Search NASA⌕ Search

SEARCH · Search NASA

Results for “Model initialization”

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 577 records · Page 32
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Prediction of laser beam spatial profiles in a high-energy laser facility by use of deep learning

We adapt the significant advances achieved recently in the field of generative artificial intelligence/machine-learning to laser performance modeling in multipass, high-energy laser systems with application to high-shot-rate facilities relevant to inertial fusion energy. Advantages of neural-network architectures include rapid prediction capability, data-driven processing, and the possibility to implement such architectures within future low-latency, low-power consumption photonic networks. Four models were investigated that differed in their generator loss functions and utilized the U-Net encoder/decoder architecture with either a reconstruction loss alone or combined with an adversarial network loss. We achieved inference times of 1.3 ms for a 256 × 256 pixel near-field beam with errors in predicted energy of the order of 1% over most of the energy range. It is shown that prediction errors are significantly reduced by ensemble averaging the models with different weight initializations. These results suggest that including the temporal dimension in such models may provide accurate, real-time spatiotemporal predictions of laser performance in high-shot-rate laser systems.

47 OTHER INSTRUMENTATION↗
DOE OSTI2024

NuGraph2: A Graph Neural Network for Neutrino Event Reconstruction

Neutrino experiments are set to probe some of the most important open questions in physics, from CP violation and the nature of dark matter. The technology of choice for many of these experiments is the liquid argon time projection chamber (LArTPC). In current LArTPC experiments, reconstruction performance often represents a limiting factor for the sensitivity. New developments are therefore needed to unlock the full potential of LArTPC experiments. NuGraph2 is a state of the art Graph Neural Network for reconstruction of data in LArTPC experiments. NuGraph2 utilizes a heterogeneous graph structure, with separate subgraphs of 2D nodes (hits in each plane) connected across planes via 3D nodes (space points). The model provides a consistent description of the neutrino interaction across all planes. NuGraph2 is a multi-purpose network, with a common message-passing attention engine connected to multiple decoders with different classification or regression tasks. These include the classification of detector hits according to the particle type that produced them (semantic segmentation) and the separation of hits from the neutrino interaction from hits due to noise or cosmic-ray background. Additional decoders are being developed, performing tasks such as the regression of the neutrino interaction vertex position. Performance results will be presented based on publicly available samples from MicroBooNE. These include both physics performance metrics, achieving 95% accuracy for semantic segmentation and 98% classification of neutrino hits, as well as computational metrics for training and for inference on CPU or GPU. The status of the NuGraph integration in the LArSoft software framework will be presented, as well as initial studies about model interpretability and injection of domain knowledge.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

NuGraph2: A Graph Neural Network for Neutrino Event Reconstruction

Neutrino experiments are set to probe some of the most important open questions in physics, from CP violation and the nature of dark matter. The technology of choice for many of these experiments is the liquid argon time projection chamber (LArTPC). In current LArTPC experiments, reconstruction performance often represents a limiting factor for the sensitivity. New developments are therefore needed to unlock the full potential of LArTPC experiments. NuGraph2 is a state of the art Graph Neural Network for reconstruction of data in LArTPC experiments. NuGraph2 utilizes a heterogeneous graph structure, with separate subgraphs of 2D nodes (hits in each plane) connected across planes via 3D nodes (space points). The model provides a consistent description of the neutrino interaction across all planes. NuGraph2 is a multi-purpose network, with a common message-passing attention engine connected to multiple decoders with different classification or regression tasks. These include the classification of detector hits according to the particle type that produced them (semantic segmentation) and the separation of hits from the neutrino interaction from hits due to noise or cosmic-ray background. Additional decoders are being developed, performing tasks such as the regression of the neutrino interaction vertex position. Performance results will be presented based on publicly available samples from MicroBooNE. These include both physics performance metrics, achieving 95% accuracy for semantic segmentation and 98% classification of neutrino hits, as well as computational metrics for training and for inference on CPU or GPU. The status of the NuGraph integration in the LArSoft software framework will be presented, as well as initial studies about model interpretability and injection of domain knowledge.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

