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 217 records · Page 12
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Data-Driven Approach to Transactive Energy Systems with Commercial Buildings

A microgrid with solar, storage, and responsive load resources has been implemented and tested on an urban academic campus. Through modeling and simulation, a consensus transactive energy mechanism has been implemented, with each resource participating as a virtual battery. Most owners of large buildings don't have the information and expertise to develop and validate suitable models of their buildings using available tools. To mitigate this adoption barrier, a data-driven building model has been implemented and validated. It uses 5-minute weather data, 3-second revenue meter data, energy audit information, and a load reduction test conducted by the building owner.

Buildings, data-driven modeling, deep learning, en↗
DOE OSTI2023

Aerobic respiration controls on shale weathering, Geochimica et Cosmochimica Acta, 2023: Dataset

This data package was generated in order to support the development of a deep-time weathering model and to assess the coupling between shale weathering and aerobic respiration in the paper “Aerobic respiration controls on shale weathering” by Stolze et al., Geochimica et Cosmochimica Acta (2023). The package contains two csv files providing the average CO2(g) concentration profiles [ppm] and mineral concentration profiles [wt%], respectively. The CO2(g) concentration profiles were measured in the vicinity of the monitoring well PLM2 between January 2018 and April 2019. The gas samples were collected in the unsaturated zone to a depth of 1.52 m. The mineral concentration profiles were determined by X-Ray diffraction (XRD). The XRD measurements were performed on sub-core samples collected in the monitoring well PLM3 down to a depth of 7.01 m. The dataset additionally 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.Update on 2024-05-28: Revised versions of the CSV data files (CO2_data_GCA_Stolze_et_al_2023.csv and XRD_data_GCA_Stolze_et_al_2023.csv) were made to apply ESS-DIVE's CSV reporting format guidelines. Updated versions of the File Level Metadata (v2_20240528_flmd.csv) and Data Dictionary (v2_20240528_dd.csv) files were updated to reflect the changes made to the CSV files.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1988

Ocean alkalinity and the Cretaceous/Tertiary boundary

A biogeochemical cycle model resolving ocean carbon and alkalinity content is applied to the Maestrichtian and Danian. The model computes oceanic concentrations and distributions of Ca(2+), Mg(2+), and Sigma-CO2. From these values an atmospheric pCO2 value is calculated, which is used to estimate rates of terrestrial weathering of calcite, dolomite, and calcium and magnesium silicates. Metamorphism of carbonate rocks and the subsequent outgassing of CO2 to the atmosphere are parameterized in terms of carbonate rock reservoir sizes, total land area, and a measure of overall tectonic activity, the sea-floor generation rate. The ocean carbon reservoir computed by the model is used with Deep Sea Drilling Project (DSDP) C-13 data to estimate organic detrital fluxes under a variety of ocean mixing rate assumptions. Using Redfield ratios, the biogenic detrital flux estimate is used to partition the ocean carbon and alkalinity reservoirs between the mixed layer and deep ocean. The calcite flux estimate and carbonate ion concentrations are used to determine the rate of biologically mediated CaCO3 titration. Oceanic productivity was severely limited for approximately 500 kyr following the K/T boundary resulting in significant increases in total ocean alkalinity. As productivity returned to the ocean, excess carbon and alkalinity was removed from the ocean as CaCO3. Model runs indicate that this resulted in a transient imbalance in the other direction. Ocean chemistry returned to near-equilibrium by about 64 mybp.

Caldeira, K. G.↗
NASA NTRS2019

Pixel-Based Smoke Detection with Neural Network

Exposure to biomass burning has been linked to respiratory and cardiovascular illnesses in humans. Traditional satellite based visual and multispectral smoke detection methods are not scalable to capabilities of new generations of remote sensing platforms. We develop a scalable, deep learning based detection model capable of identifying smoke pixels using GOES-16 shortwave reflectance data and present a operational web-based tool to visualize smoke predictions.

Ramasubramanian, M.↗
DOE OSTI2022

Bat554: Machine Learning for Accelerated Life Prediction and Cell Design

This presentation describes work associated with the early prediction of life and failure modes using machine learning. Key areas which will be highlighted are the generation of synthetic data, the use of deep learning and other machine learning models and the use of machine learning to train life prediction models.

