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.

SourceResearch connectionFrom yearThrough year

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

At least 739 records · Page 41

All sources NASA NTRS DOE OSTI

DOE OSTI2024

3D-Reconstruction of Tau Neutrinos in LArTPC Detectors

The Deep Underground Neutrino Experiment (DUNE) is a next-generation neutrino experiment currently under construction. DUNE will consist of two high-resolution neutrino interaction imaging detectors exposed to the world’s most intense neutrino beam, with the Near Detector at Fermilab and the Far Detector 1,300 km away in the Sanford Underground Research Facility in South Dakota, US. The high statistics and excellent resolution capabilities of DUNE's $^{40}$Ar detector will allow us to make precision studies of oscillation parameters capable of searching for CP violation in the lepton sector, testing interaction models, and studying phenomena that have until now, seemed too complex to measure, like $\nu_\tau$ detection and therefore, providing the completion of the 3-flavor neutrino paradigm. Knowledge of the $\nu_\tau$ detection can impact a broad spectrum of open questions. These include searching for non-standard neutrino interactions, constraining the unitarity of the PMNS matrix, searching for sterile neutrinos, and studying neutrino interactions. In the case of LArTPC data, the detector hits can be considered nodes in a graph, and the edges represent the spatial and temporal relationships between them. By using graph neural networks, it is possible to exploit these relationships and improve the accuracy of particle identification and reconstruction. During my presentation and specifically for tau neutrino reconstruction, I will show the effectiveness and reliability of our in-house developed graph neural network (GNN), NuGraph. This GNN classifies detector hits based on the particle type responsible for their production, assuring that the system accurately identifies and categorizes information based on its unique characteristics.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2022

Deep reaction network exploration at a heterogeneous catalytic interface

Characterizing the reaction energies and barriers of reaction networks is central to catalyst development. However, heterogeneous catalytic surfaces pose several unique challenges to automatic reaction network characterization, including large sizes and open-ended reactant sets, that make ad hoc network construction the current state-of-the-art. Here, we show how automated network exploration algorithms can be adapted to the constraints of heterogeneous systems using ethylene oligomerization on silica-supported single-site Ga 3+ as a model system. Using only graph-based rules for exploring the network and elementary constraints based on activation energy and size for identifying network terminations, a comprehensive reaction network is generated and validated against standard methods. The algorithm (re)discovers the Ga-alkyl-centered Cossee-Arlman mechanism that is hypothesized to drive major product formation while also predicting several new pathways for producing alkanes and coke precursors. These results demonstrate that automated reaction exploration algorithms are rapidly maturing towards general purpose capability for exploratory catalytic applications.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2023

Physics informed deep neural network embedded in a chemical transport model for the Amazon rainforest

Secondary organic aerosols (SOA) are fine particles in the atmosphere, which interact with clouds, radiation and affect the Earth’s energy budget. SOA formation involves chemistry in gas phase, aqueous aerosols, and clouds. Simulating these chemical processes involve solving a stiff set of differential equations, which are computationally expensive steps for three-dimensional chemical transport models. Deep neural networks (DNNs) are universal function approximators that could be used to represent the complex nonlinear changes in aerosol physical and chemical processes; however, key challenges such as generalizability to extended time periods, preservation of mass balance, simulating sparse model outputs, and maintaining physical constraints have limited their use in atmospheric chemistry. Here, we develop an approach of using a physics-informed DNN that overcomes previous such challenges and demonstrates its applicability for the chemical formation processes of isoprene epoxydiol SOA (IEPOX-SOA) over the Amazon rainforest. The DNN is trained with data generated by simulating IEPOX-SOA over the entire atmospheric column, using the Weather Research and Forecasting Model coupled with Chemistry (WRF-Chem). The trained DNN is then embedded within WRF-Chem to replace the computationally expensive default solver of IEPOX-SOA formation. The trained DNN predictions generalizes well with the default model simulation of the IEPOX-SOA mass concentrations and its size distribution (20 size bins) over several days of simulations in both dry and wet seasons. The embedded DNN reduces the computational expense of WRF-Chem by a factor of 2. Our approach shows promise in terms of application to other computationally expensive chemistry solvers in climate models.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2021

Learning the structure of wind: A data-driven nonlocal turbulence model for the atmospheric boundary layer

