SEARCH · Engineering Papers

Results for “Generative models”

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 OSTI2025

Physics-Driven Construction of Compact Primitive Gaussian Density Fitting Basis Sets

We present a model-assisted density fitting (MADF) basis set generator, an algorithm for generating primitive atomic Gaussian density fitting (DF) basis sets (DFBSs) from a contracted Gaussian orbital basis set (OBS). The MADF algorithm produces DFBSs suitable for accurate robust DF approximation of 2-particle interactions in mean-field and correlated electronic structures. The algorithm is designed to (a) saturate the OBS product space by a large regularized set of primitive solid-harmonic Gaussian shells with nonuniform distribution of exponents, followed by (b) pruning of the shells according to their contributions to the 2- body energy of a correlated atomic ensemble. Building the DFBS generator model almost exclusively on mathematical and physical principles allows one to limit the number of parameters that control the density fitting error to three, with a single set of parameters sufficient for computations with all basis cardinal numbers, with and without correlation of core electrons, with and without scalar and spin-dependent relativistic effects, spanning almost all of the Periodic Table. Performance assessment included basis sets up to quadruple-ζ quality from several major basis set families, using molecules composed of main-group, d-block, and f-block elements. The resulting DF errors in Hartree−Fock and second-order MP2 energies (with relativistic all-electron treatments, when appropriate) were on the order of 20 and 10 μE h per electron, respectively.

Approximation↗

DOE OSTI2024

Evaluation of Converter Performance Considering Static and Dynamic Device Part-to-Part Variability

This paper presents a methodology to incorporate and analyze the impact of semiconductor device part-to-part variation on power converter performance. By integrating extensive static and dynamic device characterization data with an automated compact model generation process that reflects manufacturing variability, device models with inherent variability features are utilized in converter simulations for a comprehensive assessment of performance impacts. The traditional converter performance evaluation process typically yields fixed efficiency values, often dismissing the inherent part-to-part variability caused by the manufacturing process of semiconductor devices. To address this limitation, a large population of devices was characterized to capture variations in static parameters-such as transfer, output, and capacitance characteristics-as well as dynamic behaviors, including switching losses. This data-driven approach enables the development of individual compact models, which were then integrated into converter simulations to evaluate efficiency ranges rather than single point estimated values. The converter simulation results show that part-to-part component variation can lead to significant efficiency deviations, exceeding several percentage points in high-power conversion applications. By offering a more accurate representation of converter behavior under real-world manufacturing conditions, this methodology enables designers to anticipate performance variability, improving the robustness of power converter designs.

device characterization↗

DOE OSTI2022

Balrog

SAND2022-12602 O Balrog uses mathematical methods to calculate spectral line shapes in high-energy-density plasmas. Line shapes are a result of atomic perturbations by their surrounding plasmas, and this model captures much of the physics and generates model spectrum that can be used in data analysis and as an input for other models. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Gomez, Thomas↗

DOE OSTI2023

Performance Evaluation of the Starfire nGen-350 DD Neutron Generator for use with the Fast Neutron Coincidence Collar

The Starfire Industries nGen-350 DD neutron generator has been developed for use with the Fast Neutron Coincidence Collar (FNCL). The generator was adapted from the nGen-310 model generator by Starfire to accommodate an International Atomic Energy Agency (IAEA) request to incorporate a neutron yield stabilization mechanism. An nGen-350 generator was obtained by Oak Ridge National Laboratory (ORNL) for evaluation as a potential neutron interrogation source replacement for Am(Li) in the FNCL. This report provides a summary of the performance of the nGen-350.

43 PARTICLE ACCELERATORS↗

DOE OSTI2024

IMPACT: Design of Integrated Multiphysics Producible Additive Components for Turbomachinery

The overall objective of the IMPACT program was to enable a dramatic reduction in design maturation time for an additive hot-section turbomachinery component through the following: • A fast crack-risk producibility surrogate model generated from machine learning applied to additive process simulation data generated via exascale computing, • Linking this surrogate model to multi-physics topology optimization (TO) to enable the creation of producible, near-optimal structural/thermal designs for additive hot-section components, • Maturing this toolset to reduce hot-section component design-for-manufacturing iterations by a large fraction, and eventually, • Using these tools to develop more efficient gas turbines in much shorter design cycle times.

