Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “inverse modeling”

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 829 records · Page 46
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Translation and rotation equivariant normalizing flow (TRENF) for optimal cosmological analysis

ABSTRACT Our Universe is homogeneous and isotropic, and its perturbations obey translation and rotation symmetry. In this work, we develop translation and rotation equivariant normalizing flow (TRENF), a generative normalizing flow (NF) model which explicitly incorporates these symmetries, defining the data likelihood via a sequence of Fourier space-based convolutions and pixel-wise non-linear transforms. TRENF gives direct access to the high dimensional data likelihood p(x|y) as a function of the labels y, such as cosmological parameters. In contrast to traditional analyses based on summary statistics, the NF approach has no loss of information since it preserves the full dimensionality of the data. On Gaussian random fields, the TRENF likelihood agrees well with the analytical expression and saturates the Fisher information content in the labels y. On non-linear cosmological overdensity fields from N-body simulations, TRENF leads to significant improvements in constraining power over the standard power spectrum summary statistic. TRENF is also a generative model of the data, and we show that TRENF samples agree well with the N-body simulations it trained on, and that the inverse mapping of the data agrees well with a Gaussian white noise both visually and on various summary statistics: when this is perfectly achieved the resulting p(x|y) likelihood analysis becomes optimal. Finally, we develop a generalization of this model that can handle effects that break the symmetry of the data, such as the survey mask, which enables likelihood analysis on data without periodic boundaries.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

β -delayed neutron emission of Mn 64 , Cr 62 , and Fe 65

The β-decay properties of nuclei near the second nuclear “island of inversion” around neutron rich nuclei with neutron number 40 are important tests of nuclear structure models and interactions. In particular, the β-delayed neutron emission branch (Ρ n ), is useful for investigating β-strength and neutron-γ competition above the neutron separation energies of the daughter nuclei. We report new constraints for Ρ n values for three nuclei in the region: 62 Cr (Ρ n <1%), 64 Mn (Ρ n = 1.5⁢(6)%), and 65 Fe (Ρ n < 1%), measured with the Neutron Emission Ratio Observer (NERO) neutron long counter system and the Beta Counting Station (BCS) at the National Superconducting Cyclotron Laboratory (NSCL). Our results resolve the large discrepancy between previous direct and indirect measurements for 64 Mn and confirm the predictions of global theoretical models when a statistical treatment of the γ and neutron decays of the daughter states is included. Here we also obtain improved half-lives for 62 Cr [206(5) ms] and the short-lived isomer in the 62 Fe daughter [112(7) ms] from β-delayed γ emission data obtained in the same experiment with the Summing NaI (SuN) total absorption spectrometer. Finally, we use γ emission data to obtain a new upper limit for the 62 Cr β-decay population of the long-lived isomeric state in 62 Mn.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2025

Machine learning-driven design and self-sensing capabilities of automotive bumper lattices for adaptive impact response

We present a novel approach to design an automotive bumper energy absorber using carbon fiber reinforced polymer composites, optimized to meet conflicting performance requirements for two distinct impact scenarios. The design must satisfy both a low-speed (2.5 mph) pendulum intrusion test, simulating vehicle-to-vehicle collisions, and a high-speed (25 mph) leg flexion test, replicating pedestrian impacts. These tests demand opposing deformation characteristics: high flexibility (deformation < 85 mm) for the former and high stiffness (deformation < 22 mm) for the latter. To address these contradictory requirements, we developed a machine learning (ML) framework for inverse optimization of lattice designs and material selection. Unlike traditional iterative design processes, our ML model directly outputs optimal design parameters and material choices based on target performance inputs. The energy absorber was fabricated using advanced additive manufacturing techniques, including extrusion deposition and digital light processing. The integration of carbon fibers provides multifunctionality to the bumper structure, enabling self-sensing capabilities through changes in electrical resistivity under compression. This electrical response demonstrates high repeatability under multiple cycles at 2% compression and exhibits distinct signatures during crack formation under high deformation. This research offers adaptive performance through innovative design methodologies and smart material integration. The approach has potential applications in various fields requiring adaptive energy absorption and real-time structural health monitoring.