NuGraph2: A Graph Neural Network for Neutrino Event Reconstruction

Neutrino experiments are set to probe some of the most important open questions in physics, from CP violation and the nature of dark matter. The technology of choice for many of these experiments is the liquid argon time projection chamber (LArTPC). In current LArTPC experiments, reconstruction performance often represents a limiting factor for the sensitivity. New developments are therefore needed to unlock the full potential of LArTPC experiments. NuGraph2 is a state of the art Graph Neural Network for reconstruction of data in LArTPC experiments [https://arxiv.org/abs/2403.11872]. NuGraph2 utilizes a heterogeneous graph structure, with separate subgraphs of 2D nodes (hits in each plane) connected across planes via 3D nodes (space points). The model provides a consistent description of the neutrino interaction across all planes. NuGraph2 is a multi-purpose network, with a common message-passing attention engine connected to multiple decoders with different classification or regression tasks. These include the classification of detector hits according to the particle type that produced them (semantic segmentation) and the separation of hits from the neutrino interaction from hits due to noise or cosmic-ray background. Additional decoders are being developed, performing tasks such as the regression of the neutrino interaction vertex position. Performance results will be presented based on publicly available samples from MicroBooNE. These include both physics performance metrics, achieving 95% accuracy for semantic segmentation and 98% classification of neutrino hits, as well as computational metrics for training and for inference on CPU or GPU. The status of the NuGraph integration in the LArSoft software framework will be presented, as well as initial studies about model interpretability and injection of domain knowledge.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2021

Phase-field modeling of rate-dependent fluid-driven fracture initiation and propagation

The rate-dependent behavior associated with deformation and fracturing of materials, such as natural rocks, poses significant challenges for modeling. In addition to the complications of the viscoelastic response, the speed of fracture propagation reflects micromechanical mechanisms in the fracture process zone(FPZ). In order to represent these complicated behaviors, a thermodynamically consistent, rate-dependent fracture model is required. Based on rigorous thermodynamic principles, we derive a rate-dependent phase-field mechanical model coupled with single-phase fluid flow in both the matrix and the fracture. The model is guaranteed to satisfy energy conservation during fracture propagation. Here, the system of equations is solved using the introduced solution procedure and a novel preconditioner that accounts for the complex fluid-structure interaction. The proposed phase-field model is tested against several benchmark problems on solid-fluid coupling, fluid-driven fracture propagation and rate-dependent viscoelastic deformation. The model serves as a strong basis for investigating rate-dependent fracturing experiments and for making predictions of material behaviors under new conditions.

42 ENGINEERING↗
DOE OSTI2018

Predicting Near-Term Changes in the Earth System: A Large Ensemble of Initialized Decadal Prediction Simulations Using the Community Earth System Model

The objective of near-term climate prediction is to improve our fore-knowledge, from years to a decade or more in advance, of impactful climate changes that can in general be attributed to a combination of internal and externally forced variability. Predictions initialized using observations of past climate states are tested by comparing their ability to reproduce past climate evolution with that of uninitialized simulations in which the same radiative forcings are applied. A new set of decadal prediction (DP) simulations has recently been completed using the Community Earth System Model (CESM) and is now available to the community. This new large-ensemble (LE) set (CESM-DPLE) is composed of historical simulations that are integrated forward for 10 years following initialization on 1 November of each year between 1954 and 2015. CESM-DPLE represents the “initialized” counterpart to the widely studied CESM Large Ensemble (CESM-LE); both simulation sets have 40-member ensembles, and they use identical model code and radiative forcings. Comparing CESM-DPLE to CESM-LE highlights the impacts of initialization on prediction skill and indicates that robust assessment and interpretation of DP skill may require much larger ensembles than current protocols recommend. CESM-DPLE exhibits significant and potentially useful prediction skill for a wide range of fields, regions, and time scales, and it shows widespread improvement over simpler benchmark forecasts as well as over a previous initialized system that was submitted to phase 5 of the Coupled Model Intercomparison Project (CMIP5). Finally, the new DP system offers new capabilities that will be of interest to a broad community pursuing Earth system prediction research.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Power handling in a highly-radiative negative triangularity pilot plant