42 ENGINEERING↗

DOE OSTI2025

Is tokenization needed for masked particle modeling?

In this work, we significantly enhance masked particle modeling (MPM), a self-supervised learning scheme for constructing highly expressive representations of unordered sets relevant to developing foundation models for high-energy physics. In MPM, a model is trained to recover the missing elements of a set, a learning objective that requires no labels and can be applied directly to experimental data. We achieve significant performance improvements over previous work on MPM by addressing inefficiencies in the implementation and incorporating a more powerful decoder. We compare several pre-training tasks and introduce new reconstruction methods that utilize conditional generative models without data tokenization or discretization. We show that these new methods outperform the tokenized learning objective from the original MPM on a new test bed for foundation models for jets, which includes using a wide variety of downstream tasks relevant to jet physics, such as classification, secondary vertex finding, and track identification.

conditional generative models↗

DOE OSTI2020

Hypernetwork science via high-order hypergraph walks

We propose high-order hypergraph walks as a framework to generalize graph-based network science techniques to hypergraphs. Edge incidence in hypergraphs is quantitative, yielding hypergraph walks with both length and width. Graph methods which then generalize to hypergraphs include connected component analyses, graph distance-based metrics such as closeness centrality, and motif-based measures such as clustering coefficients. We apply high-order analogs of these methods to real world hypernetworks, and show they reveal nuanced and interpretable structure that cannot be detected by graph-based methods. Lastly, we apply three generative models to the data and find that basic hypergraph properties, such as density and degree distributions, do not necessarily control these new structural measurements. Our work demonstrates how analyses of hypergraph-structured data are richer when utilizing tools tailored to capture hypergraph-native phenomena, and suggests one possible avenue towards that end.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

Separable shape tensors for aerodynamic design

Airfoil shape design is a classical problem in engineering and manufacturing. In this work, we combine principled physics-based considerations for the shape design problem with modern computational techniques using a data-driven approach. Modern and traditional analyses of two-dimensional (2D) and three-dimensional (3D) aerodynamic shapes reveal a flow-based sensitivity to specific deformations that can be represented generally by affine transformations (rotation, scaling, shearing, and translation). We present a novel representation of shapes that decouples affine-style deformations over a submanifold and a product submanifold principally of the Grassmannian. As an analytic generative model, the separable representation, informed by a database of physically relevant airfoils, offers: (i) a rich set of novel 2D airfoil deformations not previously captured in the data, (ii) an improved low-dimensional parameter domain for inferential statistics informing design/manufacturing, and (iii) consistent 3D blade representation and perturbation over a sequence of nominal 2D shapes.

42 ENGINEERING↗

DOE OSTI2022

SBbadger: biochemical reaction networks with definable degree distributions

Abstract Motivation An essential step in developing computational tools for the inference, optimization and simulation of biochemical reaction networks is gauging tool performance against earlier efforts using an appropriate set of benchmarks. General strategies for the assembly of benchmark models include collection from the literature, creation via subnetwork extraction and de novo generation. However, with respect to biochemical reaction networks, these approaches and their associated tools are either poorly suited to generate models that reflect the wide range of properties found in natural biochemical networks or to do so in numbers that enable rigorous statistical analysis. Results In this work, we present SBbadger, a python-based software tool for the generation of synthetic biochemical reaction or metabolic networks with user-defined degree distributions, multiple available kinetic formalisms and a host of other definable properties. SBbadger thus enables the creation of benchmark model sets that reflect properties of biological systems and generate the kinetics and model structures typically targeted by computational analysis and inference software. Here, we detail the computational and algorithmic workflow of SBbadger, demonstrate its performance under various settings, provide sample outputs and compare it to currently available biochemical reaction network generation software. Availability and implementation SBbadger is implemented in Python and is freely available at https://github.com/sys-bio/SBbadger and via PyPI at https://pypi.org/project/SBbadger/. Documentation can be found at https://SBbadger.readthedocs.io. Supplementary information Supplementary data are available at Bioinformatics online.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2023