25 ENERGY STORAGE↗
DOE OSTI2024

DRDMannTurb: A Python package for scalable, data-driven synthetic turbulence

Synthetic turbulence models (STMs) are used in wind engineering to generate realistic flow fields and are employed as inputs to industrial wind simulations. Examples include prescribing inlet conditions in large eddy simulations that model loads on wind turbines and tall buildings. We are interested in STMs capable of generating fluctuations based on prescribed second-moment statistics since such models can simulate environmental conditions that closely resemble on-site observations. To this end, the widely used Mann model (see Mann, 1994, 1998) is the inspiration for DRDMannTurb. The Mann model is described by three physical parameters: a magnitude parameter influencing the global variance of the wind field and corresponding to the Kolmogorov constant multiplied by the rate of viscous dissipation of the turbulent kinetic energy to the two-thirds, αϵ 2/3 , a turbulence length scale parameter L, and a nondimensional parameter Γ related to the lifetime of the eddies. A number of studies, as well as international standards (e.g., those by the International Electrotechnical Commission (IEC)), include recommended values for these three parameters with the goal of standardizing wind simulations according to observed energy spectra. Yet, having only three parameters, the Mann model faces limitations in accurately representing the diversity of observable spectra. This Python package enables users to extend the Mann model and more accurately fit field measurements through flexible neural network models of the eddy lifetime function. Following Keith et al. (2021), we refer to this class of models as Deep Rapid Distortion (DRD) models. DRDMannTurb also includes a general module implementing an efficient method for synthetic turbulence generation based on a domain decomposition technique. This technique is also described in Keith et al. (2021).

17 WIND ENERGY↗
NASA NTRS2022

SatNet: A Benchmark for Satellite Scheduling Optimization

Satellites provide essential services such as networking and weather tracking, and the number of near-earth and deep space satellites are expected to grow rapidly in the coming years. Communications with terrestrial ground stations is one of the critical functionalities of any space mission. Satellite scheduling is a problem that has been scientifically investigated since the 1970s. A central aspect of this problem is the need to consider resource contention and satellite visibility constraints as they require line of sight. Due to the combinatorial nature of the problem, prior solutions such as linear programs and evolutionary algorithms require extensive compute capabilities to output a feasible schedule for each scenario. Machine learning based scheduling can provide an alternative solution by training a model with historical data and generating a schedule quickly with model inference. We present SatNet, a benchmark for satellite scheduling optimization based on historical data from the NASA Deep Space Network. We propose formulation of the satellite scheduling problem as a Markov Decision Process and use reinforcement learning (RL) policies to generate schedules. The nature of constraints imposed by SatNet differ from other combinatorial optimization problems such as vehicle routing studied in prior literature. Our initial results indicate that RL is an alternative optimization approach that can generate candidate solutions of comparable quality to existing state-of-the-practice results. However, we also find that RL policies overfit to the training dataset and do not generalize well to new data, thereby necessitating continued research on reusable and generalizable agents.

Wilson, Brian↗
DOE OSTI2025

Large Language Model Integration for Knowledge Retrieval and Interaction for the DUNE Experiment

The Deep Underground Neutrino Experiment (DUNE) is a next-generation neutrino experiment that will generate an unprecedented volume of heterogeneous information-from documentation and technical notes to experimental data and reconstruction pipelines. Efficient knowledge retrieval and contextual understanding are increasingly critical for collaboration-wide productivity and onboarding. In this work, we present DUNE-GPT, a prototype framework that leverages large language models (LLMs) and retrieval-augmented generation (RAG) to enable natural-language querying of DUNE's internal documentation and technical resources. The system provides an intelligent interface for DUNE collaborators to interact with experiment-specific knowledge while maintaining data privacy and infrastructure compliance within Fermilab computing resources.

Rafique, A. [Argonne (main)]↗
DOE OSTI2023

3D-equivariant graph neural networks for protein model quality assessment

Quality assessment (QA) of predicted protein tertiary structure models plays an important role in ranking and using them. With the recent development of deep learning end-to-end protein structure prediction techniques for generating highly confident tertiary structures for most proteins, it is important to explore corresponding QA strategies to evaluate and select the structural models predicted by them since these models have better quality and different properties than the models predicted by traditional tertiary structure prediction methods. We develop EnQA, a novel graph-based 3D-equivariant neural network method that is equivariant to rotation and translation of 3D objects to estimate the accuracy of protein structural models by leveraging the structural features acquired from the state-of-the-art tertiary structure prediction method—AlphaFold2. We train and test the method on both traditional model datasets (e.g. the datasets of the Critical Assessment of Techniques for Protein Structure Prediction) and a new dataset of high-quality structural models predicted only by AlphaFold2 for the proteins whose experimental structures were released recently. Our approach achieves state-of-the-art performance on protein structural models predicted by both traditional protein structure prediction methods and the latest end-to-end deep learning method—AlphaFold2. It performs even better than the model QA scores provided by AlphaFold2 itself. The results illustrate that the 3D-equivariant graph neural network is a promising approach to the evaluation of protein structural models. Integrating AlphaFold2 features with other complementary sequence and structural features is important for improving protein model QA.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Entropy-Infused Deep Learning Loss Function for Capturing Extreme Values in Wind Power Forecasting

