Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “deep 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.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 523 records · Page 29
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Automatic Detection of Defects in High-Reliability Components

Disastrous consequences can result from defects in manufactured parts—particularly the high consequence parts developed at Sandia. Identifying flaws in as-built parts can be done with nondestructive means, such as X-ray Computed Tomography (CT). However, due to artifacts and complex imagery, the task of analyzing the CT images falls to humans. Human analysis is inherently unreproducible, unscalable, and can easily miss subtle flaws. We hypothesized that deep learning methods could improve defect identification, increase the number of parts that can effectively be analyzed, and do it in a reproducible manner. We pursued two methods: 1) generating a defect-free version of a scan and looking for differences (PandaNet), and 2) using pre-trained models to develop a statistical model of normality (Feature-based Anomaly Detection System: FADS). Both PandaNet and FADS provide good results, are scalable, and can identify anomalies in imagery. In particular, FADS enables zero-shot (training-free) identification of defects for minimal computational cost and expert time. It significantly outperforms prior approaches in computational cost while achieving comparable results. FADS’ core concept has also shown utility beyond anomaly detection by providing feature extraction for downstream tasks.

47 OTHER INSTRUMENTATION↗
DOE OSTI2023

From Femtoseconds to Gigaseconds: The SolDeg Project to Analyze and Mitigate the Performance Degradation of a-Si:H/c-Si Heterojunction Solar Cells

We pursued two major projects. In the first, we studied the degradation of a-Si/c-Si Heterojunction cells. We substantially expanded our previously developed SolDeg platform, and incorporated hydrogen into it. We also expanded the capabilities of SolDeg by developing a three channel model for a very efficient way to determine long term junction dynamics. The first main result was that we identified hydrogen drifting away from the interface as a key degradation channel. The second was that we developed and proposed the concept of a “reverse silicon density gradient” in the a-Si layer that had the promise of reducing degradation by up to 80%. In the second project, we studied the degradation of TOPCon solar cells. We again identified a key degradation channel, which in these cells was the formation of pinholes across the oxide layer. We then again were able to develop and to propose a way to dramatically reduce degradation by discovering that tuning the hydrogen concentration forms a sharp and deep minimum of the recombination current.

14 SOLAR ENERGY↗
DOE OSTI2020

Evaluation of low-exergy heating and cooling systems and topology optimization for deep energy savings at the urban district level

District energy systems have the potential to achieve deep energy savings by leveraging the density and diversity of loads in urban districts. However, planning and adoption of district thermal energy systems is hindered by the analytical burden and high infrastructure costs. It is hypothesized that network topology optimization would enable wider adoption of advanced (ambient temperature) district thermal energy systems, resulting in energy savings. In this study, energy modeling is used to compare the energy performance of “conventional” and “advanced” district thermal energy systems at the urban district level, and a partial exhaustive search is used to evaluate a heuristic for the topology optimization problem. For the prototypical district considered, advanced district thermal energy systems mated with low-exergy building heating and cooling systems achieved a source energy use intensity that was 49% lower than that of conventional systems. The minimal spanning tree heuristic was demonstrated to be effective for the network topology optimization problem in the context of a prototypical district, and contributes to mitigating the problem’s computational complexity. The work presented in this paper demonstrates the potential of advanced district thermal energy systems to achieve deep energy savings, and advances to addressing barriers to their adoption through topology optimization.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2026

Accelerating Traction Motor Optimization Design with AI Surrogate Models

The advancement of artificial intelligence systems enables the use of data-driven physics-based surrogate models to explore design spaces rapidly and deeply for engineering projects. This work presents a surrogate model workflow that accelerates electric traction motor design optimization by replacing finite element analysis (FEA) with an artificial neural network (ANN) and using this model in a genetic algorithm for design optimization. A baseline interior permanent-magnet motor is parameterized and sampled to generate FEA-labeled training data, after which a feed-forward ANN predicts key outputs (e.g., loss components and weight). The validated surrogate enables genetic-algorithm optimization and deep search over the design space without new FEA runs, producing Pareto-optimal trade-offs between weight and losses and set of optimized designs for rapid downselection of manufacturable motor designs.