Modeling Global Vegetation Gross Primary Productivity, Transpiration and Hyperspectral Canopy Radiative Transfer Simultaneously Using a Next Generation Land Surface Model—CliMA Land

Recent progress in satellite observations has provided unprecedented opportunities to monitor vegetation activity at global scale. However, a major challenge in fully utilizing remotely sensed data to constrain land surface models (LSMs) lies in inconsistencies between simulated and observed quantities. For example, gross primary productivity (GPP) and transpiration (T) that traditional LSMs simulate are not directly measurable from space, although they can be inferred from spaceborne observations using assumptions that are inconsistent with those LSMs. In comparison, canopy reflectance and fluorescence spectra that satellites can detect are not modeled by traditional LSMs. To bridge these quantities, we presented an overview of the next generation land model developed within the Climate Modeling Alliance (CliMA), and simulated global GPP, T, and hyperspectral canopy radiative transfer (RT; 400–2,500 nm for reflectance, 640–850 nm for fluorescence) at hourly time step and 1° spatial resolution using CliMA Land. CliMA Land predicts vegetation indices and outgoing radiances, including solar-induced chlorophyll fluorescence (SIF), normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), and near infrared reflectance of vegetation (NIRv) for any given sun-sensor geometry. The spatial patterns of modeled GPP, T, SIF, NDVI, EVI, and NIRv correlate significantly with existing data-driven products (mean R 2 = 0.777 for 9 products). CliMA Land would be also useful in high temporal resolution simulations, for example, providing insights into when GPP, SIF, and NIRv diverge.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2026

SimLBR: Learning to Detect Fake Images by Learning to Detect Real Images

The rapid advancement of generative models has made the detection of AI-generated images a critical challenge for both research and society. Recent works have shown that most state-of-the-art fake image detection methods overfit to their training data and catastrophically fail when evaluated on curated hard test sets with strong distribution shifts. In this work, we argue that it is more principled to learn a tight decision boundary around the real image distribution and treat the fake category as a sink class. To this end, we propose SimLBR, a simple and efficient framework for fake image detection with Latent Blending Regularization (LBR). Our method significantly improves cross-generator generalization, achieving up to +24.85% accuracy and +69.62% recall on the challenging Chameleon benchmark. SimLBR is also highly efficient, training orders of magnitude faster than existing approaches. Furthermore, we emphasize the need for reliability-oriented evaluation in fake image detection, introducing risk-adjusted metrics and worst-case estimates to better assess model robustness. All the code and models are availabe at: https://github.com/mvrl/SimLBR

Dhakal, Aayush [Washington University, St. Louis]↗

DOE OSTI2023

Machine learning simulation of finite element analysis in augmented reality

Media, method and system for approximating a finite element analysis texture map for an object. To accomplish this, the object is converted to a computer generated model and finite element analysis is performed for a plurality of different simulated inputs to generate a plurality of simulated mappings. Each simulated mapping is converted into a simulated texture map. A machine learning model is trained on the simulated inputs and simulated texture maps to generate a texture map which approximates a finite element analysis. The machine learning model receives a user input and generates the texture map therefrom. The texture map is then wrapped to the object and displayed.