In this work, we develop a novel data-driven approach to modeling the atmospheric boundary layer. This approach leads to a nonlocal, anisotropic synthetic turbulence model which we refer to as the deep rapid distortion (DRD) model. Our approach relies on an operator regression problem that characterizes the best fitting candidate in a general family of nonlocal covariance kernels parameterized in part by a neural network. This family of covariance kernels is expressed in Fourier space and is obtained from approximate solutions to the Navier–Stokes equations at very high Reynolds numbers. Each member of the family incorporates important physical properties such as mass conservation and a realistic energy cascade. The DRD model can be calibrated with noisy data from field experiments. After calibration, the model can be used to generate synthetic turbulent velocity fields. To this end, we provide a new numerical method based on domain decomposition which delivers scalable, memory-efficient turbulence generation with the DRD model as well as others. We demonstrate the robustness of our approach with both filtered and noisy data coming from the 1968 Air Force Cambridge Research Laboratory Kansas experiments. Using these data, we witness exceptional accuracy with the DRD model, especially when compared to the International Electrotechnical Commission standard.

17 WIND ENERGY↗

DOE OSTI2024

Optical potential uncertainties on exclusive tens-of-MeV Ar cross sections

Large liquid argon time projection chamber (LArTPC) neutrino detectors, such as those planned for the Deep Underground Neutrino Experiment (DUNE), show considerable promise as a platform for next-generation measurements of supernova neutrinos. Thanks to the neutron excess in $^{40}$Ar as well as the detailed tracking possible with LArTPCs, these detectors are expected to be uniquely capable of measuring supernova electron neutrinos with high statistics and minimal backgrounds. However, these technological advantages come at the price of complexities in data interpretation; reconstruction of the incident energies of supernova neutrinos in a future LArTPC-based analysis will be subject to a variety of systematic uncertainties related to nuclear interaction modeling. In this talk, we present a study of a subset of these uncertainties as implemented in the MARLEY event generator used by DUNE and other LArTPC neutrino experiments. Variations to the optical potential used in the MARLEY nuclear de-excitation model are applied to the calculation of exclusive tens-of-MeV neutrino-nucleus cross sections for the first time, and the impact on observables of interest for supernova neutrino detection is examined. We also discuss prospects for quantifying other sources of uncertainty on the MARLEY neutrino interaction model.

El-Haj, Luca Abu↗

DOE OSTI2022

Invertible Neural Networks for Airfoil Design

We report the airfoil design problem, in which an engineer seeks a shape with desired performance characteristics, is fundamental to aerodynamics. Design workflows traditionally rely on iterative optimization methods using low-fidelity integral boundary-layer methods as higher-fidelity adjoint-based computational fluid dynamics methods are computationally expensive. Surrogate-based approaches can accelerate the design process but still rely on some iterative inverse design procedure. In this work, we leverage emerging invertible neural network (INN) tools to enable the rapid inverse design of airfoil shapes for wind turbines. INNs are specialized deep-learning models with well-defined inverse mappings. When trained appropriately, INN surrogate models are capable of forward prediction of aerodynamic and structural quantities for a given airfoil shape as well as inverse recovery of airfoil shapes with specified aerodynamic and structural characteristics. The INN approach offers a roughly 100 times speed-up compared to adjoint-based methods for inverse design. We demonstrate the INN tool for inverse design on three test cases of 100 airfoils each that satisfy the performance characteristics close to those of airfoils used in wind-turbine blades. All generated shapes satisfy the desired aerodynamic characteristics, demonstrating the success of the INN approach for inverse design of airfoils.

17 WIND ENERGY↗

DOE OSTI2021

Deep learning for Alzheimer's disease: Mapping large-scale histological tau protein for neuroimaging biomarker validation

Abnormal tau inclusions are hallmarks of Alzheimer's disease and predictors of clinical decline. Several tau PET tracers are available for neurodegenerative disease research, opening avenues for molecular diagnosis in vivo. However, few have been approved for clinical use. Understanding the neurobiological basis of PET signal validation remains problematic because it requires a large-scale, voxel-to-voxel correlation between PET and (immuno) histological signals. Large dimensionality of whole human brains, tissue deformation impacting co-registration, and computing requirements to process terabytes of information preclude proper validation. We developed a computational pipeline to identify and segment particles of interest in billion-pixel digital pathology images to generate quantitative, 3D density maps. The proposed convolutional neural network for immunohistochemistry samples, IHCNet, is at the pipeline's core. We have successfully processed and immunostained over 500 slides from two whole human brains with three phospho-tau antibodies (AT100, AT8, and MC1), spanning several terabytes of images. Our artificial neural network estimated tau inclusion from brain images, which performs with ROC AUC of 0.87, 0.85, and 0.91 for AT100, AT8, and MC1, respectively. Introspection studies further assessed the ability of our trained model to learn tau-related features. We present an end-to-end pipeline to create terabytes-large 3D tau inclusion density maps co-registered to MRI as a means to facilitate validation of PET tracers.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2021