Abstract This work explores detailed power handling solutions for a class of high-field, highly-radiative negative triangularity (NT) reactors based around the MANTA concept (Rutherford et al 2024 Plasma Phys. Control. Fusion ). The divertor design is kept as simple as possible, opting for a standard divertor with standard leg length. FreeGS is used to create an equilibrium for the boundary region, prioritizing a short outer leg length of only ∼50 cm (∼40% of the minor radius). The UEDGE code package is used for the boundary plasma solution to track plasma temperatures and fluxes to the divertor targets. It is found that for P SOL = 25 MW and n sep = 0.96 × 10 20 m −3 , conditions consistent with initial core transport modeling, little additional power mitigation is necessary. For a fixed impurity fraction of just 0.13% Ne in the plasma, the peak heat flux density at the more heavily loaded outer targets falls to 7.8 MW m −2 , while the electron temperature T e remains just under 5 eV. Scans around the parameter space reveal that even at densities lower than in the primary operating scenario, P SOL can be increased up to 50 MW, so long as a slightly higher fraction of extrinsic radiator is used. With less than 1% neon (Ne) impurity content, the divertor still experiences less than 10 MW m −2 at the outer target. Design of the plasma-facing components includes a close-fitting vacuum vessel with a tungsten inner surface as well as FLiBe-carrying cooling channels fashioned into the VV wall directly behind the divertor targets. For the seeded heat flux profile, Ansys Fluent heat transfer simulations estimate that the outer target temperature remains at just below 1550 ∘ C. Initial scoping of advanced divertor designs shows that for an X-divertor, detachment of the outer target becomes much simpler, and plasma fluxes to the targets drop considerably with only 0.01% Ne content.

Miller, M. A. (ORCID:0000000265406533)↗
DOE OSTI2026

A&L Annual Report: Kapton Strength

Exploding Foil Initiator (EFI) systems function by launching a polymer flyer (often referred to as a slapper) at a high explosive (HE) pellet, which is intended to shock-initiate the HE. Flyer characteristics (e.g., shape, planarity, etc.) are extremely important in this respect, governing their ability to initiate the HE. Polymer models currently used in these systems are largely unvalidated, potentially compromising the integrity of magneto-hydrodynamic (MHD) predictions of EFI function. An improved EFI MHD modelling capability is expected to: (1) expedite optimization of EFI-based initiation system design, (2) enable consideration of EFI-related aging and corresponding lifetime predictions, and (3) potentially minimize the number of costly experiments required to certify EFI designs. To address this knowledge gap, several years ago we began a computational project to develop a higher-fidelity model for Kapton, a Polyimide copolymer most commonly employed as the flyer material in EFI devices.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Coupled geomechanical investigation of depletion-induced fault reactivation

Fault reactivation during subsurface fluid production pose significant challenges to safe and sustainable resource extraction. Here, this study presents a three-dimensional coupled geomechanical framework to investigate the processes driving fault reactivation, capturing the interactions between reservoir dynamics and geomechanical responses. Verification against theoretical estimations based on linear poroelasticity confirms the model's capacity in representing reservoir background stress responses. However, the study reveals that relying solely on background stress states can underestimate or overestimate fault reactivation potential, emphasizing the importance of including localized stress perturbations such as differential compaction and stress redistribution. Applied to a fault (M1) inspired by the geological characteristics of the Groningen field, the model shows slip initiation at 2965 m depth with 16.0 MPa depletion, aligning with field observations where seismicity occurred at approximately 3 km depth after 15.8 MPa depletion. Parametric studies reveal: (1) inelastic reservoir compaction delays fault reactivation and mitigates fault slip by reducing stress concentration, (2) higher intermediate in-situ stress magnitudes decrease the Coulomb Failure Stress (CFS) increase rate and reduce fault slip, (3) larger fault offsets amplify shear stress near the offset zone, promoting earlier reactivation and longer rupture propagation, and (4) fault permeability significantly influences pressure diffusion, with low-permeability faults leading to sharper stress changes and earlier fault destabilization. These insights highlight the critical role of geological and mechanical parameters in fault reactivation and provide a predictive framework for mitigating induced seismicity risks.

