performance test, in situ infrared spectroscopy, and operando X-ray absorption spectroscopy, we could unveil the catalytic nature of Pt-SAs and their stability. By utilizing the heterogeneous interface formed between TiO 2 supporting particles and CeO x clusters formed on the surface of TiO 2 , we preferentially synthesized Pt-SAs pinned at the CeO x –TiO 2 interfaces on CeO x –TiO 2 hybrid-oxide supports. The strong electronic coupling between the Pt-SAs and the Ce ions at the CeO x –TiO 2 interfaces enhanced the catalytic activity toward CO oxidation of Pt-SAs and improved the long-term stability under CO oxidation conditions. The CO oxidation activity of Pt-SAs stabilized at CeO x –TiO 2 was improved by 13.5 times at 200 °C compared with the Pt-SAs on TiO 2 . The results present how to easily improve the activity and stability of Pt-SAs using a simple interface control method. Moreover, we demonstrate that the catalytic activity and the stability of Pt-SAs can be monitored through the chemical state of the interfaces. Finally, our study provides comprehensive understanding about the catalytic nature as well as a novel strategy toward applications of Pt-SAs, enabling sustainable use of Pt in heterogeneous catalysts.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2023

The cosmic DANCe of Perseus

Context. Star-forming regions are excellent benchmarks for testing and validating theories of star formation and stellar evolution. The Perseus star-forming region, being one of the youngest (< 10 Myr), closest (280-320 pc), and most studied in the literature, is a fundamental benchmark. Aims. We aim to study the membership, phase-space structure, mass, and energy (kinetic plus potential) distribution of the Perseus star-forming region using public catalogues (Gaia, APOGEE, 2MASS, and Pan-STARRS). Methods. We used Bayesian methodologies that account for extinction to identify the Perseus physical groups in the phase-space, retrieve their candidate members, derive their properties (age, mass, 3D positions, 3D velocities, and energy), and attempt to reconstruct their origin. Results. We identify 1052 candidate members in seven physical groups (one of them new) with ages between 3 and 10 Myr, dynamical super-virial states, and large fractions of energetically unbounded stars. Their mass distributions are broadly compatible with that of Chabrier for masses ≳0.1 M ⊙ and do not show hints of over-abundance of low-mass stars in NGC 1333 with respect to IC 348. These groups’ ages, spatial structure, and kinematics are compatible with at least three generations of stars. Future work is still needed to clarify if the formation of the youngest was triggered by the oldest. Conclusions. The exquisite Gaia data complemented with public archives and mined with comprehensive Bayesian methodologies allow us to identify 31% more members than previous studies, discover a new physical group (Gorgophone: 7 Myr, 191 members, and 145 M ⊙ ), and confirm that the spatial, kinematic, and energy distributions of these groups support the hierarchical star formation scenario.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

DOE OSTI2022

Insights into the origin of halo mass profiles from machine learning

ABSTRACT The mass distribution of dark matter haloes is the result of the hierarchical growth of initial density perturbations through mass accretion and mergers. We use an interpretable machine-learning framework to provide physical insights into the origin of the spherically-averaged mass profile of dark matter haloes. We train a gradient-boosted-trees algorithm to predict the final mass profiles of cluster-sized haloes, and measure the importance of the different inputs provided to the algorithm. We find two primary scales in the initial conditions (ICs) that impact the final mass profile: the density at approximately the scale of the haloes’ Lagrangian patch RL ($R\sim 0.7\, R_L$) and that in the large-scale environment (R ∼ 1.7 RL). The model also identifies three primary time-scales in the halo assembly history that affect the final profile: (i) the formation time of the virialized, collapsed material inside the halo, (ii) the dynamical time, which captures the dynamically unrelaxed, infalling component of the halo over its first orbit, (iii) a third, most recent time-scale, which captures the impact on the outer profile of recent massive merger events. While the inner profile retains memory of the ICs, this information alone is insufficient to yield accurate predictions for the outer profile. As we add information about the haloes’ mass accretion history, we find a significant improvement in the predicted profiles at all radii. Our machine-learning framework provides novel insights into the role of the ICs and the mass assembly history in determining the final mass profile of cluster-sized haloes.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2022