Scherer, Derek Carl↗

DOE OSTI2024

DiffLense: a conditional diffusion model for super-resolution of gravitational lensing data

Abstract Gravitational lensing data is frequently collected at low resolution due to instrumental limitations and observing conditions. Machine learning-based super-resolution techniques offer a method to enhance the resolution of these images, enabling more precise measurements of lensing effects and a better understanding of the matter distribution in the lensing system. This enhancement can significantly improve our knowledge of the distribution of mass within the lensing galaxy and its environment, as well as the properties of the background source being lensed. Traditional super-resolution techniques typically learn a mapping function from lower-resolution to higher-resolution samples. However, these methods are often constrained by their dependence on optimizing a fixed distance function, which can result in the loss of intricate details crucial for astrophysical analysis. In this work, we introduce DiffLense , a novel super-resolution pipeline based on a conditional diffusion model specifically designed to enhance the resolution of gravitational lensing images obtained from the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP). Our approach adopts a generative model, leveraging the detailed structural information present in Hubble space telescope (HST) counterparts. The diffusion model, trained to generate HST data, is conditioned on HSC data pre-processed with denoising techniques and thresholding to significantly reduce noise and background interference. This process leads to a more distinct and less overlapping conditional distribution during the model’s training phase. We demonstrate that DiffLense outperforms existing state-of-the-art single-image super-resolution techniques, particularly in retaining the fine details necessary for astrophysical analyses.

Computer Science↗

DOE OSTI2026

AI-Driven Frameworks for Characterizing Urban Energy Systems

We develop AI-driven frameworks to characterize urban energy systems with the goal of transforming planning by reducing the labor of model generation, scaling scenario exploration, and improving accuracy for localized analysis. The approach integrates top-down and bottom-up data to train different AI models that predict missing information and generate inputs and targeted scenarios for district-scale models. The result is a scalable framework that provides actionable insights for reliable and efficient planning.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

DOE OSTI2022

SDSS-IV MaStar: Stellar parameter determination with continuum-supplemented full-spectrum fitting

We present a stellar parameter catalog built to accompany the MaStar Stellar Library, which is a comprehensive collection of empirical, medium-resolution stellar spectra. We constructed this parameter catalog by using a multicomponent χ 2 fitting approach to match MaStar spectra to models generated by interpolating the ATLAS9-based BOSZ model spectra. The total χ 2 for a given model is defined as the sum of components constructed to characterize narrow-band features of observed spectra (e.g., absorption lines) and the broadband continuum shape separately. Extinction and systematics due to flux calibration were taken into account in the fitting. The χ 2 distribution for a given region of model space was sampled using a Markov chain Monte Carlo (MCMC) algorithm, the data from which were then used to extract atmospheric parameter estimates (T eff , log g, [Fe/H], and [α/Fe]), their corresponding uncertainties, and direct extinction measurements. Two methods were used to extract parameters and uncertainties: one that accepts the MCMC’s prescribed minimum-χ 2 result, and one that uses Bayesian inference to compute a likelihood-weighted mean from the χ 2 distribution sampled by the MCMC. Results were evaluated for internal consistency using repeat observations where available and by comparing them with external data sets (e.g., APOGEE-2 and Gaia DR2). Our spectral-fitting exercise reveals possible deficiencies in current theoretical model spectra, illustrating the potential power of MaStar spectra for helping to improve the models. This paper represents an update to the parameters that were originally presented with SDSS-IV DR17. The MaStar parameter catalog containing our BestFit results is available on the SDSS-IV DR17 website as part of version 2 of the MaStar stellar parameter value-added catalog.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2024

The 1.3 Å resolution structure of the truncated group Ia type IV pilin from Pseudomonas aeruginosa strain P1

