Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “inference accelerators”

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 199 records · Page 11
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Causes of fuel–ablator mix inferred from modeling of monochromatic time-gated radiography of OMEGA cryogenic implosions

Here, we present evidence, in the context of OMEGA cryogenic target implosions, that laser imprint, known to be capable of degrading laser-direct-drive target performance, plays a major role in generating fuel–ablator mix. OMEGA cryogenic target implosions show a performance boundary correlated with acceleration-phase shell stability; for sufficiently low adiabats (where the adiabat is the ratio of the pressure to the Fermi pressure) and high in-flight aspect ratios (IFAR's), the neutron-weighted shell areal density and neutron yield relative to the clean simulated values sharply decline. Direct evidence of Rayleigh–Taylor fuel–ablator mixing was previously obtained using a Si Heα backlighter driven by an ~20-ps short pulse generated by OMEGA EP. The shadow cast by the shell shortly prior to stagnation, as diagnosed using backlit radiographs, shows a softening near the limb, which is evidence of an ablator–fuel mix region for a low-adiabat implosion (α ~ 1.9, IFAR = 14) but not for a moderate adiabat implosion (α ~ 2.5, IFAR = 10). We find good agreement between experimental and synthetic radiographs in simulations that model laser imprint and account for uncertainty in the initial ablator thickness. Furthermore, we further explore the role of other mechanisms such as classical instability growth at the fuel–ablator interface, species concentration diffusion, and long-wavelength drive and target asymmetries.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2026

Strong Lensing Cosmology with Population-level Calibrated Neural Ratio Estimation

Strong gravitational lensing contains key information about cosmic acceleration. Modern and next-generation galaxy imaging surveys are expected to provide high-quality data on $\mathcal{O}(10^5)$ galaxy-galaxy lensing systems. The plethora and complexity of the data are likely to present computational challenges for parameter inference methods for fitting high-dimensional likelihoods, which are often analytically intractable. Neural Ratio Estimation (NRE) efficiently computes individual likelihood ratios that can be combined into population-level posteriors. We use simulations to study the capacity of NRE to jointly predict the dark energy equation-of-state parameter $w$ and the total matter density $Ω_{m}$ from lensing images and companion spectroscopic information. We also introduce a post hoc posterior coverage calibration procedure that mitigates the model overconfidence that is typically found in neural density estimation applications. Our experiments show that the errors on both parameters decrease with increasing inference population sizes. In particular, for 100 lenses in a standard $Λ$CDM Universe, our calibrated NRE model achieves median fractional uncertainty of $22.8\%$ in $w$ and $2.9\%$ in $Ω_{m}$. This proof of concept demonstrates a potentially scalable approach for efficient cosmological parameter inference with large populations of galaxy-scale lenses observed in future surveys.

Jarugula, Sreevani [Fermilab] (ORCID:0000000253867↗
DOE OSTI2022

A Framework for Neural Network Inference on FPGA-Centric SmartNICs

FPGA-based SmartNICs offer great potential to significantly improve the performance of high-performance computing and warehouse data processing by tightly coupling support for reconfigurable data-intensive computation with cross-node communication, thereby mitigating the von Neumann bottleneck. Existing work, however, has been generally been limited in that it assumes an accelerator model where kernels are offloaded to SmartNICs, but most control tasks are left to the CPUs. This leads to frequent waiting, inferior performance, and scaling challenges. In this work, we propose a new distributive data-centric computing framework, named FCsN, for reconfigurable SmartNIC-based systems. Through a lightweight task circulation execution model and its implementation architecture, FCsN allows the complete detaching of kernel execution, control logic, system scheduling, and network communication to the SmarNICs. This boosts performance by: (i) avoiding the control dependency with CPUs and (ii) supporting streaming kernel execution and network communication at line rate and in a very fine-grained manner. We demonstrate the efficiency and flexibility of FCsN using various types of neural network applications including graph neural networks; as these last are both irregular and data intensive they offer an especially robust demonstration. Evaluations using commonly-used neural network models and graph datasets show that a system with the support of FCsN can achieve, on average, 144 speedups over the MPI-based standard CPU baselines.