EJFAT: Towards Intelligent Compute Destination Load Balancing

To handle increased data flow, Jefferson Lab (JLab) is partnering with ESnet for development of an AI/ML directed compute work Load Balancer (LB) of UDP streamed data. The LB is FPGA based featuring dynamically configurable, low latency and high throughput destination address switching. The LB provides integration of edge and core computing to support JLab experimental programs, the Electron-Ion Collider, as well as data centers of the future. In the ESnet/JLab FPGA Accelerated Transport (EJFAT) initiative, the function of the LB Data Plane (DP) is to redirect data streams to selectable (but unknown to sender) destination hosts based on current worload and within that host to destination ports as a function of sub- stream id. This effects hierarchical scaling, first across compute machines for processing over a series of events and second, across ports so different data source sub-streams may be assigned to different processors for further parallelization. The LB Control Plane (CP) programs the DP using compute farm telemetry to direct and balance workloads across a compute cluster as the operating conditions require. While Proportional/Integrative/Derivative (PID) controllers are often seen in similar applications, here we investigate the feasibility of a Reinforcement Learning (RL) based schedule manager running in the CP to provide dynamic updates to the DP scheduling policy.

Lawrence, David↗

NASA NTRS1991

An optimal sensing strategy for recognition and localization of 3-D natural quadric objects

An optimal sensing strategy for an optical proximity sensor system engaged in the recognition and localization of 3-D natural quadric objects is presented. The optimal sensing strategy consists of the selection of an optimal beam orientation and the determination of an optimal probing plane that compose an optimal data collection operation known as an optimal probing. The decision of an optimal probing is based on the measure of discrimination power of a cluster of surfaces on a multiple interpretation image (MII), where the measure of discrimination power is defined in terms of a utility function computing the expected number of interpretations that can be pruned out by a probing. An object representation suitable for active sensing based on a surface description vector (SDV) distribution graph and hierarchical tables is presented. Experimental results are shown.

Lee, Sukhan↗

DOE OSTI2020

Hierarchical Membrane Dynamics in Phase-Separated Model Membranes

Lipid membranes, the primary matrix of cell membranes, exhibit a hierarchy of dynamics from molecular motions to collective undulations, which control spatiotemporal membrane phenomena and regulate various cellular functions. Thus, understanding how various dynamic modes influence membrane behavior is crucial to uncovering the design rules of cell membranes and implementing them in future technologies. Previous membrane dynamics studies using neutron spin-echo spectroscopy (NSE) have shed light on the dependence of collective membrane fluctuations on viscoelastic membrane properties, demonstrating that thickness fluctuation modes (100ns) are dictated by membrane viscosity, which directly governs fast (ps) molecular diffusive motions. Here, we focus on phase-separating lipid membranes, composed of DMPC:DSPC mixtures, to understand the effect of domain formation/growth on hierarchical matrix dynamics using selective lipid deuteration and different neutron spectroscopy techniques. NSE measurements on DMPC:DSPC-d83 (70:30 mol%) show that bending rigidity of the DMPC-rich matrix graduallyincreases with decreasing temperature - as the fully-fluid membrane approaches the upper phase transition (i.e. with the formation of transient DSPC clusters) - and increases further in the phase coexistence region with the growth of DSPC domains. Comparatively, NSE measurements on DMPC-d54:DSPC-d83 vesicles show pronounced matrix thickness fluctuations in the fully-fluid phase and drastic suppression of this fluctuation mode in the gel-fluid coexistence phase, even when the DMPC-rich matrix is still in its fluid phase. This suppression in fluctuations suggests an increase in the matrix viscosity with the growth of gel DSPC domains, which is further investigated by selectively measuring lipid diffusion in the DMPC-rich matrix using neutron backscattering spectroscopy. Comparisons of the experimental results with coarse-grained MD simulationsillustrate the underlying coupling mechanisms. Overall, these observations indicate a clear mechanical and viscoelastic coupling betweenlipid domains and their host matrix, suggesting that domain formation influences membrane properties beyond local effects.

