Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “deep well”

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 811 records · Page 45
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Accelerated CO2 Storage Optimization Using Multi-Resolution Fourier Neural Operator at the Illinois Basin Decatur Project (IBDP)

This paper presents a deep learning-based approach for optimizing CO2 injection in carbon capture and storage (CCS) operations. We developed a multi-resolution machine learning model to significantly reduce data generation costs. Utilizing this proxy model, we implemented a multi-objective genetic algorithm to optimize well control during the CO2 injection process. The proposed approach was applied to the Illinois Basin Decatur Project (IBDP), successfully optimizing the CO2 injection schedule based on three key objectives: maximizing the amount of CO2 stored, maximizing sweep efficiency, and minimizing pressure increase. The use of the proxy model accelerated the optimization workflow by two orders of magnitude, while the cost of data generation for the proxy model was reduced by 90% by utilizing a coarse-scale model.

accelerated CO2 storage optimization↗
DOE OSTI2023

Predicting Failure Using Deep Learning SAND Report

Accurate prediction of ductile failure is critical to Sandia’s NW mission, but the models are computationally heavy. The costs of including high-fidelity physics and mechanics that are germane to the failure mechanisms are often too burdensome for analysts either because of the person-hours it requires to input them or because of the additional computational time, or both. In an effort to deliver analysts a tool for representing these phenomena with minimal impact to their existing workflow, our project sought to develop modern data-driven methods that would add microstructural information to business-as-usual calculations and expedite failure predictions. The goal is a tool that receives as input a structural model with stress and strain fields, as well as a machine-learned model, and output predictions of structural response in time, including failure. As such, our project spent substantial time performing high-fidelity, three-dimensional experiments to elucidate materials mechanisms of void nucleation and evolution. We developed crystal-plasticity finite-element models from the experimental observations to enrich the findings with fields not readily measured. We developed engineering length-scale simulations of replicated test specimens to understand how the engineering fields evolve in the presence of fine-scale defects. Finally, we developed deep learning convolutional neural networks, and graph-based neural networks to encode the findings of the experiments and simulations and make forward predictions in time for structural performance. This project demonstrated the power of data-driven methods for model development, which have the potential to vastly increase both the accuracy and speed of failure predictions. These benefits and the methods necessary to develop them are highlighted in this report. However, many challenges remain to implementing these in real applications, and these are discussed along with potential methods for overcoming them.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Deep Learning for Multigroup Cross-Section Representation in Two-Step Core Calculations

Here we investigate using deep learning, a type of machine-learning algorithm employing multiple layers of artificial neurons, for the mathematical representation of multigroup cross sections for use in the Griffin reactor multiphysics code for two-step deterministic neutronics calculations. A three-dimensional fuel element typical of a high-temperature gas reactor as well as a two-dimensional sodium-cooled fast reactor lattice are modeled using the Serpent Monte Carlo code, and multigroup macroscopic cross sections are generated for various state parameters to produce a training data set and a separate validation data set. A fully connected, feedforward neural network is trained using the open-source PyTorch machine-learning framework, and its accuracy is compared against the standard piecewise linear interpolation model. Additionally, we provide in this work a generic technique for propagating the cross-section model errors up to the k eff using sensitivity coefficients with the first-order uncertainty propagation rule. Quantifying the eigenvalue error due to the cross-section regression errors is especially practical for appropriately selecting the mathematical representation of the cross sections. We demonstrate that the artificial neural network model produces lower errors and therefore enables better accuracy relative to the piecewise linear model when the cross sections exhibit nonlinear dependencies; especially when a coarse grid is employed, where the errors can be halved by the artificial neural network. However, for linearly dependent multigroup cross sections as found for the sodium-cooled fast reactor case, a simpler linear regression outperforms deeper networks.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2026