Guo, Anqi↗
DOE OSTI2023

Laser Wakefield Accelerator modelling with Variational Neural Networks

A machine learning model was created to predict the electron spectrum generated by a GeV-class laser wakefield accelerator. The model was constructed from variational convolutional neural networks, which mapped the results of secondary laser and plasma diagnostics to the generated electron spectrum. An ensemble of trained networks was used to predict the electron spectrum and to provide an estimation of the uncertainty of that prediction. It is anticipated that this approach will be useful for inferring the electron spectrum prior to undergoing any process that can alter or destroy the beam. In addition, the model provides insight into the scaling of electron beam properties due to stochastic fluctuations in the laser energy and plasma electron density.

43 PARTICLE ACCELERATORS↗
DOE OSTI2020

Geophysical Methods for Stratigraphic Identification

The goal of this work is to identify geophysical methods that can be used to delineate (in 3D) a large-scale paleochannel that has been inferred to exist between the 200 West and 200 East Areas of the Hanford Site’s Central Plateau . This paleochannel is of particular importance because it will significantly impact flow and contaminant transport from 200 West to 200 East. Its high permeability can both accelerate contaminant transport and lower contaminant concentrations through spreading. Information about the spatial extent and structure of such features is needed to inform subsurface flow and contaminant transport models that support remedy decisions at Hanford.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2020

Design of a multichannel photonic crystal dielectric laser accelerator

To be useful for most scientific and medical applications, compact particle accelerators will require much higher average current than enabled by current architectures. For this purpose, we propose a photonic crystal architecture for a dielectric laser accelerator, referred to as a multi-input multi-output silicon accelerator (MIMOSA), that enables simultaneous acceleration of multiple electron beams, increasing the total electron throughput by at least 1 order of magnitude. To achieve this, we show that the photonic crystal must support a mode at the Γ point in reciprocal space, with a normalized frequency equal to the normalized speed of the phase-matched electron. We show that the figure of merit of the MIMOSA can be inferred from the eigenmodes of the corresponding infinitely periodic structure, which provides a powerful approach to design such devices. Additionally, we extend the MIMOSA architecture to electron deflectors and other electron manipulation functionalities. These additional functionalities, combined with the increased electron throughput of these devices, permit all-optical on-chip manipulation of electron beams in a fully integrated architecture compatible with current fabrication technologies, which opens the way to unconventional electron beam shaping, imaging, and radiation generation.

43 PARTICLE ACCELERATORS↗
DOE OSTI2025

Bayesian calibration of bubble size dynamics applied to CO 2 gas fermenters

To accelerate the scale-up of gaseous CO 2 fermentation reactors, computational models need to predict gas-to-liquid mass transfer which requires capturing the bubble size dynamics, i.e. bubble breakup and coalescence. However, the applicability of existing models beyond air–water mixtures remains to be established. Here, an inverse modeling approach, accelerated with a neural network surrogate, calibrates the breakup and coalescence closure models, that are used in class methods for population balance modeling (PBM). The calibration is performed based on experimental results obtained in a CO 2 -air–water-coflowing bubble column reactor. Bayesian inference is used to account for noise in the experimental dataset and bias in the simulation results. To accurately capture gas holdup and interphase mass transfer, the results show that the breakage rate needs to be increased by one order of magnitude. In conclusion, the inferred model parameters are then used on a separate configuration and shown to also improve bubble size distribution predictions.

09 BIOMASS FUELS↗
DOE OSTI2024

Rapid assessment of the creep rupture life of metals: A model enabling experimental design

Prediction of the creep rupture life of engineering metals is critical for qualification and design of new materials. The use of long-term creep tests and the need to quantify the performance variability in a priori similar systems hinder the rapid creep assessment of a given material. Therefore, it is essential to develop methods that can extrapolate the long-term performance of alloys and the associated variability from short-term experiments. To this end, this study introduces a new model which enables the estimation of the rupture life of a material for a given stress and temperature. This model relies on two components. First, a new relation for the minimum creep rate (MCR) of materials is introduced. It includes a stress dependent stress exponent allowing the model to capture the variation of MCR across a wide range of temperatures and stresses. Second, employing the Monkman-Grant (MG) law, we establish a relation between stress, temperature and creep rupture life. Together, these two elements yield a new closed-form mathematical expression for the Larson Miller parameter as a function of stress and temperature. This expression captures the creep rupture time for many metals (Gr91, Copper, Gr122 and 347H) and compares favorably with alternate empirical approaches. The model is then used to assess the minimum duration of creep rates necessary to qualify the material up to 100000h. Furthermore, it is found that depending on the material system, creep tests as few as five limited to 5000 h for steels (Gr91, Gr122, 347H) and 100 h for copper are sufficient to model creep lifetimes. Finally, using a Bayesian inference-based approach to calibrate the model, we demonstrate that variability in rupture life can be captured via the quantification of the uncertainty in the model parameters and extrapolated from a limited number of short to moderately short creep tests; thereby paving the way for accelerated creep testing.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Harnessing ML Privacy by Design Through Crossbar Array Non-idealities