The type IV pilus is a diverse molecular machine capable of conferring a variety of functions and is produced by a wide range of bacterial species. The ability of the pilus to perform host-cell adherence makes it a viable target for the development of vaccines against infection by human pathogens such as Pseudomonas aeruginosa . Here, the 1.3 Å resolution crystal structure of the N-terminally truncated type IV pilin from P. aeruginosa strain P1 (ΔP1) is reported, the first structure of its phylogenetically linked group (group I) to be discussed in the literature. The structure was solved from X-ray diffraction data that were collected 20 years ago with a molecular-replacement search model generated using AlphaFold ; the effectiveness of other search models was analyzed. Examination of the high-resolution ΔP1 structure revealed a solvent network that aids in maintaining the fold of the protein. On comparing the sequence and structure of P1 with a variety of type IV pilins, it was observed that there are cases of higher structural similarities between the phylogenetic groups of P. aeruginosa than there are between the same phylogenetic group, indicating that a structural grouping of pilins may be necessary in developing antivirulence drugs and vaccines. These analyses also identified the α–β loop as the most structurally diverse domain of the pilins, which could allow it to serve a role in pilus recognition. Studies of ΔP1 in vitro polymerization demonstrate that the optimal hydrophobic catalyst for the oligomerization of the pilus from strain K122 is not conducive for pilus formation of ΔP1; a model of a three-start helical assembly using the ΔP1 structure indicates that the α–β loop and the D-loop prevent in vitro polymerization.

Bragagnolo, Nicholas↗

DOE OSTI2021

Integrating deep neural networks with full-waveform inversion: Reparameterization, regularization, and uncertainty quantification

Full-waveform inversion (FWI) is an accurate imaging approach for modeling the velocity structure by minimizing the misfit between recorded and predicted seismic waveforms. However, the strong nonlinearity of FWI resulting from fitting oscillatory waveforms can trap the optimization in local minima. We have adopted a neural-network-based full-waveform inversion (NNFWI) method that integrates deep neural networks with FWI by representing the velocity model with a generative neural network. Neural networks can naturally introduce spatial correlations as regularization to the generated velocity model, which suppresses noise in the gradients and mitigates local minima. Furthermore, the velocity model generated by neural networks is input to the same partial differential equation (PDE) solvers used in conventional FWI. The gradients of the neural networks and PDEs are calculated using automatic differentiation, which back propagates gradients through the acoustic PDEs and neural network layers to update the weights of the generative neural network. Experiments on 1D velocity models, the Marmousi model, and the 2004 BP model determine that NNFWI can mitigate local minima, especially for imaging high-contrast features such as salt bodies, and it significantly improves the inversion in the presence of noise. Adding dropout layers to the neural network model also allows analyzing the uncertainty of the inversion results through Monte Carlo dropout. NNFWI opens a new pathway to combine deep learning and FWI for exploiting the characteristics of deep neural networks and the high accuracy of PDE solvers. Because NNFWI does not require extra training data and optimization loops, it provides an attractive and straightforward alternative to conventional FWI.

58 GEOSCIENCES↗

DOE OSTI2020

Variability and Diversity Load Model Tool [SWR-20-03]

The motivation for the development of this tool and the underlying algorithms and methods was to enable the development of high-temporal resolution, realistic time-series data for quasi-static time-series (QSTS) analysis of distribution systems. Often, aggregated load profile data for a distribution circuit is available (e.g. feeder loading data collected via SCADA at the utility substation) and, while this data is typically accurate it masks the considerable variability of the 100’s or 1000’s of individual loads connected on the circuit. This tool was developed to model both the increased variability expected for these individual loads (e.g. the load of a single distribution transformer connected to 8-12 houses) and the expected diversity between loads on the circuit. It is important to note that the difference in variability and diversity, in the context of this tool, is that variability modeling only adds representative variability due to disaggregated load characteristics (e.g. the presence in the load profile of loads turning off and on like an air conditioner/oven) while the average energy profile remains the same as the user supplied power profile. Diversity modeling generates multiple individual load profiles which, in aggregate, sum to the user supplied power profile. Diversity is effectively variability in the energy usage over longer periods of time than seen in the variability model. Put another way, variability modeling supplies the expected variability due to the operation of various end-use loads and diversity modeling supplies the usage differences due to human behavior, schedules, etc. This load modeling tool was developed for use in generating data for distribution systems. Modeling is summarized by two major functions: 1) taking low resolution load profiles and adding intra-seconds variability onto the profiles, and 2) taking a user supplied load profile and distribution factors and adding both diversity and variability to the user supplied profile.

Zhu, Xiangqi↗