Covalent Adaptable Networks: Reprocessable Cross-Linked Polymers

Thermoset polymers have desirable properties, such as excellent thermal and mechanical stability, but their covalent cross-links typically prevent repair or recycling. By enabling and controlling dynamic exchange reactions within polymer networks, their covalent bonds rearrange and allow the polymer to be reshaped. These viscoelastic polymer networks, now known as covalent adaptable networks (CANs), are an important frontier for improving plastic circularity, as well as for designing valuable stimuli-responsive materials. This Review describes the history of CANs, dating back to the early days of polymer science, and the evolution of their classification and nomenclature. A comprehensive survey of dynamic reactions and linkage chemistries is provided, as well as methods to characterize and reprocess CANs. Beyond straightforward reprocessing, many advanced applications of CANs and their composites are now emerging. Lastly, we provide perspective on how the development of new chemistries, strategies to control stimuli-responsive bond exchange and mechanical properties, and a deep understanding of exchange reactions will advance this field toward scalable, sustainable, and high-value materials.

Sun, Molly [Northwestern University, Evanston, IL ↗
DOE OSTI2024

Unraveling the impact of initial choices and in-loop interventions on learning dynamics in autonomous scanning probe microscopy

The current focus in Autonomous Experimentation (AE) is on developing robust workflows to conduct the AE effectively. This entails the need for well-defined approaches to guide the AE process, including strategies for hyperparameter tuning and high-level human interventions within the workflow loop. This paper presents a comprehensive analysis of the influence of initial experimental conditions and in-loop interventions on the learning dynamics of Deep Kernel Learning (DKL) within the realm of AE in scanning probe microscopy. We explore the concept of the “seed effect,” where the initial experiment setup has a substantial impact on the subsequent learning trajectory. Additionally, we introduce an approach of the seed point interventions in AE allowing the operator to influence the exploration process. Using a dataset from Piezoresponse Force Microscopy on PbTiO 3 thin films, we illustrate the impact of the “seed effect” and in-loop seed interventions on the effectiveness of DKL in predicting material properties. The study highlights the importance of initial choices and adaptive interventions in optimizing learning rates and enhancing the efficiency of automated material characterization. This work offers valuable insights into designing more robust and effective AE workflows in microscopy with potential applications across various characterization techniques.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Metagenome-assembled genomes from Wind River Basin floodplain sediments Riverton, Wyoming site (August 2015)

Microorganisms play a key role in cycling nutrients and contaminants in the terrestrial environment depending on their genetic potential. Here we present metagenome-assembled genomes (MAGs) for the bacterial and archaeal community in floodplain sediment samples taken August 29, 2015 at a location (KB1) close to DOE Legacy Management well 855 at the Riverton, Wyoming floodplain site in the Wind River Basin (WRB). The groundwater at this site exhibits persistent U, Mo, and sulfate plumes and is one of the field sites in focus for the SLAC Groundwater Quality SFA program. Sediment samples from a deep soil pit were collected from 0 to 234 cm depth below surface at discrete depths every ~10-20 cm for microbial analyses. 13 metagenomes were sequenced through JGI and can be found under Gold sequencing project: Gs0131241. Metagenomes were assembled, binned, and refined using metawrap to generate MAGs (>50% complete and < 10% contamination based on checkM scores).This dataset includes a zip file of 2216 MAG fasta files and a csv file with quality, taxonomic classification (GTDB RS220), and metagenome accessions for MAGs. This dataset also includes a file-level metadata (flmd.csv) file that lists each file contained in the dataset with associated metadata and a data dictionary (dd.csv) file that contains column/row headers used throughout the files along with a definition, units, and data type

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

JLab E12-14-012 (e,e'p) cross section measurements for Ar and Ti

One of the limitations of accelerator based neutrino oscillation experiments including the Deep Underground Neutrino Experiment (DUNE), is the lack of understanding on how neutrino interact in matter. As an indispensable part of the neutrino physics future, DUNE is an accelerator based experiment that will use Argon as the neutrino target. Argon is a complicated target and a well defined theoretical model for neutrino interaction on Argon is needed. Thus, the JLab E12-14-012 experiment was performed in Hall A at Jefferson Lab in Newport, News, VA to help develop a neutrino-nucleus model for Argon. Data was collected on five targets (Argon, titanium, carbon, dummy and optical) for five different kinematic set-ups. The primary goal of this experiment is to measure the spectral functions of Argon and Titanium through (e,e?p) reactions. This thesis will present an overview of the experimental setup and results from the data analysis