Deep Neural Networks (DNNs), handling computeand data-intensive tasks, often utilize accelerators like Resistiveswitching Random-access Memory (RRAM) crossbar for energyefficient in-memory computation. Despite RRAM’s inherent nonidealities causing deviations in DNN output, this study transforms the weakness into strength. By leveraging RRAM non-idealities, the research enhances privacy protection against Membership Inference Attacks (MIAs), which reveal private information from training data. RRAM non-idealities disrupt MIA features, increasing model robustness and revealing a privacy-accuracy tradeoff. Empirical results with four MIAs and DNNs trained on different datasets demonstrate significant privacy leakage reduction with a minor accuracy drop (e.g., up to 2.8% for ResNet-18 with CIFAR-100).

artificial intelligence↗
DOE OSTI2025

MOOSE ProbML: Parallelizable Probabilistic Machine Learning and Uncertainty Quantification Capabilities

The Multiphysics Object Oriented Simulation Environment (MOOSE) is a widely used open- source finite element software for performing multiphysics multiscale simulations in a massively parallel fashion. Recently, the computational team at Idaho National Laboratory (INL) has implemented Probabilistic Machine Learning (ProbML) capabilities in MOOSE—in a parallelized fashion—and enable active learning with large-scale computational models for tasks such as surrogate model development, scale bridging, forward/inverse uncertainty quantification (UQ), Bayesian optimization, etc. This presentation summarizes these developments in MOOSE along with demonstrations on several real applications relevant to nuclear energy. At the fundamental level, samplers like Monte Carlo/Latin Hypercube, variance reduction, parallelized Markov Chain Monte Carlo (MCMC) support uncertainty propagation in both forward and inverse settings. These samplers can be integrated with the Gaussian processes (GP) suite in MOOSE, which offer several variants like scalar GPs, multi-output GPs, and deep GPs, to enable active learning. These GPs can be tuned using gradient-based optimization methods like Adam and its variants or gradient-free methods like the elliptical slice sampler (a variant of MCMC adept under Gaussian settings) for more complex covariance kernels or likelihoods whose gradient computations can be cumbersome. A variety of batch acquisition functions permit parallelized evaluation of the computational model and support different learning objectives with high efficiency like Bayesian inference, global surrogate development, optimization, etc. Furthermore, libtorch integration supports training, evaluation, and re-training of neural networks and other complex machine learning models in active learning settings. The impacts of these developments are shown on several real applications: (1) nuclear fuel inverse UQ and model inadequacy assessment using the Kennedy O’Hagan framework; (2) uncertainty aware surrogate modeling for additive manufacturing to predict field quantities; (3) nuclear reactor rare events analysis; and (4) complex fluid flow prediction using a global surrogate with quantified prediction uncertainty. Finally, the outlook of MOOSE ProbML is discussed for both outer-loop and inner-loop computations in the broad view to accelerate fuels and materials qualification, address gaps in knowledge and data, and assess new reactor/fuel systems.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

Evaluation of data driven low-rank matrix factorization for accelerated solutions of the Vlasov equation