Coupled geomechanical modeling↗
DOE OSTI2026

Quark flavor equilibration of the quark-gluon plasma

The early stage of a heavy-ion collision is marked by rapid entropy production and the transition from a gluon saturated initial condition to a plasma of quarks and gluons that evolves hydrodynamically. However, during the early times of the hydrodynamic evolution, the chemical composition of the QCD medium is still largely unknown. We present a study of quark chemical equilibration in the (Q)GP using a novel model of viscous hydrodynamic evolution in partial chemical equilibrium. Motivated by the success of gluon saturated initial condition models, we initialize the QCD medium as a completely gluon dominated state. Local quark production during the hydrodynamic phase is then simulated through the evolution of time-dependent fugacities for each independent quark flavor, with the timescales set as free parameters to compare different rates of equilibration. We present the results of complete heavy-ion collision simulations using this partial chemical equilibrium model, and show the effects on hadronic and electromagnetic observables. In particular, we show that the development of flow is sensitive to the equilibration timescale, providing an empirical way to probe the chemical equilibration of the QCD medium.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

A method for examining ensemble averaging forms during the transition to turbulence in HED systems for application to RANS models

This paper discusses a strategy to initialize a two-dimensional (2D) Reynolds-averaged Navier–Stokes model [LANL's Besnard–Harlow–Rauenzahn (BHR) model] in order to describe an unsteady transitional Richtmyer–Meshkov (RM)-induced flow observed in on-going high-energy-density ensemble experiments performed on the OMEGA-EP facility. The experiments consist of a nominal single-mode perturbation (initial amplitude a 0 ≈ 10 and wavelength $λ$ = 100μm) with target-to-target variations in the surface roughness subjected to the RM instability with delayed Rayleigh–Taylor in a heavy-to-light configuration. Our strategy leverages high-resolution three-dimensional (3D) implicit large eddy simulations (ILES) simulations to initialize BHR-relevant parameters and subsequently validate the 2D BHR results against the 3D ILES simulations. A suite of five 3D ILES simulations corresponding to five experimental target profiles is undertaken to generate an ensemble dataset. Using ensemble averages from the 3D simulations to initialize the turbulent kinetic energy in the BHR model ( K 0 ) demonstrates the ability of the model to predict the time evolution of the interface as well as the density-specific-volume covariance, b . To quantify the sensitivity of the BHR results to the choice of K 0 and the initial turbulent length scale, S 0 , we execute a parameter sweep spanning four orders of magnitude for both S 0 and K 0 , generating a parameter space consisting of 26 simulations. The Pearson's correlation coefficient is used as a measure of discrepancy between the 2D BHR and 3D ILES simulations and reveals that the ranges 8≲S 0 ≲20 μm and 10 9 ≲K 0 ≲10 10 cm 2 /s 2 produce predictions that agree best with the 3D ILES results.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2024

Machine Learning Design of Perovskite Catalytic Properties

Abstract Discovering new materials that efficiently catalyze the oxygen reduction and evolution reactions is critical for facilitating the widespread adoption of solid oxide fuel cell and electrolyzer (SOFC/SOEC) technologies. Here, machine learning (ML) models are developed to predict perovskite catalytic properties critical for SOFC/SOEC applications, including oxygen surface exchange, oxygen diffusivity, and area specific resistance (ASR). The models are based on trivial‐to‐calculate elemental features and are more accurate and dramatically faster than the best models based on ab initio‐derived features, potentially eliminating the need for ab initio calculations in descriptor‐based screening. The model of ASR enables temperature‐dependent predictions, has well calibrated uncertainty estimates and online accessibility. Use of temporal cross‐validation reveals the model to be effective at discovering new promising materials prior to their initial discovery, demonstrating the model can make meaningful predictions. Using the SHapley Additive ExPlanations (SHAP) approach, detailed discussion of different approaches of model featurization is provided for ML property prediction. Finally, the model is used to screen more than 19 million perovskites to develop a list of promising cheap, earth‐abundant, stable, and high performing materials, and find some top materials contain mixtures of less‐explored elements (e.g., K, Bi, Y, Ni, Cu) worth exploring in more detail.