Ribeiro, Pedro [ORNL] (ORCID:0009000921026641)↗
DOE OSTI2026

Constraining the Stellar-to-Halo Mass Relation with Galaxy Clustering and Weak Lensing from DES Year 3 Data

We develop a framework to study the relation between the stellar mass of a galaxy and the total mass of its host dark matter halo using galaxy clustering and galaxy-galaxy lensing measurements. We model a wide range of scales, roughly from $\sim 100 \; {\rm kpc}$ to $\sim 100 \; {\rm Mpc}$, using a theoretical framework based on the Halo Occupation Distribution and data from Year 3 of the Dark Energy Survey (DES) dataset. The new advances of this work include: 1) the generation and validation of a new stellar mass-selected galaxy sample in the range of $\log M_\star/M_\odot \sim 9.6$ to $\sim 11.5$; 2) the joint-modeling framework of galaxy clustering and galaxy-galaxy lensing that is able to describe our stellar mass-selected sample deep into the 1-halo regime; and 3) stellar-to-halo mass relation (SHMR) constraints from this dataset. In general, our SHMR constraints agree well with existing literature with various weak lensing measurements. We constrain the free parameters in the SHMR functional form $\log M_\star (M_h) = \log(εM_1) + f\left[ \log\left( M_h / M_1 \right) \right] - f(0)$, with $f(x) \equiv -\log(10^{αx}+1) + δ[\log(1+\exp(x))]^γ/ [1+\exp(10^{-x})]$, to be $\log M_1 = 11.506^{+0.325}_{-0.404}$, $\log ε= -1.632^{+0.306}_{-0.181}$, $α= -1.638^{+0.108}_{-0.099}$, $γ= 0.596^{+0.251}_{-0.210}$ and $δ= 3.810^{+2.045}_{-1.811}$. The inferred average satellite fraction is within $\sim 5-35\%$ for our fiducial results and we do not see any clear trends with redshift or stellar mass. Furthermore, we find that the inferred average galaxy bias values follow the generally expected trends with stellar mass and redshift. Our study is the first SHMR in DES in this mass range, and we expect the stellar mass sample to be of general interest for other science cases.

Zacharegkas, G. [Argonne; Chicago U., KICP] (ORCID↗
DOE OSTI2023

DL-TODA: A Deep Learning Tool for Omics Data Analysis

Metagenomics is a technique for genome-wide profiling of microbiomes; this technique generates billions of DNA sequences called reads. Given the multiplication of metagenomic projects, computational tools are necessary to enable the efficient and accurate classification of metagenomic reads without needing to construct a reference database. The program DL-TODA presented here aims to classify metagenomic reads using a deep learning model trained on over 3000 bacterial species. A convolutional neural network architecture originally designed for computer vision was applied for the modeling of species-specific features. Using synthetic testing data simulated with 2454 genomes from 639 species, DL-TODA was shown to classify nearly 75% of the reads with high confidence. The classification accuracy of DL-TODA was over 0.98 at taxonomic ranks above the genus level, making it comparable with Kraken2 and Centrifuge, two state-of-the-art taxonomic classification tools. DL-TODA also achieved an accuracy of 0.97 at the species level, which is higher than 0.93 by Kraken2 and 0.85 by Centrifuge on the same test set. Application of DL-TODA to the human oral and cropland soil metagenomes further demonstrated its use in analyzing microbiomes from diverse environments. Compared to Centrifuge and Kraken2, DL-TODA predicted distinct relative abundance rankings and is less biased toward a single taxon.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Mitigating Algorithmic Bias in Cancer Site Classification Models