Carrillo, Jan Michael↗

DOE OSTI2026

From Cell to System: Accelerated hpc Simulations of BESS Aging under Frequency Regulation and Arbitrage use cases

Lithium-ion battery energy storage systems (BESS) packs have emerged as a leading solution for grid-scale energy storage, enhancing resiliency and balancing load fluctuations. Yet, experimental characterization of large-format LIB packs-particularly to assess performance and degradation over hundreds of cycles - demands substantial hardware investment and multi-year testing campaigns. In this work, we couple a hierarchical, physics-based modeling framework agnostic to electrode chemistries with high-performance computing to accelerate systems level evaluation by upto two orders of magnitude. Building on the open-source liionpack platform, we implement cell, module, and pack-scale electrochemical models enriched with mechanistic aging mechanisms and deploy them on an HPC cluster to simulate 150−200kWh systems over 500 - 1,000 cycles with in days. We subject these virtual B ESS to both constant-current cycling and realistic grid service profiles spanning frequency regulation, ramp-rate support, and energy arbitrage-and quantify the resulting degradation patterns. Our results reveal that localized cell aging can induce substantial nonuniformity at module and pack levels, with service-specific cycling protocols driving distinct aging modes. This rapid, multiscale modeling approach provides a powerful design-space exploration tool for optimizing electrical architecture, control strategies, and operational schedules to prolong pack lifetime and lower total cost of ownership.

Ayalasomayajula, Surya [ORNL] (ORCID:0009000860788↗

DOE OSTI2022

Controlled synthesis of hierarchically-structured hybrid materials through peptoid engineering

This disclosure concerns a method for making a hybrid organic/inorganic material having an architecture. The method includes combining a peptoid comprising a sequence of N-substituted glycine residues and an inorganic material or inorganic material precursor to form a hybrid organic/inorganic material comprising the peptoid and the inorganic material, the hybrid organic/inorganic material having an architecture based at least in part on the peptoid sequence. The hybrid organic/inorganic materials include a cluster of nanoparticles of an inorganic material and peptoids, the cluster formed by random attachment of nanoparticles to one another by peptoid-peptoid and peptoid-nanoparticle surface interactions, wherein the hybrid material has an architecture based at least in part on the peptoid sequence.

Chen, Chun-Long↗

DOE OSTI2026

FitCache: A Transparent Drop-In Framework for Multi-Tier Caching to Accelerate Distributed Deep Learning Workloads

Training in Deep learning (DL) remains highly compute- and data-intensive, with I/O becoming a critical bottleneck as models and datasets scale. Recent studies report that data loading can dominate training time, especially on large-scale HPC systems with shared parallel file systems (PFS). Existing caching approaches either rely on single-tier designs or require intrusive modifications to training pipelines, limiting their portability and effectiveness. In this work, we present FitCache, a transparent drop-in framework for multi-tier caching to accelerate distributed DL training by coordinating fast local memory (e.g., DRAM, Persistent Memory (PMem)) and NVMe as hierarchical caches atop PFS. Our design adapts to hardware diversity, i.e., if NVMe is missing, memory transparently acts as a caching tier, ensuring stable performance. FitCache transparently intercepts I/O requests and issues concurrent fetches across all tiers, returning data from the fastest responder without centralized metadata or static redirection paths. FitCache adapts to dynamic workloads and heterogeneous clusters while maintaining POSIX compatibility. Experiments on Frontier (2048 GPUs) and smaller research clusters show that FitCache reduces training time by up to 40% and per-batch I/O latency by up to 71.6% compared to Lustre Orion PFS, offering a drop-in solution for scalable DL training.

Hu, Guangxing [ORNL] (ORCID:0009000283203614)↗

DOE OSTI2022

Two-Stage Assembly of Nanoparticle Superlattices with Multiscale Organization

Self-assembly processes, while promising for enabling the fabrication of complexly organized nanomaterials from nanoparticles, are often limited in creating structures with multiscale order. These limitations are due to difficulties in practically realizing the assembly processes required to achieve such complex organizations. For a long time, a hierarchical assembly attracted interest as a potentially powerful approach. However, due to the experimental limitations, intermediate-level structures are often heterogeneous in composition and structure, which significantly impacts the formation of large-scale organizations. Here, we introduce a two-stage assembly strategy: DNA origami frames scaffold a coordination of nanoparticles into designed 3D nanoclusters, and then these clusters are assembled into ordered lattices whose types are determined by the clusters’ valence. Through modulating the nanocluster architectures and intercluster bindings, we demonstrate the successful formation of complexly organized nanoparticle crystals. To conclude, the presented two-stage assembly method provides a powerful fabrication strategy for creating nanoparticle superlattices with prescribed unit cells.