Chawla, Komal [ORNL] (ORCID:0000000190327565)↗
DOE OSTI2024

Quadrature Based Neural Network Learning of Stochastic Hamiltonian Systems

Hamiltonian Neural Networks (HNNs) provide structure-preserving learning of Hamiltonian systems. In this paper, we extend HNNs to structure-preserving inversion of stochastic Hamiltonian systems (SHSs) from observational data. We propose the quadrature-based models according to the integral form of the SHSs’ solutions, where we denoise the loss-by-moment calculations of the solutions. The integral pattern of the models transforms the source of the essential learning error from the discrepancy between the modified Hamiltonian and the true Hamiltonian in the classical HNN models into that between the integrals and their quadrature approximations. This transforms the challenging task of deriving the relation between the modified and the true Hamiltonians from the (stochastic) Hamilton–Jacobi PDEs, into the one that only requires invoking results from the numerical quadrature theory. Meanwhile, denoising via moments calculations gives a simpler data fitting method than, e.g., via probability density fitting, which may imply better generalization ability in certain circumstances. Numerical experiments validate the proposed learning strategy on several concrete Hamiltonian systems. The experimental results show that both the learned Hamiltonian function and the predicted solution of our quadrature-based model are more accurate than that of the corrected symplectic HNN method on a harmonic oscillator, and the three-point Gaussian quadrature-based model produces higher accuracy in long-time prediction than the Kramers–Moyal method and the numerics-informed likelihood method on the stochastic Kubo oscillator as well as other two stochastic systems with non-polynomial Hamiltonian functions. Moreover, the Hamiltonian learning error εH arising from the Gaussian quadrature-based model is lower than that from Simpson’s quadrature-based model. These demonstrate the superiority of our approach in learning accuracy and long-time prediction ability compared to certain existing methods and exhibit its potential to improve learning accuracy via applying precise quadrature formulae.

Mathematics↗
DOE OSTI2023

Evaluation of dual-weighted residual and machine learning error estimation for projection-based reduced-order models of steady partial differential equations

Projection-based reduced-order models (pROMs) show great promise as a means to accelerate many-query applications such as forward error propagation, solving inverse problems, and design optimization. In order to deploy pROMs in the context of high-consequence decision making, accurate error estimates are required to determine the region(s) of applicability in the parameter space. The following paper considers the dual-weighted residual (DWR) error estimate for pROMs and compares it to another promising pROM error estimate, machine learned error models (MLEM). Here, we show how DWR can be applied to ROMs and then evaluate DWR on two partial differential equations (PDEs): a two-dimensional linear convection–reaction–diffusion equation, and a three-dimensional static hyper-elastic beam. It is shown that DWR is able to estimate errors for pROMs extrapolating outside of their training set while MLEM is best suited for pROMs used to interpolate within the pROM training set.

42 ENGINEERING↗
DOE OSTI2023

Impurity gas monitoring using ultrasonic sensing and neural networks: forward and inverse problems

Ultrasonic sensing is a non-invasive technique for monitoring impurity gas composition in various industrial applications where safety and regulatory compliance are crucial. In this study, ultrasonic sensing and neural networks were used to analyze impurity gases (i.e., air and argon) in helium. An experimental platform was established to acquire ultrasonic data. In the forward problem, an artificial neural network (ANN) model was used to forecast the response and time-of-flight (TOF) based on the excitation, and argon and air concentrations. The inverse problem was solved using a convolutional neural network (CNN) to predict the argon and air concentrations given the ultrasonic response and excitation. The results showed that the ANN accurately predicted the ultrasonic response and the change in TOF with concentration. As the air concentration was increased from 0 to 9.8%, the TOF sensitivity to detect argon decreased by 39.8% and 16.1% from ANN and sound speed theory, respectively. The CNN demonstrated high accuracy in predicting concentrations for inputs in the testing dataset. The application of the trained CNN indicated that it over-predicts air concentration while under-predicting the argon concentration. To improve accuracy, the predicted air and argon concentrations should be corrected by -0.992% and 1.027% bias, respectively.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