Low-rank methods have shown success in accelerating simulations of a collisionless plasma described by the Vlasov equation, but still rely on computationally costly linear algebra every time step. We propose a data-driven factorization method using artificial neural networks, specifically with convolutional layer architecture, that trains on existing simulation data. At inference time, the model outputs a low-rank decomposition of the distribution field of the charged particles, and we demonstrate that this step is faster than the standard linear algebra technique. Numerical experiments show that the method achieves comparable reconstruction accuracy for interpolation tasks, generalizing to unseen test data in a manner beyond just memorizing training data; patterns in factorization also inherently followed the same numerical trend as those within algebraic methods (e.g., truncated singular-value decomposition). However, when training on the first 70% of a time-series data and testing on the remaining 30%, the method fails to meaningfully extrapolate. Despite this limiting result, the technique may have benefits for simulations in a statistical steady-state or otherwise showing temporal stability. These results suggest that while the model offers a computationally efficient alternative for datasets with temporal stability, its current formulation is best suited for interpolation rather than for predicting future states in time-evolving systems. This study thus lays the groundwork for further refinement of neural network-based approaches to low-rank matrix factorization in high-dimensional plasma simulations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

DeepLensSBI: Deep inference of simulated strong lenses in ground-based surveys

This code is used to train and test machine learning models and generate results and plots presented in 2501.08524 [astro-ph.IM]. The code is written in python. The goal of this work is to train ML models trained on simulated images of strong gravitational lenses. The trained model can then quickly infer properties of the lensed objects with uncertainty quantification.

Poh, Jason [Univ. of Chicago, IL (United States)] ↗
DOE OSTI2024

A Strong Gravitational Lens Is Worth a Thousand Dark Matter Halos: Inference on Small-scale Structure Using Sequential Methods

Strong gravitational lenses are a singular probe of the Universe’s small-scale structure—they are sensitive to the gravitational effects of low-mass (<10 10 M ⊙ ) halos even without a luminous counterpart. Recent strong-lensing analyses of dark matter structure rely on simulation-based inference (SBI). Modern SBI methods, which leverage neural networks as density estimators, have shown promise in extracting the halo-population signal. However, it is unclear whether the constraints from these models are limited by the methodology or the data. In this study, we introduce an accelerator-optimized simulation pipeline that can generate lens images with realistic subhalo populations in milliseconds. Leveraging this simulator, we identify the main limitation of our fiducial SBI analysis: training set size. We then adopt a sequential neural posterior estimation (SNPE) approach, allowing us to refine the training distribution to align with the observed data. Using only one-fifth as many mock Hubble Space Telescope images, SNPE matches the constraints on the low-mass halo population produced by our best nonsequential model. Our experiments suggest that an over 3 order-of-magnitude increase in training set size and GPU hours would be required to achieve an equivalent result without sequential methods. While the full potential of the existing lens sample remains to be explored, the notable improvement in constraining power enabled by our sequential approach highlights that current constraints are limited primarily by methodology and not the data itself. Moreover, our results emphasize the need to treat training set generation and model optimization as interconnected stages of any cosmological analysis using SBI.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2026

A Nonparametric Method for the Inference of Halo Occupation Distributions

The galaxy–halo connection traces processes by which galaxies form and evolve. The halo occupation distribution (HOD) describes the relationship between galaxies and their host dark matter haloes. Measurements of the galaxy two-point correlation function (2PCF) allow us to extract information about the HODs of observed galaxy samples. Several parametric HOD models have been proposed in the literature, but the choice of parameterization restricts the space of possible HODs. To resolve this issue, we introduce a nonparametric HOD fitting method in which we train an emulator to learn the mappings among the galaxy 2PCF, physical properties used to select galaxy samples, and the HOD, all obtained from simulated past light cones constructed with the Santa Cruz semianalytic model. Implementing this emulator within a likelihood analysis framework, we derive constraints on the HOD of a galaxy sample when provided with a measurement of its 2PCF. Using the emulator to accelerate likelihood evaluations, we test the nonparametric HOD approach on a set of 2PCFs for mock galaxy samples drawn from the TNG100-1 simulation and selected above threshold values of stellar mass and star formation rate. Our framework is able to recover TNG100-1 HODs within 0.2 dex. We use the TNG100-1 mocks to tune the reported uncertainties to estimate those expected in the analysis of observations. Comparing to parametric HOD modelling routines applied to the same mock galaxy samples, our approach consistently infers the HOD with comparable or greater precision and accuracy.