Gu, Linjie↗
DOE OSTI2022

Autoencoder Neural Network for chemically reacting systems

Incorporating detailed chemical kinetic models is critical for accurate simulations of reacting flows. However, detailed models involve a large number of thermochemical (TC) state variables. Solving the governing equations to evolve these TC variables becomes impractical for real-world applications. In this work, we propose an autoencoder (AE) neural network (NN)-based reduced model to accelerate such simulations. The AE NN is first trained to find a low-dimensional latent representation of the TC states. Then, the evolving state of a chemical system can be tracked by solving the equations of the latent variables instead of the original TC equations. We demonstrate the reduced model in a syngas CO/H 2 combustion system, using training data collected from canonical perfectly stirred reactors (PSRs). It is found that the AE model can reduce the dimension of the combustion system from 12 to 2 while maintaining low reconstruction error and excellent elemental mass conservation for the test dataset. In the a posteriori test, the combustion states obtained from solving the two latent equations are compared to those from solving the 12 equations of the full model. The AE reduced method is found to be able to capture the diverse combustion states on the top two branches of the S-curve well including the extinction turning point, but with higher prediction errors for states near the ignition turning point.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Autonomous nondestructive evaluation of resistance spot welded joints

The application of non-destructive evaluation approaches has attracted strong interests in modern automotive industries. Here, we present an autonomous deep-computing framework to analyze raw videos from infrared systems and to predict weld nugget shape and size with unprecedented accuracy and speed. In a comprehensive training and testing experiment with 90 videos (seven sets of welding material stack-ups), a new method was developed to assemble sufficient datasets for neural network training. Our framework successfully predicts all the nugget shapes with F1 scores that range from 0.84 to 0.92. The total training time on Nvidia DGX station takes less than 10 min for each set of welding material stack-up. The real inference time of an individual dataset (with 30 video frames) takes about 0.005 s. The procedure and methods developed in the study can be applied to other image-based weld property prediction, as well as other manufacturing processes. Furthermore, our well-trained neural networks take limited memory resources (2.3 MB) and are suitable for embedded microprocessors for in-situ welding quality control as edge computing within an intelligent welding framework.

42 ENGINEERING↗
DOE OSTI2025

Friction Self-piercing Riveting Process for 3T Configuration of AA7075-AA7075-AA6022

Friction self-piercing riveting (F-SPR), as a solid-state joining process that generates frictional heat during the process, was utilized to join a 3 T configuration of low-ductility AA7075-AA7075-AA6022 sheets. A multi-step F-SPR process was employed to precisely control the mechanical interlocking and solid-state joining, ensuring a strong joint formation. With control of the plunge depth when the transition from rivet penetration to rivet flaring, two joining conditions were developed to assess mechanical joint integrity: one bearing deep penetration with a smaller interlock, and the other yielding a shallow penetration with a larger interlock. Both strategies met the criteria for a sound rivet joint. Customized lap shear tensile configurations were designed to assess the joint strength at interface between first and second layers, as well as between second and third layers. Both conditions demonstrated great joint strength between the first and second layers, while deeper penetration approach showed stronger solid-state joint, leading to improved joint strength between the second and third layers.

Wang, Tianzhao [ORNL] (ORCID:0000000315337602)↗
DOE OSTI2021

Modulating solvation interactions of deep eutectic solvents formed by ammonium salts and carboxylic acids through varying the molar ratio of hydrogen bond donor and acceptor

Deep eutectic solvents (DESs) have gained increasing popularity in separation science due to the fact that their physico-chemical properties can be easily fine-tuned by varying the type or ratio of hydrogen bond acceptor (HBA) and hydrogen bond donor (HBD). While it is well-known that the molar ratio of HBA/HBD affects the melting point of a eutectic mixture, much less is understood regarding its effect on the magnitude of individual solvation interactions. This is largely due to the fact that established solvatochromic dye methods lack sensitivity when the HBA/HBD ratio is varied slightly in a eutectic mixture. Herein, this study is the first to measure the variation of DES solvation interactions with small changes in the molar ratio of HBA/HBD using inverse gas chromatography (IGC). Solute-solvent interactions of three different DES systems comprised of ammonium salts and organic acids were examined. The probe molecules were studied for 18 eutectic mixtures of varied HBA and HBD composition. We report DES hydrogen bond basicity, hydrogen bond acidity, and dispersive-type interactions exhibited the greatest change when the molar ratio of HBA/HBD was varied in the eutectic mixture. Results from this study demonstrate that the HBA/HBD ratio can be used to modulate the solvation characteristics for this class of DESs in separations and that the stoichiometric ratio of the HBA/HBD is important in ensuring their reproducible preparation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Analysis of neptunium oxides produced through modified direct denitration