Mountain waves can impact wind power generation

Abstract. Mountains can modify the weather downstream of the terrain. In particular, when stably stratified air ascends a mountain barrier, buoyancy perturbations develop. These perturbations can trigger mountain waves downstream of the mountains that can reach deep into the atmospheric boundary layer where wind turbines operate. Several such cases of mountain waves occurred during the Second Wind Forecast Improvement Project (WFIP2) in the Columbia River basin in the lee of the Cascade Range bounding the states of Washington and Oregon in the Pacific Northwest of the United States. Signals from the mountain waves appear in boundary layer sodar and lidar observations as well as in nacelle wind speeds and power observations from wind plants. Weather Research and Forecasting (WRF) model simulations also produce mountain waves and are compared to satellite, lidar, and sodar observations. Simulated mountain wave wavelengths and wave propagation speeds (group velocities) are analyzed using the fast Fourier transform. We found that not all mountain waves exhibit the same speed and conclude that the speed of propagation, magnitudes of wind speeds, or wavelengths are important parameters for forecasters to recognize the risk for mountain waves and associated large drops or surges in power. When analyzing wind farm power output and nacelle wind speeds, we found that even small oscillations in wind speed caused by mountain waves can induce oscillations between full-rated power of a wind farm and half of the power output, depending on the position of the mountain wave's crests and troughs. For the wind plant analyzed in this paper, mountain-wave-induced fluctuations translate to approximately 11 % of the total wind farm output being influenced by mountain waves. Oscillations in measured wind speeds agree well with WRF simulations in timing and magnitude. We conclude that mountain waves can impact wind turbine and wind farm power output and, therefore, should be considered in complex terrain when designing, building, and forecasting for wind farms.

17 WIND ENERGY↗

DOE OSTI2023

High Performance Adaptive Deep-Reinforcement-Learning-based Real-time Emergency Control (HADREC) to Enhance Power Grid Resilience in Stochastic Environment (Final Report)

The HADREC technology developed in this project features real-time, validated, and predictive emergency control capabilities needed to effectively safeguard the grid against costly disturbances from extreme weather and other disruptive events. It has the potential to revolutionize the emergency control by providing faster and more effective solution for power grid emergency control. The technology can reduce the grid emergency control action time from minutes to seconds and mitigate the impact of outages and reduce system recovery time. The evaluation results using the models and data from PacifiCorp system have shown ~20% improvement in responding time and 70% improved in tripped generator output, comparing against today’s industrial application. This performance can help reduce preventive measurements and improve network utilization and market efficiency, and help lead to a more reliable and resilient power grid with renewable integration. The project has generated significant interest from industry, academic and commercial companies.

24 POWER TRANSMISSION AND DISTRIBUTION↗

DOE OSTI2023

A Process Model for ITCZ Narrowing under Warming Highlights Clear-Sky Water Vapor Feedbacks and Gross Moist Stability Changes in AMIP Models

Tropical areas with mean upward motion—and as such the zonal-mean intertropical convergence zone (ITCZ)—are projected to contract under global warming. To understand this process, a simple model based on dry static energy and moisture equations is introduced for zonally symmetric overturning driven by sea surface temperature (SST). Processes governing ascent area fraction and zonal mean precipitation are examined for insight into Atmospheric Model Intercomparison Project (AMIP) simulations. Bulk parameters governing radiative feedbacks and moist static energy transport in the simple model are estimated from the AMIP ensemble. Uniform warming in the simple model produces ascent area contraction and precipitation intensification—similar to observations and climate models. Contributing effects include stronger water vapor radiative feedbacks, weaker cloud-radiative feedbacks, stronger convection-circulation feedbacks, and greater poleward moisture export. The simple model identifies parameters consequential for the inter-AMIP-model spread; an ensemble generated by perturbing parameters governing shortwave water vapor feedbacks and gross moist stability changes under warming tracks inter-AMIP-model variations with a correlation coefficient ~0.46. Here, the simple model also predicts the multimodel mean changes in tropical ascent area and precipitation with reasonable accuracy. Furthermore, the simple model reproduces relationships among ascent area precipitation, ascent strength, and ascent area fraction observed in AMIP models. A substantial portion of the inter-AMIP-model spread is traced to the spread in how moist static energy and vertical velocity profiles change under warming, which in turn impact the gross moist stability in deep convective regions—highlighting the need for observational constraints on these quantities.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2026

Harnessing Ocean Thermal Gradients Using Thermoelectric Based Submersibles for Ocean Power Applications