25 ENERGY STORAGE↗
DOE OSTI2024

Getting allometry right at the Oak Ridge free‐air CO 2 enrichment experiment: Old problems and new opportunities for global change experiments

Societal Impact Statement Free‐air CO 2 enrichment (FACE) experiments provide essential data on forest responses to increasing atmospheric CO 2 for evaluations of climate change impacts on humanity. Understanding and reducing the uncertainty in the experimental results is critical to ensure scientific and public confidence in the models and policy initiatives that derive therefrom. One source of uncertainty is the estimation of tree biomass using mathematical relationships between biomass and easily obtained and non‐destructive measurements (allometry). We evaluated the robustness of the allometric relationships established at the beginning of a FACE experiment and discuss the challenges and opportunities for the new generation of FACE experiments. Summary Long‐term field experiments to elucidate forest responses to rising atmospheric CO 2 concentration require allometric equations to estimate tree biomass from non‐destructive measurements of tree size. We analyzed whether the allometric equations established at the beginning of a free‐air CO 2 enrichment (FACE) experiment in a Liquidambar styraciflua plantation were still valid at the end of the 12 year experiment. Aboveground woody biomass was initially predicted by an equation that included bole diameter, taper, and height, assuming that including taper and height as predictors would accommodate changes in tree structure that might occur over time and in response to elevated CO 2 . At the conclusion of the FACE experiment, we harvested 23 trees, measured dimensions and dry mass of boles and branches, and extracted and measured the woody root mass of 10 trees. Although 10 of the harvested trees were larger than the trees used to establish the allometric relationship, measured aboveground woody biomass was well predicted by the original allometry. The initial linear equation between bole basal area and woody root biomass underestimated final root biomass by 28%, but root biomass was just 21% of total wood mass, and errors in aboveground and belowground estimates were offsetting. The allometry established at the beginning of the experiment provided valid predictions of tree biomass throughout the experiment. New allometric approaches using terrestrial laser scanning should reduce an important source of uncertainty in decade‐long forest experiments and in assessments of centuries‐long forest biomass accretion used in evaluating carbon offsets and climate mitigation.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

A2SD: Accelerating Scientific Innovation Through Autonomous Discovery Systems

The 2025 Advancing Autonomous Scientific Discovery (A2SD) workshop convened researchers from academia, national laboratories, and industry to explore the transformative role of autonomy in scientific discovery. The workshop highlighted a convergence of artificial intelligence, robotics, and computational workflows into autonomous systems capable of accelerating the scientific process. Presentations and discussions spanned autonomous experimentation, intelligent workflow orchestration, digital twins, and agent-based systems for managing complex research ecosystems. Key challenges discussed included interoperability across heterogeneous infrastructures, near real-time data management under FAIR principles, reproducibility, and the integration of human oversight. The workshop also emphasized the need for modular software interfaces, federated learning models, and education initiatives to support a next-generation scientific workforce.

Taufer, Michela [University of Tennessee, Knoxvill↗
DOE OSTI2025

Simulated effect of defect volume and location on very high cycle fatigue of laser beam powder bed fused AlSi10Mg