INTEGRATE – Inverse Network Transformations for Efficient Generation of Robust Airfoil and Turbine Enhancements

The INTEGRATE (Inverse Network Transformations for Efficient Generation of Robust Airfoil and Turbine Enhancements) project developed a new inverse-design capability for the aerodynamic design of wind turbine rotors using invertible neural networks. Training data was obtained from improved turbulence and transition models for RANS and hybrid RANS/LES solvers with machine-learned physics-based data-augmented corrections and then using the resulting neural-network(s) augmented RANS model to run thousands of 2-D and 3-D CFD simulations.

17 WIND ENERGY↗
DOE OSTI2022

Mean-field model for the Curie-Weiss temperature dependence of coherence length in metallic liquids

The coherence length of the medium-range order (MRO) in metallic liquids is known to display a Curie-Weiss temperature dependence; its inverse is linearly related to temperature, and when extrapolated from temperatures above the glass transition, the coherence length diverges at a negative temperature with a critical exponent of unity. We propose a mean-field pseudospin model that explains this behavior. Specifically, we model the atoms and their local environment as Ising spins with antiferromagnetic exchange interactions. We further superimpose an exchange interaction between dynamical heterogeneities, or clusters of atoms undergoing cooperative motion. The coherence length in the metallic liquid is thus the correlation length between dynamical heterogeneities. Overall, our results reaffirm the idea that the MRO coherence length is a measure of point-to-set correlations, and that local frustrations in the interatomic interactions are prominent in metallic liquids.

36 MATERIALS SCIENCE↗
DOE OSTI2020

PAMTRA 1.0: the Passive and Active Microwave radiative TRAnsfer tool for simulating radiometer and radar measurements of the cloudy atmosphere

Forward models are a key tool to generate synthetic observations given knowledge of the atmospheric state. In this way, they are an integral part of inversion algorithms that aim to retrieve geophysical variables from observations or in data assimilation. Their application for the exploitation of the full information content of remote sensing observations becomes increasingly important when these are used to evaluate the performance of cloud-resolving models (CRMs). Herein, CRM profiles or fields provide the input to the forward model whose simulation results are subsequently compared to the observations. This paper introduces the freely available comprehensive microwave forward model PAMTRA (Passive and Active Microwave TRAnsfer), demonstrates its capabilities to simulate passive and active measurements across the microwave spectral region for upward- and downward-looking geometries, and illustrates how the forward simulations can be used to evaluate CRMs and to interpret measurements to improve our understanding of cloud processes. PAMTRA is unique as it treats passive and active radiative transfer (RT) in a consistent way with the passive forward model providing upwelling and downwelling polarized brightness temperatures and radiances for arbitrary observation angles. The active part is capable of simulating the full radar Doppler spectrum and its moments. PAMTRA is designed to be flexible with respect to instrument specifications and interfaces to many different formats of input and output, especially CRMs, spanning the range from bin-resolved microphysical output to one- and two-moment schemes, and to in situ measured hydrometeor properties. A specific highlight is the incorporation of the self-similar Rayleigh–Gans approximation (SSRGA) for both active and passive applications, which becomes especially important for the investigation of frozen hydrometeors.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Pre-main-sequence Brackett Emitters in the APOGEE DR17 Catalog: Line Strengths and Physical Properties of Accretion Columns