Purpose Integrating artificial intelligence in cancer diagnostics has improved tumor classification beyond rule-based systems. Despite these advancements, these models may still encode demographic biases. We conducted a large-scale, applied bias-probing study of a deep learning–based cancer site classifier to quantify race information encoded in document embeddings. We then evaluated how performance changes when race-correlated embedding dimensions are removed in a post-training sensitivity analysis. Methods The cancer site classifier was trained using 3.5 million electronic cancer pathology reports from six of the National Cancer Institute's SEER registries. We trained a hierarchical self-attention network to generate 400-dimensional document embeddings. These embeddings were used to train two downstream, gradient-boosted decision tree classifiers: one to classify the cancer sites and another to predict racial categories. We identified overlapping features by intersecting the top 50 feature-importance rankings from the site and race models and computed their cumulative feature importance in each model. As a post hoc sensitivity analysis, we progressively pruned these overlapping dimensions, retrained the site model, and compared overall macro-F1 and accuracy, race-stratified macro-F1, and group fairness metrics on the basis of demographic parity and equalized odds before and after pruning. Results The analysis revealed minimal feature overlap between the cancer site and race prediction models, and the cumulative importance scores indicated a negligible influence of racial information on clinical predictions. Post-training pruning of overlapping features did not compromise the models' diagnostic accuracy, with a 0.07% loss in accuracy. Conclusion Our findings demonstrate that HiSAN-generated embeddings from SEER data can be used effectively in cancer site classification without significant demographic bias influencing the outcomes. Post-training pruning therefore functions as a practical audit and sensitivity check.

Shivanna, Abhishek [ORNL] (ORCID:0009000665228593)↗
DOE OSTI2024

Modern deep neural networks for Direct Normal Irradiance forecasting: A classification approach

The escalating energy demand and the adverse environmental impacts of fossil-fuel use necessitate a shift towards cleaner and renewable alternatives. Concentrated Solar Power (CSP) technology emerges as a promising solution, offering a carbon-free alternative for power generation. The efficiency and profitability of CSP depend on the Direct Normal Irradiance (DNI) component of solar radiation; hence, accurate DNI forecasting can help optimize CSP plants’ operations and performance. The unpredictable nature of weather phenomena, particularly cloud cover, introduces uncertainty into DNI projections. Existing DNI forecasting models use meteorological factors, which are both challenging to estimate numerically over short prediction windows and expensive to model through data at a sufficiently high spatial and temporal resolution. This research addresses the challenge by presenting a novel approach that formulates DNI prediction as a multi-class classification problem, departing from conventional regression-based methods. The primary objective of this classification framework is to identify optimal periods aligning with specific operational thresholds for CSP plants, contributing to enhanced dispatch optimization strategies. We model the DNI classification problem using four advanced deep neural networks – rectified linear unit (ReLU) networks, 1D residual networks (ResNets), bidirectional long short-term memory (BiLSTM) networks, and transformers – achieving accuracies up to 93.5% without requiring meteorological parameters.

14 SOLAR ENERGY↗
DOE OSTI2025

Deciphering the small-angle scattering of polydisperse hard spheres using deep learning

We introduce a deep learning approach for analyzing the scattering function of the polydisperse hard sphere system. We use a variational autoencoder-based neural network to learn the bidirectional mapping between the scattering function and the system parameters, including the volume fraction and polydispersity. Such that the trained model serves both as a generator that produces a scattering function from the system parameters and an inferrer that extracts system parameters from the scattering function. We first generate a scattering dataset by carrying out molecular dynamics simulations of the polydisperse hard spheres modeled by the truncated-shifted Lennard-Jones model, then analyze the scattering function dataset using singular value decomposition to confirm the feasibility of dimensional compression. Then, we split the dataset into training and testing sets and train our neural network on the training set only. Our generator model produces a scattering function with significantly higher accuracy compared to the traditional Percus–Yevick approximation and β correction, and the inferrer model can extract the volume fraction and polydispersity with much higher accuracy than traditional model functions.