Kennedy, Jacob [Rutgers Univ., Piscataway, NJ (Uni↗
DOE OSTI2020

Effect of bars on evolution of SDSS spiral galaxies

ABSTRACT We explore the significance of bars in triggering central star formation (SF) and active galactic nucleus (AGN) activity for spiral galaxy evolution using a volume-limited sample with 0.020 < z < 0.055, Mr < −19.5, and σ > 70 km s−1 selected from Sloan Digital Sky Survey Data Release 7. On a central SF rate–σ plane, we measure the fraction of galaxies with strong bars in our sample and also the AGN fractions for barred and non-barred galaxies, respectively. The comparison between the bar and AGN fractions reveals a causal connection between the two phenomena of SF quenching and AGN activity. A massive black hole and abundant gas fuels are sufficient conditions to trigger AGNs. We infer that the AGNs triggered by satisfying the two conditions drive the strong AGN feedback, suddenly suppressing the central SF and leaving the SF sequence. We find that in galaxies where either of the two conditions is not sufficient, bars are a great help for the AGN triggering, accelerating the entire process of evolution, which is particularly evident in pseudo-bulge galaxies. All of our findings are obtained only when plotted in terms of their central velocity dispersion and central SFR (not galactic scale SFR), indicating that the AGN-driven SF quenching is confined in the central kpc region.

Kim, Minbae↗
DOE OSTI2025

Materials Learning Algorithms (MALA): Scalable machine learning for electronic structure calculations in large-scale atomistic simulations

We present the Materials Learning Algorithms (MALA) package, a scalable machine learning framework designed to accelerate density functional theory (DFT) calculations suitable for large-scale atomistic simulations. Using local descriptors of the atomic environment, MALA models efficiently predict key electronic observables, including local density of states, electronic density, density of states, and total energy. The package integrates data sampling, model training and scalable inference into a unified library, while ensuring compatibility with standard DFT and molecular dynamics codes. We demonstrate MALA's capabilities with examples including boron clusters, aluminum across its solid-liquid phase boundary, and predicting the electronic structure of a stacking fault in a large beryllium slab. Scaling analyses reveal MALA's computational efficiency and identify bottlenecks for future optimization. With its ability to model electronic structures at scales far beyond standard DFT, MALA is well suited for modeling complex material systems, making it a versatile tool for advanced materials research.

Density functional theory↗
DOE OSTI2021

Measurement of U 238 ( n , n ' γ ) cross section data and their impact on reaction models

A better knowledge of (n,xn) reaction cross sections is important for both reaction modeling and energy applications. This article focuses on inelastic scattering of neutrons off 238 U for which improvements are needed to better constrain evaluations and solve inconsistencies in nuclear power reactor calculations. A new precise measurement of (n,xnγ) reaction cross sections on 238 U has been performed at the GELINA (Geel Electron LINear Accelerator) neutron facility operated by EC-JRC-Geel (Belgium) with the GRAPhEME (GeRmanium array for Actinides PrEcise MEasurements) setup. The prompt γ-ray spectroscopy method coupled to time-of-flight measurements is used to extract (n,xnγ) cross section values which can be further combined to infer the total neutron inelastic scattering cross section. Cross section data for 18 γ transitions (five never measured before) are presented and compared to the data in the literature. Here, emphasis is especially given to the uncertainty determination to produce partial cross section data as accurate as possible. Due to intrinsic limitations of the experimental method, the use of additional nuclear structure information coupled with theoretical modeling is required to determine the total (n, n') cross section over the whole neutron energy range. We have investigated modeling aspects of the 238 U(n,n'γ) cross sections related to the description of compound nucleus and preequilibirum mechanisms as well as the discrete part of nuclear structure. Through comparison between experimental and calculated (n,n'γ) cross sections, we pinpoint inaccuracies in the description of specific reaction mechanisms and challenge recently implemented models. This helps improving the whole modeling of the (n, n') reaction.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2024

Hydra: computer vision for data quality monitoring

Hydra, initially developed for Hall-D in 2019, is a system that utilizes computer vision to perform near real time data quality monitoring. Since then, it has been deployed across all experimental halls at Jefferson Lab, with the CLAS12 collaboration in Hall-B being the first outside of GlueX to fully utilize Hydra. The system comprises back end processes that manage the models, their inferences, and the data flow. Finally, the front-end components, accessible via web pages, allow detector experts and shift crews to view and interact with the system.

47 OTHER INSTRUMENTATION↗