Very young (t ≲ 10 Myr) stars possess strong magnetic fields that channel ionized gas from the interiors of their circumstellar disks to the surface of the star. Upon impacting the stellar surface, the shocked gas recombines and emits hydrogen spectral lines. To characterize the density and temperature of the gas within these accretion streams, we measure equivalent widths of Brackett (Br) 11–20 emission lines detected in 1101 APOGEE spectra of 326 likely pre-main-sequence accretors. For sources with multiple observations, we measure median epoch-to-epoch line strength variations of 10% in Br11 and 20% in Br20. We also fit the measured line ratios to predictions of radiative transfer models by Kwan & Fischer. We find characteristic best-fit electron densities of n e = 10 11 –10 12 cm -3 , and excitation temperatures that are inversely correlated with electron density (from T ~ 5000 K for n e ~ 10 12 cm -3 to T ~ 12,500 K at n e ~ 10 11 cm -3 ). These physical parameters are in good agreement with predictions from modeling of accretion streams that account for the hydrodynamics and radiative transfer within the accretion stream. We also present a supplementary catalog of line measurements from 9733 spectra of 4255 Brackett emission-line sources in the APOGEE Data Release 17 data set.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Evaluation of assumptions made by Hygroscopic Tandem Differential Mobility Analyzer inversion routines

The Hygroscopic Tandem Differential Mobility Analyzer (H-TDMA) measures the hygroscopicity of atmospheric particles, and many atmospheric processes that change this hygroscopicity also change the atmospheric size distribution. Two assumptions made during H-TDMA inversion create spurious hygroscopic trends as a function of the changing inlet size distribution. These two assumptions—that the particles exiting the first Differential Mobility Analyzer (DMA1) are singly charged and that the inlet size distribution has a slope of zero (flat)— generate Multi-Charge Dispersion (MCD) bias and Slope bias, respectively. First, we use a model, named TAO, to show that the inlet size distribution could theoretically change the measured ammonium sulfate hygroscopicity by 10%–20% as a function of diameter or experimental time with no change in relative humidity. Secondly, we show experimentally that aerosol emitted from the flaming combustion of grass creates MCD bias. Here, in this experiment, we measure the CPC response of the first three charges and invert these responses using a new routine named Junior. Junior's inversion of each charge shows that one growth factor distribution describes all measured diameters (no growth dependence on diameter). As in the modeling study above, previous publications of this aerosol system, using traditional inversion assumptions, report a decrease in hygroscopicity as DMA1 diameter increases. Unlike traditional inversions, Junior's inversion does not assume the particles are singly charged nor does it make the flat inlet size distribution assumption. Instead, both the inlet size distribution and each charge's CPC response are measured quantities. Thus, the discrepancy between our inversion results and previous publications is likely due to the traditional inversion routine assumptions. This underscores the importance for accounting for Slope and MCD bias during inversions. Experimental results should be carefully analyzed when reporting hygroscopic trends with respect to diameter or experimental time when using the traditional inversion assumptions.

42 ENGINEERING↗
DOE OSTI2023

Exploring for Superhot Geothermal Targets in Magmatic Settings: 2022 Field Campaign at Newberry Volcano

This paper presents preliminary results from a subset of work carried out as part of a multinational research project entitled DErisking Exploration for multiple geothermal Plays in magmatic ENvironments (DEEPEN), supported by the U.S. Department of Energy (DOE) and Geothermica, a joint effort by EU member states and associated countries. The DEEPEN project will develop a customized approach to exploration for supercritical and superhot geothermal plays in magmatic systems, which will be applied to two demonstration sites. This paper summarizes field activities carried out at the U.S. demonstration site, Newberry Volcano in central Oregon. The objective of this work effort is to refine the subsurface model of Newberry Volcano, with special focus on deeper zones including the magmatic plumbing system and other key geologic elements. New data collection included gravity and wideband magnetotelluric (MT) surveys, as well as reinstallation of a seismic network. The National Renewable Energy Laboratory (NREL) and Enthalpion Energy LLC (Enthalpion) worked with the Deschutes National Forest Fort Rock District to use a low ground disturbance method of MT deployment to collect MT data inside the caldera and other restricted areas inside the National Volcanic Monument. This opened these areas to geophysical exploration for the first time in decades. Sites along and adjacent to the south rim of the volcano constituted the primary survey objectives. A team from Lawrence Berkeley National Laboratory (LBNL), the U.S. Geological Survey (USGS), and AltaRock also began the process of reinstalling the seismic network from the AltaRock enhanced geothermal system (EGS) demonstration in anticipation of further development activities at the site. The data ingestion, reduction, and analysis phase of the project is ongoing. We are currently processing the MT and gravity data and are developing a new, highly GPU-accelerated, 3D joint MT and gravity inversion to better localize the south rim/south flank conductive target and better understand its relationship to deep heat, fluid sources, and surface extrusive features. Joint inversions, which have not yet been undertaken at Newberry, will allow us to obtain constraints on the geologic model that cannot be determined from each method in isolation, improving our ability to image key geologic features at depth.