36 MATERIALS SCIENCE↗

DOE OSTI2020

Rapid Generation of Hierarchically Porous Metal–Organic Frameworks through Laser Photolysis

Hierarchically porous metal–organic frameworks (HP-MOFs) facilitate mass transfer due to mesoporosity while preserving the advantage of microporosity. This unique feature endows HP-MOFs with remarkable application potential in multiple fields. Recently, new methods such as linker labilization for the construction of HP-MOFs have emerged. To further enrich the synthetic toolkit of MOFs, we report a controlled photolytic removal of linkers to create mesopores within microporous MOFs at tens of milliseconds. Ultraviolet (UV) laser has been applied to eliminate “photolabile” linkers without affecting the overall crystallinity and integrity of the original framework. Presumably, the creation of mesopores can be attributed to the missing-cluster defects, which can be tuned through varying the time of laser exposure and ratio of photolabile/robust linkers. Upon laser exposure, MOF crystals shrank while metal oxide nanoparticles formed giving rise to the HP-MOFs. In addition, photolysis can also be utilized for the fabrication of complicated patterns with high precision, paving the way towards MOF lithography, which has enormous potential in sensing and catalysis.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2021

Hierarchical excitations from correlated spin tetrahedra on the breathing pyrochlore lattice

The hierarchy of the coupling strengths in a physical system often engenders an effective model at low energies where the decoupled high-energy modes are integrated out. Here, using neutron scattering, we show that the spin excitations in the breathing pyrochlore lattice compound CuInCr 4 S 8 are hierarchical and can be approximated by an effective model of correlated tetrahedra at low energies. At higher energies, intratetrahedron excitations together with strong magnon-phonon couplings are observed, which suggests the possible role of the lattice degree of freedom in stabilizing the spin tetrahedra. Finally, our work illustrates the spin dynamics in CuInCr 4 S 8 and demonstrates a general effective-cluster approach to understand the dynamics on the breathing-type lattices.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

DOE OSTI2023

CEGANN: CRYSTAL EDGE GRAPH ATTENTION NEURAL NETWORK

SF-22-156 Machine learning (ML) models and applications in materials design and discovery typically involve the use of feature representations or descriptors followed by a learning algorithm that maps them to user desired properties of interest. Most popular mathematical formulation-based descriptors are not unique across atomic environments and suffer from transferability issues across different application domains and/or material classes. The CEGANN code provides a unified interface to facilitate material characterization across materials across multiple scales (from atomic to mesoscale) and diverse classes of materials ranging from metals oxides, non-metals, and even hierarchical materials such as zeolites and semi ordered materials such as mesophases. CEGANN implements a Graph Attention Network (GAT) type convolution architecture. The details of network architecture can be found in the paper https://doi.org/10.48550/arXiv.2207.10168. The software comes with pretrained examples and dataset for the classification of the following representative systems: (1) Structure-level representation such as space group (2) Structural dimensionality (e.g., bulk, 2D, clusters etc.) (3) Grain boundary identification (4) Nucleation and growth of a zeolite polymorph (5) Characterization of binary mesophases and their phase transitions (6) Growth of ice. The code is written in python programming language.

CHAN, HENRYT↗

DOE OSTI2024

Construction of Reconfigurable and Polymorphic DNA Origami Assemblies with Coiled‐Coil Patches and Patterns

Abstract DNA origami nanodevices achieve programmable structure and tunable mechanical and dynamic properties by leveraging the sequence‐specific interactions of nucleic acids. Previous advances have also established DNA origami as a useful building block to make well‐defined micron‐scale structures through hierarchical self‐assembly, but these efforts have largely leveraged the structural features of DNA origami. The tunable dynamic and mechanical properties also provide an opportunity to make assemblies with adaptive structures and properties. Here the integration of DNA origami hinge nanodevices and coiled‐coil peptides are reported into hybrid reconfigurable assemblies. With the same dynamic device and peptide interaction, it is made multiple higher‐order assemblies (i.e., polymorphic assembly) by organizing clusters of peptides into patches or arranging single peptides into patterns on the surfaces of DNA origami to control the relative orientation of devices. The coiled‐coil interactions are used to construct circular and linear assemblies whose structure and mechanical properties can be modulated with DNA‐based reconfiguration. Reconfiguration of linear assemblies leads to micron scale motions and ≈2.5‐10‐fold increase in bending stiffness. The results provide a foundation for stimulus‐responsive hybrid assemblies that can adapt their structure and properties in response to nucleic acid, peptide, protein, or other triggers.

59 BASIC BIOLOGICAL SCIENCES↗