Extreme scenarios in wind power generation occur with higher frequency and larger magnitude in the recent years due to the ever-increasing extreme meteorological factors. Accurate forecasting of the occurrence of extreme values in wind power generation is of great concern to ensure reliable power system operation. Recently, deep learning models have surged in popularity for wind power forecasting, with the mean squared error (MSE) loss function being commonly used. However, the MSE loss function, being sensitive to extreme values, disproportionately penalizes larger errors, cannot adequately capture the extreme values present in wind energy data, and novel loss functions have seldom been tailored for wind power forecasting. To this end, in this paper, we introduce a novel loss function specifically crafted to capture extreme values in wind power forecasting. The experimental results with four fundamental deep learning methods on open source wind power dataset validate that the new loss function is efficient and superior in all cases compared to MSE in capturing extreme values while maintaining forecasting performance.

17 WIND ENERGY↗
DOE OSTI2025

WaveDenoiser

We developed a robust deep learning model to effectively reduce background noise in the time domain from seismic waveforms. The deep learning model processes a 57-second three-component seismogram to predict and generate a denoised seismogram. The training was conducted using the benchmark STEAD dataset, which comprises globally distributed earthquake signals recorded at local distances ranging from 0 to 350 kilometers.

Chai, Chengping [Oak Ridge National Laboratory (OR↗
NASA NTRS2024

Development of a Data Fusion Methodology for Lineload Aerodynamic Databases for a Launch Vehicle during Liftoff and Transition

The need for databases for the distributed loading on launch vehicles during the early portion of flight necessitates the use of expensive computational flows in regimes where wake effects dominate. While also being expensive, this is a regime that computational tools tend to historically have problems simulating accurately. To help tackle this problem, a method of data fusion to combine computational results to wind tunnel derived force and moment data is developed. Using this method, significant reduction in computational costs and increases in confidence of the final product is possible and has been used to generate several databases for the Space Launch System (SLS) at NASA. While the full details of database generation are not part of this work, the crucial method at its core is developed here. Two SLS geometries are used throughout the work to demonstrate the techniques. These are two of the larger geometries and represent both planned crewed missions to the Moon as well as potential cargo missions to deep space. The method uses principal component analysis (PCA) to generate a reduced ordered model (ROM) to help fill in the full parameter space. Other similar techniques are explored, but were not found to have a significant result on the predictions of the ROM. Because the full number of components are kept to generate the model, this lack of difference is expected. This method is then extended to ensure that predicted surfaces match trusted force and moment data derived from wind tunnel testing. This extension is done by setting up a constrained optimization problem in order to minimize the deviation from the surface resolved computational data while still integrating to the desired values. When generating the constrained optimization problem, a weighting factor to balance these competing needs is introduced. The work compares previously introduced weighting terms from similar work to the proposed terms and shows that the previously used terms do not have as desirable behavior in this flow regime. This method is then expanded by developing a technique to incorporate uncertainty quantification into the developed data fusion methodology. This expansion takes a two pronged approach. One examines transferring the uncertainties in the force and moment database and characterizes how those adjustments change the predicted lineloads. The second looks at model form error and looks how rebuilding the model using slightly different data changes the predictions. These two terms are then combined in order to create an uncertainty model that takes both effects into account. The limitations of the proposed methods is then discussed as well as possible techniques to address these shortcomings.

Launch Vehicles↗
DOE OSTI2022

Accelerating Hydraulic Fracture Imaging by Deep Transfer Learning

Deep transfer learning has a great success story in computer vision (CV), natural language processing, and many other fields. In this communication, we are going to push forward the deep transfer learning to the hydraulic fracture imaging problem by proposing a two-step approach: 1) train a convolutional neural network (CNN) to reconstruct target geometries by a relatively large amount of approximated field patterns generated from a simplified model and 2) fine-tune the top layers of transferred CNN by a small amount of true field patterns generated through a full model. The advantages include the rapid generation of large amount of data through the simplified model and the high reconstruction accuracy through a careful design of the deep transfer learning. Here, the CNN trained through the deep transfer learning can accurately reconstruct the lateral extent and direction of fractures with unseen conductivity and white Gaussian noise, showing a notable acceleration/accuracy enhancement over the previous CNN trained by mixing a small number of true data into the approximated data as data augmentation.