This study quantifies the interaction between volumetric defect location and size on the very high cycle fatigue (VHCF) of laser beam powder bed fused (LB-PBF) AlSi10Mg. Crystal plasticity finite element method (CPFEM) simulations were used to investigate the effects of defect location and size on the driving force for crack initiation. The CPFEM model was calibrated against uniaxial and cyclic experimental data of LB-PBF AlSi10Mg. Defect characteristics were informed by experimental data from the specimens produced in various geometries to create realistic representative volume elements (RVEs) with equivalent volume fractions of defects. By embedding defects of varying sizes and locations within the RVEs, fatigue indicator parameters (FIPs) were calculated to analyze the impact of defects’ characteristics on fatigue performance. Different combinations of defect volume and locations were generated for various microstructure instantiations, providing insight into extreme value fatigue responses. Larger defect volumes located on free surfaces consistently generated the highest FIPs, suggesting defect size and boundary proximity intensify stress concentration effects. RVEs with multiple smaller defects produced lower FIPs than those with single large critical defects. These findings underscore the critical role of defect characteristics on fatigue life, providing a foundation for future predictive modeling in fatigue-sensitive AM applications.

AlSi10Mg↗
DOE OSTI2021

Crystal structure of the $τ_{11}$ Al 4 Fe 1.7 Si phase from neutron diffraction and ab initio calculations

The intermetallic τ 11 Al 4 Fe 1.7 Si phase is of interest for high-temperature structural application due to its combination of low density and high strength. We determine the crystal structure of the τ 11 phase through a combination of powder neutron diffraction and density functional theory calculations. Using Pawley and Rietveld refinements of the neutron diffraction data provides an initial crystal structure model. Since Al and Si have nearly identical neutron scattering lengths, we use density-functional calculations to determine their preferred site occupations. The τ 11 phase exhibits a hexagonal crystal structure with space group P6 3 /mmc and lattice parameters of a = 7.478 Å and c = 7.472 Å. The structure comprises five Wyckoff positions; Al occupies the 6h and 12k sites, Fe the 2a and 6h sites, and Si the 2a sites. Here we observe site disorder and partial occupancies on all sites with a large fraction of 80% Fe vacancies on the 2d sites, indicating an entropic stabilization of the τ 11 phase at high temperature.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Enabling topography-resolving structural dynamic contact simulation

Damping of structures and systems is often dominated by frictional dissipation in connections, the prediction of which remains a longstanding scientific challenge. Previous studies have shown that the actual topography of contact interfaces may have a strong effect, especially in the partial slip/liftoff regime. We recently proposed a multi-scale method, which couples finite element and boundary element modeling. The primary benefit of this approach is that it permits to analyze the effect of the actual contact topography on the dynamics of jointed structures. While this multi-scale modeling method was initially developed for quasi-static analysis, we demonstrate herein how it can be used for time step integration and Harmonic Balance analysis. We cross-verify those fully dynamic analysis methods against each other and quasi-static results, for the S4 Beam benchmark. We compare the multi-scale method against state-of-the-art full-FE analysis, in terms of numerical damping and computational performance. Some discrepancy is found to be of physical origin. Depending on the load history, it is shown that the system settles to a slightly different equilibrium. Finally, transient multi-scale simulations enable the prediction of this interesting phenomenon, for the first time, for a structure with bolted joints.

Frictional-unilateral contact↗
DOE OSTI2026

Probing gluon saturation with forward di-hadron correlations in proton-nucleus collisions

We present a detailed numerical investigation of semi-inclusive forward di-hadron production in proton–nucleus collisions employing the Color Glass Condensate effective theory. We focus on the regime where di-hadrons are produced nearly back-to-back in the transverse plane, thereby justifying a transverse-momentum-dependent factorization approach in terms of small-x gluon distributions. Our computation integrates several key elements: i) non-linear rapidity evolution via the Balitsky–Kovchegov equation with running coupling, ii) both perturbative and non-perturbative Sudakov resummation, and iii) a phenomenologically constrained model for the initial conditions for small-x gluon distributions. We compare this phenomenological framework to experimental data from the STAR Collaboration on azimuthal correlations in forward di-pion production in both proton–proton and proton–gold collisions. We analyze the systematic theoretical uncertainties associated with the saturation scales of nuclei at the initial scale for rapidity evolution and with those associated with the hadronization process. Finally, we make predictions for the kinematics anticipated to be covered by the ALICE Forward Calorimeter (FoCal) upgrade at the Large Hadron Collider.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