Production of neptunium-237 ( 237 Np) target materials for plutonium-238 ( 238 Pu) radioisotope thermoelectric generators (RTGs) for deep space exploration requires advanced chemistry and engineering development. Currently, the domestic Pu-238 Supply Program at Oak Ridge National Laboratory produces neptunium dioxide (NpO 2 ) for target material using a modified direct denitration (MDD) flowsheet. Although the chemistry, reaction mechanisms, and product characteristics of MDD are well understood for uranium, corresponding studies of the neptunium system are still needed to continue optimization of target material properties, production equipment design, and production flowsheets. Here, the objective of this work is to characterize crystalline phases, morphology, surface texture, and particle size of NpO 2 produced via MDD reactions. Solid-phase characterization techniques, including powder X-ray diffraction (pXRD) and scanning electron microscopy with energy-dispersive spectroscopy (SEM-EDS), were employed to achieve this objective. Subsequent data processing using the Morphological Analysis for Material Attribution (MAMA) software was performed to analyze particle morphology and size. Broadly, the powders were found to contain a mixture of NpO 2 and Np 2 O 5 after denitration with a variety of morphologies. After high-firing, the product was found to be NpO 2 with a typical polycrystalline oxide morphology and a grain size ranging from 0.72 to 0.94 µm. These analyses provide knowledge on the reaction pathway for a non-traditional NpO 2 synthesis method and offer additional unique insight into production-scale environments for transuranic materials.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2024

Quasi-Classical Trajectory Calculation of Rate Constants Using an Ab Initio Trained Machine Learning Model (aML-MD) with Multifidelity Data

Machine learning (ML) provides a great opportunity for the construction of models with improved accuracy in classical molecular dynamics (MD). However, the accuracy of a ML trained model is limited by the quality and quantity of the training data. Generating large sets of accurate ab initio training data can require significant computational resources. Furthermore, inconsistent or incompatible data with different accuracies obtained using different methods may lead to biased or unreliable ML models that do not accurately represent the underlying physics. Recently, transfer learning showed its potential for avoiding these problems as well as for improving the accuracy, efficiency, and generalization of ML models using multifidelity data. In this work, ab initio trained ML-based MD (aML-MD) models are developed through transfer learning using DFT and multireference data from multiple sources with varying accuracy within the Deep Potential MD framework. Further, the accuracy of the force field is demonstrated by calculating rate constants for the H + HO 2 → H 2 + 3 O 2 reaction using quasi-classical trajectories. We show that the aML-MD model with transfer learning can accurately predict the rate constants while reducing the computational cost by more than five times compared to the use of more expensive quantum chemistry training data sets. Hence, the aML-MD model with transfer learning shows great potential in using multifidelity data to reduce the computational cost involved in generating the training set for these potentials.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

How Well Does the DOE Global Storm Resolving Model Simulate Clouds and Precipitation Over the Amazon?

This study assesses a 40-day 3.25-km global simulation of the Simple Cloud-Resolving E3SM Model (SCREAMv0) using high-resolution ground-based observations from the Atmospheric Radiation Measurement (ARM) Green Ocean Amazon (GoAmazon) field campaign. SCREAMv0 reasonably captures the diurnal timing of boundary layer clouds yet underestimates the boundary layer cloud fraction and mid-level congestus. SCREAMv0 well replicates the precipitation diurnal cycle, however it exhibits biases in the precipitation cluster size distribution compared to scanning radar observations. Specifically, SCREAMv0 overproduces clusters smaller than 128 km, and does not form enough large clusters. Such biases suggest an inhibition of convective upscale growth, preventing isolated deep convective clusters from evolving into larger mesoscale systems. This model bias is partially attributed to the misrepresentation of land-atmosphere coupling. This study highlights the potential use of high-resolution ground-based observations to diagnose convective processes in global storm resolving model simulations, identify key model deficiencies, and guide future process-oriented model sensitivity tests and detailed analyses.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators

It is widely known that neural networks (NNs) are universal approximators of continuous functions. However, a less known but powerful result is that a NN with a single hidden layer can accurately approximate any nonlinear continuous operator. This universal approximation theorem of operators is suggestive of the structure and potential of deep neural networks (DNNs) in learning continuous operators or complex systems from streams of scattered data. Here, in this work, we thus extend this theorem to DNNs. We design a new network with small generalization error, the deep operator network (DeepONet), which consists of a DNN for encoding the discrete input function space (branch net) and another DNN for encoding the domain of the output functions (trunk net). We demonstrate that DeepONet can learn various explicit operators, such as integrals and fractional Laplacians, as well as implicit operators that represent deterministic and stochastic differential equations. We study different formulations of the input function space and its effect on the generalization error for 16 different diverse applications.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

DCTRGAN: improving the precision of generative models with reweighting

Significant advances in deep learning have led to more widely used and precise neural network-based generative models such as Generative Adversarial Networks (Gans). We introduce a post-hoc correction to deep generative models to further improve their fidelity, based on the Deep neural networks using the Classification for Tuning and Reweighting (Dctr) protocol. The correction takes the form of a reweighting function that can be applied to generated examples when making predictions from the simulation. We illustrate this approach using Gans trained on standard multimodal probability densities as well as calorimeter simulations from high energy physics. We show that the weighted Gan examples significantly improve the accuracy of the generated samples without a large loss in statistical power. This approach could be applied to any generative model and is a promising refinement method for high energy physics applications and beyond.

47 OTHER INSTRUMENTATION↗
DOE OSTI2020

The Impact of Parameterized Lateral Mixing on the Antarctic Circumpolar Current in a Coupled Climate Model

This study examines the impact of changing the lateral diffusion coefficient ARedi on the transport of the Antarctic Circumpolar Current (ACC). The lateral diffusion coefficient ARedi is poorly constrained, with values ranging across an order of magnitude in climate models. The ACC is difficult to accurately simulate, and there is a large spread in eastward transport in the Southern Ocean (SO) in these models. This paper examines how much of that spread can be attributed to different eddy parameterization coefficients. A coarse-resolution, fully coupled model suite was run with A Redi = 400, 800, 1200, and 2400 m 2 s -1 . Additionally, two simulations were run with two-dimensional representations of the mixing coefficient based on satellite altimetry. Relative to the 400 m 2 s -1 case, the 2400 m 2 s -1 case exhibits 1) an 11% decrease in average wind stress from 50° to 65°S, 2) a 20% decrease in zonally averaged eastward transport in the SO, and 3) a 14% weaker transport through the Drake Passage. The decrease in transport is well explained by changes in the thermal current shear, largely due to increases in ocean density occurring on the northern side of the ACC. In intermediate waters these increases are associated with changes in the formation of intermediate waters in the North Pacific. We hypothesize that the deep increases are associated with changes in the wind stress curl allowing Antarctic Bottom Water to escape and flow northward.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Data and Code for Understanding Generative AI Content with Embedding Models

This repository contains code for the experiments in the paper "Understanding Generative AI Content with Embedding Models". Constructing high-quality features is critical to any quantitative data analysis. While feature engineering was historically addressed by carefully hand-crafting data representations based on domain expertise, deep neural networks (DNNs) now offer a radically different approach. DNNs implicitly engineer features by transforming their input data into hidden feature vectors called embeddings. For embedding vectors produced by foundation models -- which are trained to be useful across many contexts -- we demonstrate that simple and well-studied dimensionality-reduction techniques such as Principal Component Analysis uncover inherent heterogeneity in input data concordant with human-understandable explanations. Of the many applications for this framework, we find empirical evidence that there is intrinsic separability between real samples and those generated by artificial intelligence (AI).

Vargas, Max [Pacific Northwest National Laboratory↗