02 PETROLEUM↗
NASA NTRS2019

Using Deep Learning to Automate Inference of Meteoroid Pre-Entry Properties

Properly assessing the asteroid threat depends on the knowledge of asteroid pre-entry parameters, such as size, velocity, mass, density, and strength. Although a vast number of possible bodies to study exist, such characterization of asteroid populations is currently limited by substantial costs associated with space rendezvous missions and rare meteorite findings. As asteroids fragment, ablate, and decelerate in the atmosphere, they emit light detectable by ground-based and space-borne instruments. Earth’s atmosphere, thus, becomes an accessible laboratory that enables impactor risk assessments by facilitating inference of the pre-entry parameters. These asteroid pre-entry conditions are typically deduced by modeling the entry and breakup physics that best reproduce the observed light or energy deposition curve. However, this process requires extensive manual trial-and-error of uncertain modeling parameters. Automating meteor modeling and inference would improve property distributions used in risk assessments and enable population characterization as more light curves become more readily available through the presence of space assets and ground-based camera networks. We previously developed a genetic algorithm to automate meteor modeling by using the fragment-cloud model (FCM) to search for the values of the FCM input parameters (e.g., diameter) that generate energy deposition profiles that match the observed one. Now, we apply deep learning to infer asteroid diameter, velocity, and density from observed energy deposition curves. We trained and tested our neural network models with synthetic energy deposition curves modeled using the FCM rubble pile implementation. We present an application of a 1D convolutional neural network and compare its performance to other attempted regressors and machine learning techniques, such as a fully connected neural network and Random Forest regression, to demonstrate its capabilities. We validate our model weights and approach using the Chelyabinsk, Tagish Lake, Benešov, Košice, and Lost City meteors.

Tarano, Ana Maria↗
DOE OSTI2022

Deep learning to estimate permeability using geophysical data

Time-lapse electrical resistivity tomography (ERT) is a popular geophysical method to estimate three-dimensional (3D) permeability fields from electrical potential difference measurements. Traditional inversion and data assimilation methods are used to ingest this ERT data into hydrogeophysical models to estimate permeability. Due to ill-posedness and the curse of dimensionality, existing inversion strategies provide poor estimates and low resolution of the 3D permeability field. Recent advances in deep learning provide us with powerful algorithms to overcome this challenge. This paper presents a deep learning (DL) framework to estimate the 3D subsurface permeability from time-lapse ERT data. To test the feasibility of the proposed framework, we train DL-enabled inverse models on simulation data. Each measurement in both synthetic and field data is standardized by removing the mean and scaling the time-series to unit variance. This pre-processing step is necessary to bring simulation data closer to field observations. Subsurface process models based on hydrogeophysics are used to generate this synthetic data. Training performed on limited simulation data resulted in the DL model over-fitting. An advanced data augmentation based on mixup is implemented to generate additional training samples to overcome this issue. This mixup technique creates weakly labeled (low-fidelity) samples from strongly labeled (high-fidelity) data. The weakly labeled training data is then used to develop DL-enabled inverse models and reduce over-fitting. As both time-lapse ERT (1133048 features/realization) and 3D permeability (585453 features/realization) data samples are from a high-dimensional space, principal component analysis (PCA) is employed to reduce dimensionality. Encoded ERT and encoded permeability are generated using the trained PCA estimators. A deep neural network is then trained to map the encoded ERT to encoded permeability. This mixup training and unsupervised learning allowed us to build a fast and reasonably accurate DL-based inverse model under limited simulation data. Results show that proposed weak supervised learning can capture salient spatial features in the 3D permeability field. Quantitatively, the average mean squared error (in terms of the natural log) on the strongly labeled training, validation, and test datasets is less than 0.5. The R 2 -score (global metric) is greater than 0.75, and the percent error in each cell (local metric) is less than 10%. Finally, an added benefit in terms of computational cost is that the proposed DL-based inverse model is at least O(10 4 ) times faster than running a forward model once it is trained. Data generation, DL model training, and hyperparameter tuning to identify optimal neural network architectures utilized high-performance computing resources while the DL inference is performed on a standard laptop. Approximately, O(10 5 ) processor hours are used for generating data and DL tuning and training. We acknowledge that the data generation and DL model development are expensive. But once a DL model is trained, it can be re-used for inversion rapidly for the given system, with set physics and domain. Note that traditional inversion may require multiple forward model simulations (e.g., in the order of 10 to 1000), which are very expensive. This computational savings ≈ O(10 5 ) – O(10 7 )) makes the proposed DL-based inverse model attractive for subsurface imaging and real-time ERT monitoring applications due to fast and yet reasonably accurate estimations of permeability field.