The urgent need for energy solutions in marine environments has accelerated the development of innovative technologies capable of leveraging natural resources for power generation. This study introduces a buoyancy-driven submersible system designed to harness ocean thermal gradients using thermoelectric generators (TEGs) and phase change materials (PCMs). The technology aims to provide autonomous power to offshore aquaculture farms, unmanned underwater vehicles (UUVs), offshore platform illumination, and ocean sensors, significantly reducing dependence on fossil fuels. Ocean thermal gradients, especially prevalent in mid-latitude regions, exhibit temperature differences between surface and deep waters ranging from 7 degrees Celsius to 30 degrees Celsius depending on seasonal variations. The proposed submersible technology utilizes TEGs to convert thermal energy from these gradients into electrical power, generating between 0.2 and 0.5 watts, while PCMs are employed to store and regulate this energy, ensuring a stable and continuous power supply. The buoyancy-driven mechanism of the submersible enhances its capability to navigate through varying depths, optimizing its exposure to different thermal gradients and maximizing energy harvesting. The performance of this submersible system is analyzed through detailed thermodynamic assessments and computational fluid dynamics (CFD) modeling focused on heat transfer. These analyzes consider real-world ocean temperature profiles and seek to refine the interaction between TEGs and PCMs to optimize energy extraction. The evaluation encompasses several key performance metrics, including power output and energy efficiency. Results confirm the potential of this innovative technology to provide a continuous and reliable power source for marine applications. By demonstrating the feasibility of using ocean thermal gradients for energy generation, this study contributes to the broader efforts of innovation in energy technologies for harsh, remote marine environments. The implementation of such promises is significant advancements in the autonomy of marine operations. The ongoing research will further investigate scalability ensuring that the system can be effectively adapted to various marine settings and operational demands.

16 TIDAL AND WAVE POWER↗

DOE OSTI2025

Vehicular Re-Identification from Uncontrolled Multiple Views

Vehicle re-identification (re-ID) across disparate sensing modalities remains a fundamental challenge for transportation research. In this work, we introduce a deep multi-view vehicle re-ID framework that leverages Siamese networks to compare pairs of vehicle images and produce matching scores, enabling robust association across drastically different viewpoints such as those from UAVs, surveillance cameras, and ground sensors. The model exploits convolutional neural networks to learn features that remain discriminative under changes in angle, distance, and illumination, supporting more generalizable re-ID performance. As part of this effort, we also developed an automated pipeline to synchronize roadside and UAV video streams, producing a multi-perspective dataset that complements preexisting real collections and a synthetic dataset generated in this study. Together, these contributions advance the capability to re-identify vehicles across wide viewing baselines; establish a foundation for scalable, reproducible research in vehicle re-ID; and open pathways for future applications, such as inferring routine behaviors, movement patterns, and daily habits of the individual associated with the vehicle.

convolutional neural networks↗

DOE OSTI2020

Enforcing constraints for time series prediction in supervised, unsupervised and reinforcement learning

We assume that we are given a time series of data from a dynamical system and our task is to learn the flow map of the dynamical system. We present a collection of results on how to enforce constraints coming from the dynamical system in order to accelerate the training of deep neural networks to represent the flow map of the system as well as increase their predictive ability. In particular, we provide ways to enforce constraints during training for all three major modes of learning, namely supervised, unsupervised and reinforcement learning. In general, the dynamic constraints need to include terms which are analogous to memory terms in model reduction formalisms. Such memory terms act as a restoring force which corrects the errors committed by the learned flow map during prediction. For supervised learning, the constraints are added to the objective function. For the case of unsupervised learning, in particular generative adversarial networks, the constraints are introduced by augmenting the input of the discriminator. Finally, for the case of reinforcement learning and in particular actor-critic methods, the constraints are added to the reward function. In addition, for the reinforcement learning case, we present a novel approach based on homotopy of the action-value function in order to stabilize and accelerate training. We use numerical results for the Lorenz system to illustrate the various constructions.

Stinis, Panagiotis↗

DOE OSTI2024

Application of passive vortex generators to enhance vertical mixing in an open raceway pond

A novel use of vortex generators in a raceway pond is described here, with the flow field quantitatively simulated using computational fluid dynamics using the large eddy simulation turbulence model. Persistence lengths of the swirling motion generated by the vortex generators indicate that significant vertical mixing can be achieved by placing vortex generators in the straight section opposite the paddle wheel, downstream of the first hairpin bend. Relatively simple vortex generators are capable of creating stronger swirling motions that persist for a longer distance than those created by the paddle wheel. For optimal performance, vortex generators are positioned side by side but in opposite directions, and their diameters should be equal to or slightly less than the desired liquid depth. The optimal length of a 0.18 m diameter vortex generator in a 0.2 m deep pond was determined to be 0.3 m. As a result, it has been demonstrated that a longer persistence length is achieved by inducing a swirling motion with its rotational axis parallel to the primary flow direction.