Ding, Lijie [ORNL] (ORCID:0000000227454606)↗
DOE OSTI2024

Discovering Elusive Dynamics Across Frontiers (Final Technical Report)

Profound puzzles, such as the nature of dark matter, the origins of the electroweak scale, the mechanism behind the small neutrino mass, and the strong CP problem, suggested new physics beyond the Standard Model and drove the particle physics program in search of the associated new particles. Despite extensive searches, conventional realizations of new physics have not yet provided conclusive evidence. This raised the serious possibility that new dynamics might be more elusive, perhaps due to a richer gauge and matter structure than previously considered. Notably, the existence of dark matter and advancements in understanding the naturalness problem urged exploration into sectors with complex gauge and matter structures. Through experiments like the LHC, DUNE, and small-scale experiments, the robust US HEP program played a critical role in pursuing these well-motivated but under-explored scenarios for new physics. I explored these physics opportunities in depth, focusing on novel searches and significant improvements in parameter space coverage. The elusive dynamics revealed rich information about the underlying theory and were essential in identifying observable opportunities. Understanding the observable consequences required a deep comprehension of the theory, which I also developed. The proposal included essential components aimed at coherently increasing our knowledge in well-motivated elusive dynamics models. Through research on high-quality axions, composite neutrinos, the Higgs boson as a portal to hidden strong dynamics, and new scalar potentials to generate alternative electroweak phase transitions, I focused on identifying new signatures and parameter regions in plausible elusive dynamics models. The exploration emphasized generic possibilities motivated by broad classes of elusive dynamics models. These signatures were not effectively probed previously due to various challenges such as triggering, background suppression, or experimental design. My research involved close interaction with experimental colleagues to overcome these difficulties, leveraging new theoretical and experimental possibilities. These efforts included identifying new observables such as timing information and substructure in calorimetries, new multiple-hit techniques, new scattering events, and new resonance searches in liquid argon detectors. This work created a positive feedback loop: theory and experimental work inspired each other, revealing new exciting opportunities that supported both programs.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2020

Multiple-objective Reinforcement Learning for Inverse Design and Identification

The aim of inverse chemical design is to develop new molecules with given optimized molecular properties or objectives. Recently, generative deep learning (DL) networks are considered as the state-of-the-art in inverse chemical design and have achieved early success in generating molecular structures with desired properties in the pharmaceutical and material chemistry fields. However, satisfying a large number (> 10 objectives) of molecular objectives is a limitation of current generative models. To improve the model’s ability to handle a large number of molecule design objectives, we developed a Reinforcement Learning (RL) based generative framework to optimize chemical molecule generation. Our use of Curriculum Learning (CL) to fine-tune the pre-trained generative network allowed the model to satisfy up to 21 objectives and increase the generative network’s robustness. The experiments show that the proposed multiple-objective RL-based generative model can correctly identify unknown molecules with an 83% to 100% success rate, compared to the baseline approach of 0%. Additionally, this proposed generative model is not limited to just chemistry research challenges; we anticipate that problems that utilize RL with multiple objectives will benefit from this framework.

reinforcement learning, generative model, inverse ↗
DOE OSTI2025

Expanding the physics reach of DUNE in the near and far detectors

The Deep Underground Neutrino Experiment (DUNE) is a next-generation long-baseline neutrino oscillation experiment. Its primary goal is the determination of the neutrino mass hierarchy and the CP-violating phase. The DUNE physics programme also includes the detection of astrophysical neutrinos and the search for beyond the Standard Model (BSM) phenomena. DUNE will consist of a near detector (ND) complex placed at Fermilab, and a modular Liquid Argon Time Projection Chamber (LArTPC) far detector (FD) to be built in the Sanford Underground Research Facility (SURF), approximately 1300 km away from the neutrino production point. This thesis describes three different projects within DUNE. First, a novel strategy to improve the triggering capabilities of the DUNE FD is proposed. It uses matched filters to enhance the production of online hits across all charge collection planes. Next, the possibility of detecting neutrinos coming from dark matter (DM) annihilations in the Sun with the FD is explored. The complementarity of DUNE to this kind of DM searches is shown. Finally, the simulation and reconstruction framework of ND-GAr, the gas argon ND proposed for Phase II of DUNE, is presented. A number of additions to this are described, particularly focused on the development of the particle identification (PID) capabilities of the detector. These are then used to perform the first event selection studies with an end-to-end simulation in ND-GAr, in particular the selection of pion exclusive samples in $\nu_{\mu}$ CC interactions. All three of these projects share the common goal of enhancing the physics programme of DUNE.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2022