58 GEOSCIENCES↗
DOE OSTI2024

Evidential Deep Learning for Probabilistic Modelling of Extreme Storm Events

Uncertainty quantification (UQ) methods play an important role in reducing errors in weather forecasting. Conventional approaches in UQ for weather forecasting rely on generating an ensemble of forecasts from physics-based simulations to estimate the uncertainty. However, it is computationally expensive to generate many forecasts to predict real-time extreme weather events. Evidential Deep Learning (EDL) is an uncertainty-aware deep learning approach designed to provide confidence about its predictions using only one forecast. It treats learning as an evidence acquisition process where more evidence is interpreted as increased predictive confidence. We apply EDL to storm forecasting using real-world weather datasets and compare its performance with traditional methods. Our findings indicate that EDL not only reduces computational overhead but also enhances predictive uncertainty. This method opens up novel opportunities in research areas such as climate risk assessment, where quantifying the uncertainty about future climate is crucial.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

High-Fidelity Simulation Aerodynamics Dataset of NACA 0012 and 0021 Airfoils

The repository contains time series data of pressure, viscous and moment forces for the NACA 0012 and 0021 airfoils at angles of attack (AOA), 5, 17, 30, 45, 60 and 90 degrees. The data was generated during the high-fidelity simulations of the airfoils using the open-source CFD code, Nalu-Wind (https://github.com/Exawind/nalu-wind). The research objective was to study the effects of mesh resolution and turbulence model on the three-dimensional deep stall aerodynamics of the airfoils, which has been published in the Journal of Turbulence (https://doi.org/10.1080/14685248.2023.2225141). Several meshes of the O-grid type were generated using the modified Pointwise Glyph script of Carrigan (https://github.com/pointwise/AirfoilMesh) by varying the wall normal and spanwise resolutions. Note that, the geometry was extruded by four chord lengths in the spanwise direction. In the paper, we showed that Improved Delayed Detached Eddy Simulation (IDDES) hybrid RANS/LES turbulence model with a minimum of 24 cells per chord length in the spanwise direction is necessary to correctly predict the loads in the deep stall regime. Under the NACA 0012 airfoil directory, the data for several combinations of wall-normal and spanwise resolutions are presented in each of the AOA subdirectories. The corresponding experimental data for the lift and drag polars can be found in the "exp_data" subdirectory. The directory for the NACA 0021 airfoil consists of CFD force data and airfoil polars for Reynolds numbers 2.7e5 and 2.0e6.

17 WIND ENERGY↗
DOE OSTI2024

Multi-agent voltage control in distribution systems using GAN-DRL-based approach

Active distribution grids can experience voltage fluctuations and violations due to the high penetration of variable distributed energy resources (DERs). These problems might occur because of the uncertain and variable generation natures of these resources, especially solar photovoltaic resources, during panel shadowing scenarios. Volt-VAR control (VVC) is an efficient method that controls the reactive power set-points of the inverters to regulate the voltage of distribution grids. Although several VVC approaches have been proposed recently, the performance of these approaches degrades significantly if behind-the-meter solar generation data are unobservable/missing. Therefore, it is necessary to impute missing/unobservable PV data accurately to be utilized in VVC approaches. Further, this paper proposes a model-free, data-driven, centrally trained, and decentrally executed multi-agent deep reinforcement learning-based VVC architecture to regulate the voltage of distribution networks. A generative adversarial network (GAN) is incorporated to impute the unobservable PV data accurately, which improves the performance of the proposed control architecture. The proposed multi-agent-soft-actor–critic algorithm (MASAC)-based VVC technique utilizes the actual PV dataset as well as the imputed dataset from the GAN framework to learn the optimal coordinated control policy for controlling the optimal reactive power set-points of PV inverters. The effectiveness of the proposed approach is analyzed on a modified IEEE 34-bus test case with added PV inverters. The results are compared and analyzed with a base case model with no VVC and VVC with a local droop control approach, genetic algorithm optimization, and a centralized soft actor–critic-based approach. Moreover, the performance of the proposed approach is compared with that of a multi-agent VVC framework without using the PV generation data and load information as the system state. The results illustrate that the proposed method with more state input improves the voltage profile and reduces the power loss of the network across various loading and PV generation scenarios.

14 SOLAR ENERGY↗