geophysics↗
DOE OSTI2023

Limits on Leptonic TeV Emission from the Cygnus Cocoon with Swift-XRT

$\gamma$-ray observations of the Cygnus Cocoon, an extended source surrounding the Cygnus X star-forming region, suggest the presence of a cosmic ray accelerator reaching energies up to a few PeV. The very-high-energy (VHE; 0.1-100~TeV) $\gamma$-ray emission may be explained by the interaction of cosmic-ray hadrons with matter inside the Cocoon, but an origin of inverse Compton radiation by relativistic electrons cannot be ruled out. Inverse Compton $\gamma$-rays at VHE are accompanied by synchrotron radiation peaked in X-rays. Hence, X-ray observations may probe the electron population and magnetic field of the source. We observed eleven fields in or near the Cygnus Cocoon with the Neil Gehrels Swift Observatory's X-Ray Telescope (Swift-XRT) totaling 110 ksec. We fit the fields to a Galactic and extra-galactic background model and performed a log-likelihood ratio test for an additional diffuse component. We found no significant additional emission and established upper limits in each field. By assuming that the X-ray intensity traces the TeV intensity and follows an dN/dE ∝ E -2.5 spectrum, we obtained a 90% upper limit of F X < 8.7 x 10 -11 erg cm -2 s -1 or 5.2 x10 -11 erg cm -2 s -1 on the X-ray flux of the entire Cygnus Cocoon between 2 and 10 keV depending on the choice of hydrogen column density model. This suggests that no more than one quarter of the $\gamma$-ray flux at 1 TeV is produced by inverse Compton scattering, when assuming an equipartition magnetic field of ~20 μG.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Fusion burn-propagation simulations using the collisional and radiative particle-in-cell code TRIFORCE

The ability to accurately model burn propagation in inertial confinement fusion plasmas is crucial for advancing fusion energy research. This work presents enhancements to the triforce hybrid fluid-kinetic multiphysics code, focusing on its kinetic half, which employs the particle-in-cell (PIC) method with Monte Carlo collisions (MCC). We use a moment-preserving collision model that mitigates numerical noise, particularly in spherical geometries where particle weights vary significantly. Additionally, we refine the treatment of inverse bremsstrahlung to account for electron–ion collision frequency reductions in degenerate plasmas and incorporate a blackbody radiation source to enable realistic photon injection. These improvements enable the simulation of 1-dimensional (1D) spherical fusion burn propagation in deuterium–tritium plasmas. Benchmark comparisons with the hydra radiation-hydrodynamics code confirm that triforce accurately captures the dynamics of hot-spot expansion and burn propagation, demonstrating sensitivity to ignition thresholds consistent with theoretical models. Findings show the ignition cliff to be less steep in our work compared to radiation-hydrodynamic modeling. These results highlight the role of kinetic effects in fusion ignition physics and underscore the necessity of hybrid fluid-kinetic models for advancing predictive capabilities in high-energy-density plasma systems.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2023

Diffusion Probabilistic Modeling for Video Generation

Denoising diffusion probabilistic models are a promising new class of generative models that mark a milestone in high-quality image generation. This paper showcases their ability to sequentially generate video, surpassing prior methods in perceptual and probabilistic forecasting metrics. We propose an autoregressive, end-to-end optimized video diffusion model inspired by recent advances in neural video compression. The model successively generates future frames by correcting a deterministic next-frame prediction using a stochastic residual generated by an inverse diffusion process. We compare this approach against six baselines on four datasets involving natural and simulation-based videos. We find significant improvements in terms of perceptual quality and probabilistic frame forecasting ability for all datasets.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