Occlusion-Perturbed Deep Learning for Probabilistic Solar Forecasting via Sky Images

Solar forecasting is shifting to the probabilistic paradigm due to the inherent uncertainty within the solar resource. Input uncertainty quantification is one of the widely used and best-performing ways to model solar uncertainty. However, compared to other sources of inputs, such as numerical weather prediction models, pure sky image-based probabilistic solar forecasting lags behind. In this research, an occlusion-perturbed convolutional neural network, named the PSolarNet, is developed. The PSolarNet provides very short-term deterministic forecasts, forecast scenarios, and probabilistic forecasts of the global horizontal irradiance from sky image sequences. Case studies based on 6 years of open-source data show that the developed PSolarNet is able to generate accurate 10-minute ahead deterministic forecasts with a 5.62% normalized root mean square error, realistic and diverse forecast scenarios with a 0.966 average correlation with the actual time series, and reliable and sharp probabilistic forecasts with a 2.77% normalized continuous ranked probability score.

Bayesian model averaging↗
DOE OSTI2022

Occlusion-Perturbed Deep Learning for Probabilistic Solar Forecasting via Sky Images: Preprint

Solar forecasting is shifting to the probabilistic paradigm due to the inherent uncertainty within the solar resource. Input uncertainty quantification is one of the widely used and best-performing ways to model solar uncertainty. However, compared to other sources of inputs, such as numerical weather prediction models, pure sky image-based probabilistic solar forecasting lags behind. In this research, an occlusion-perturbed convolutional neural network, named the PSolarNet, is developed. The PSolarNet provides very short-term deterministic forecasts, forecast scenarios, and probabilistic forecasts of the global horizontal irradiance from sky image sequences. Case studies based on 6 years of open-source data show that the developed PSolarNet is able to generate accurate 10-minute ahead deterministic forecasts with a 5.62% normalized root mean square error, realistic and diverse forecast scenarios with a 0.966 average correlation with the actual time series, and reliable and sharp probabilistic forecasts with a 2.77% normalized continuous ranked probability score.

Bayesian model averaging↗
DOE OSTI2020

Physical and hydrodynamic properties of deep sea mining-generated, abyssal sediment plumes in the Clarion Clipperton Fracture Zone (eastern-central Pacific)

The anthropogenic impact of polymetallic nodule harvesting in the Clarion-Clipperton Fracture Zone is expected to strongly affect the benthic ecosystem. To predict the long-term, industrial-scale impact of nodule mining on the deep-sea environment and to improve the reliability of the sediment plume model, information about the specific characteristics of deep-sea particles is needed. Discharge simulations of mining-related fine-grained (median diameter ≈ 20 µm) sediment plumes at concentrations of 35–500 mg L –1 (dry weight) showed a propensity for rapid flocculation within 10 to 135 min, resulting in the formation of large aggregates up to 1100 µm in diameter. The results indicated that the discharge of elevated plume concentrations (500 mg L –1 ) under an increased shear rate (G ≥ 2.4 s –1 ) would result in improved efficiency of sediment flocculation. Furthermore, particle transport model results suggested that even under typical deep-sea flow conditions (G ≈ 0.1 s –1 ), rapid deposition of particles could be expected, which would restrict heavy sediment blanketing (several centimeters) to a smaller fall-out area near the source, unless subsequent flow events resuspended the sediments. Planning for in situ tests of these model projections is underway.