09 BIOMASS FUELS↗

DOE OSTI2025

Uncertainty based Online Ensemble on Non-Stationary Data for Fusion Science

Machine Learning (ML) is poised to play a pivotal role in the development and operation of next-generation fusion devices. Fusion data shows non-stationary behavior due to drifts in the data. The drifts can arise from both experimental evolution and machine wear-and-tear. ML models assume stationary distribution and fail to maintain performance when encountered with non-stationary data streams.Online learning can be used to continuously adapt the models with new data as it is acquired. However, traditional online learning can suffer from short-term performance degradation, as ground truth are not available before making the prediction. To address this challenge, we propose uncertainty aware ensemble approach for online learning. We use Deep Gaussian Process Approximation (DGPA) technique for calibrated uncertainty estimation and use the uncertainty values to guide a meta-algorithm that produces predictions based on ensemble of learners. Moreover, DGPA also provides uncertainty estimation along with the predictions for decision makers. This paper demonstrates that the proposed method outperforms traditional online learning approach, and a naive ensemble without uncertainty guidance by about 7% and 6%, respectively, on B-coil deflection prediction at DIII-D Fusion Facility.

Rajput, Kishansingh [Thomas Jefferson National Acc↗

DOE OSTI2025

Uncertainty based Online Ensemble on Non-Stationary Data for Fusion Science

Rajput, Kishansingh [Thomas Jefferson National Acc↗

DOE OSTI2024

Deep learning-driven super-resolution in Raman hyperspectral imaging: Efficient high-resolution reconstruction from low-resolution data

Deep learning (DL) has become an indispensable tool in hyperspectral data analysis, automatically extracting valuable features from complex, high-dimensional datasets. Super-resolution reconstruction, an essential aspect of hyperspectral data, involves enhancing spatial resolution, particularly relevant to low-resolution hyperspectral data. Yet, the pursuit of super-resolution in hyperspectral analysis is fraught with challenges, including acquiring ground truth high-resolution data for training, generalization, and scalability. The pressing issue of extended spectral acquisition times, notably for high-resolution scans, is a significant roadblock in hyperspectral imaging. Super-resolution methods offer a promising solution by providing higher spatial resolution data to expedite data collection and yield more efficient outcomes. This paper delves into a practical application of these concepts using Raman imaging, where spectral acquisition times can be prohibitively long. In this context, DL-based super-resolution models demonstrate their efficacy by predicting and reconstructing high-resolution Raman data from low-resolution input, eliminating the need for resource-intensive high-resolution scans. While previous work often relied on substantial high-resolution datasets, this study showcases the ability to achieve similar outcomes even with limited data, presenting a more practical and cost-effective approach. In conclusion, the results offer a glimpse into the transformative potential of this technology to streamline hyperspectral imaging applications by saving valuable time and resources through the successful generation of high-resolution data from low-resolution inputs.

42 ENGINEERING↗

DOE OSTI2023

Peri-Net-Pro: the neural processes with quantified uncertainty for crack patterns

Abstract This paper develops a deep learning tool based on neural processes (NPs) called the Peri-Net-Pro, to predict the crack patterns in a moving disk and classifies them according to the classification modes with quantified uncertainties. In particular, image classification and regression studies are conducted by means of convolutional neural networks (CNNs) and NPs. First, the amount and quality of the data are enhanced by using peridynamics to theoretically compensate for the problems of the finite element method (FEM) in generating crack pattern images. Second, case studies are conducted with the prototype microelastic brittle (PMB), linear peridynamic solid (LPS), and viscoelastic solid (VES) models obtained by using the peridynamic theory. The case studies are performed to classify the images by using CNNs and determine the suitability of the PMB, LBS, and VES models. Finally, a regression analysis is performed on the crack pattern images with NPs to predict the crack patterns. The regression analysis results confirm that the variance decreases when the number of epochs increases by using the NPs. The training results gradually improve, and the variance ranges decrease to less than 0.035. The main finding of this study is that the NPs enable accurate predictions, even with missing or insufficient training data. The results demonstrate that if the context points are set to the 10th, 100th, 300th, and 784th, the training information is deliberately omitted for the context points of the 10th, 100th, and 300th, and the predictions are different when the context points are significantly lower. However, the comparison of the results of the 100th and 784th context points shows that the predicted results are similar because of the Gaussian processes in the NPs. Therefore, if the NPs are employed for training, the missing information of the training data can be supplemented to predict the results.

Mathematics↗