LATTE: Los Alamos TravelTime package based on Eikonal equation

This Fortran code focuses on traveltime computation and tomography based on eikonal equation. Specifically, the package provides three major functionalities: (1) forward modeling of traveltime from single-point or ensemble source based on factorized eikonal equation, (2) adjoint-state first-arrival traveltime tomography based on picked first arrival traveltime using steepest descent, conjugate gradient, or limited-memory BFGS inversion scheme, and (3) adjoint-state joint transmission-reflection tomography based on picked first-arrival and reflection traveltimes. The package applies to forward modeling and tomography based on traveltime in 2D and 3D isotropic regular-grid models. We name this package LATTE – Los Alamos TravelTime package based on Eikonal equation. * The code is for accompanying a journal paper under preparation. The paper will be submitted via LA-UR separately later.

Gao, Kai↗
DOE OSTI2023

Double-Atom Catalysts Featuring Inverse Sandwich Structure for CO 2 Reduction Reaction: A Synergetic First-Principles and Machine Learning Investigation

Electrocatalytic CO 2 reduction reactions (CO 2 RR) based on scalable and highly efficient catalysis provide an attractive strategy for reducing CO 2 emissions. Here in this work, we combined first-principles density functional theory (DFT) and machine learning (ML) to comprehensively explore the potential of double-atom catalysts (DACs) featuring an inverse sandwich structure anchored on defective graphene (gra) to catalyze CO 2 RR to generate C 1 products. We started with five homonuclear M 2 ⊥gra (M = Co, Ni, Rh, Ir, and Pt), followed by 127 heteronuclear MM'⊥gra (M = Co, Ni, Rh, Ir, and Pt, M' = Sc–Au). Stable DACs were screened by evaluating their binding energy, formation energy, and dissolution potential of metal atoms, as well as conducting first-principles molecular dynamics simulations with and without solvent water molecules. Based on DFT calculations, Rh 2 ⊥gra DAC was found to outperform the other four homonuclear DACs and the Rh-based single- and double-atom catalysts of noninverse sandwich structures. Out of the 127 heteronuclear DACs, 14 were found to be stable and have good catalytic performance. An ML approach was adopted to correlate key factors with the activity and stability of the DACs, including the sum of radii of metal and ligand atoms (d M–M' , d M–C , and d M'–C ), the sum and difference of electronegativity of two metal atoms (P M + P M' , P M – P M '), the sum and difference of first ionization energy of two metal atoms (I M + I M' , I M – I M '), the sum and difference of electron affinity of two metal atoms (A M + A M' , A M – A M '), and the number of d-electrons of the two metal atoms (Nd). The obtained ML models were further used to predict 154 potential electrocatalysts out of 784 possible DACs featuring the same inverse sandwich configuration. Overall, this work not only identified promising CO 2 RR DACs featuring the reported inverse sandwich structure but also provided insights into key atomic characteristics associated with high CO 2 RR activity.

30 DIRECT ENERGY CONVERSION↗
DOE OSTI2026

Generative Models for Crystalline Materials

Understanding structure-property relationships in materials is fundamental in condensed matter physics and materials science. Over the past few years, machine learning (ML) has emerged as a powerful tool for advancing this understanding and accelerating materials discovery. Early ML approaches primarily focused on constructing and screening large material spaces to identify promising candidates for various applications. More recently, research efforts have increasingly shifted toward generating crystal structures using end-to-end generative models. This review analyzes the current state of generative modeling for crystal structure prediction and de novo generation. It examines crystal representations, outlines the generative models used to design crystal structures, and evaluates their respective strengths and limitations. Furthermore, the review highlights experimental considerations for evaluating generated structures and provides recommendations for suitable existing software tools. Emerging topics, such as modeling disorder and defects, integration in advanced characterization, incorporating synthetic feasibility constraints, and model explainability are explored. Ultimately, this work aims to inform both experimental scientists looking to adapt suitable ML models to their specific circumstances and ML specialists seeking to understand the unique challenges related to inverse materials design and discovery.

Metni, Houssam [Karlsruhe Inst. of Technology (KIT↗