42 ENGINEERING↗
DOE OSTI2021

Transport and chemistry of isoprene and its oxidation products in deep convective clouds

Deep convective clouds can transport trace gases from the planetary boundary layer into the upper troposphere where subsequent chemistry may impact aerosol particle formation and growth. In this modelling study, we investigate processes that affect isoprene and its oxidation products injected into the upper troposphere by an isolated deep convective cloud in the Amazon. We run a photochemical box model with coupled cloud microphysics along hundreds of individual air parcel trajectories sampled from a cloud-resolving model simulation of a convective event. The box model simulates gas-phase chemical reactions, gas scavenging by liquid and ice hydrometeors, and turbulent dilution inside a deep convective cloud. The results illustrate the potential importance of gas uptake to anvil ice in regulating the intensity of the isoprene oxidation and associated low volatility organic vapour concentrations in the outflow. Isoprene transport and fate also depends on the abundance of lightning-generated nitrogen oxide radicals (NO x = NO + NO 2 ). If gas uptake on ice is efficient and lightning activity is low, around 30% of the boundary layer isoprene will survive to the cloud outflow after approximately one hour of transport, while all the low volatile oxidation products will be scavenged by the cloud hydrometeors. If lightning NO x is abundant and gas uptake by ice is inefficient, then all isoprene will be oxidised during transport or in the immediate outflow region, while several low volatility isoprene oxidation products will have elevated concentrations in the cloud outflow. Reducing uncertainties associated with the uptake of vapours on ice hydrometeors, especially HO 2 and oxygenated organics, is essential to improve predictions of isoprene and its oxidation products in deep convective outflows and their potential contribution to new particle formation and growth.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Characterizing Baselines and Change in Gas Hydrate Systems using EM Methods

The objective of this project was to advance our understanding of gas hydrate systems in nature by characterizing their electrical properties in the field and in the laboratory. In the laboratory measurements, methane hydrate was synthesized from pure water ice and flash frozen seawater, with varying amounts of sand or silt added. Electrical conductivity was determined by impedance spectroscopy, using equivalent circuit modeling to separate the effects of electrodes and to gain insight into conduction mechanisms. Silt and sand increase the conductivity of pure hydrate, inferred to be contaminant NaCl contributing to conduction in hydrate, to a peak conductivity in agreement with peak resistivities observed in well logs through massive hydrate (3,000--10,000 Ωm). The addition of silt and sand lowers the conductivity of hydrate synthesized from seawater, by an amount consistent with Archie's Law. All samples were characterized using cryogenic scanning electron microscopy and energy dispersive spectroscopy, which shows good connectivity of salt and brine phases. Electrical conductivity measurements of pure hydrate and hydrate mixed with silt during pressure-induced dissociation supports previous conclusions that sediment increases dissociation rate. In order to characterize gas hydrate systems in the field, we collected 360 line kilometers of controlled-source electromagnetic data on Walker Ridge 313, Orca Basin (WR100), Mad Dog (GC781), and Green Canyon 955 in the Gulf of Mexico, all areas with known or seismically inferred gas hydrate deposits and which have be drilled or targeted for future drilling. We deep-towed an EM transmitter that generates an alternating electric field which propagates through the seafloor geology. Data were recorded on 6 receivers towed behind the transmitter at distances between 550 and 1550 m. In the presence of conductive geology, the electric fields will be attenuated, and conversely, in resistive geology the fields will be preserved. Our data were inverted using a 2D inversion method that first optimizes the model-data misfit, then finds the smoothest model fitting the data. This ensures that resistivity structures present in the final model are likely necessary. At each of the proposed drilling sites we found increased resistivity, interpreted as increased hydrate concentrations. However, not only were the primary drilling sites not always more resistive than the alternate sites, at WR313 the strongest resistors were not at the locations targeted for drilling.

03 NATURAL GAS↗