SEARCH · Engineering Papers

Results for “deep potential”

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 73 records · Page 4

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Triangle Method for Dense ReLU Layers [SWR-25-72]

This software is an implementation of the methods for initializing and training neural networks to be more efficient per parameter, described more fully below and in the related publication: In theory, depth should make a ReLU network EXPONENTIALLY more efficient by enabling it to produce an exponential number of piecewise linear sections in its output. This reasoning is largely based on the work of mathematicians that have hand-constructed networks that make good use of depth. In practice however, even very deep ReLU networks that have been randomly initialized will behave identically to their shallow counterparts - missing an entire exponential dimension of efficiency. The triangle method is a first attempt at realizing the exponential potential of deep networks. Instead of randomly setting weights, we force pairs of neurons in each layer learn to build triangles (i.e. functions from [0,1] -> [0,1] that look like triangles). This is a very efficient pattern for generating lots of linear pieces because composing two triangular functions doubles the number of pieces with each composition. The triangle method is more than just a different initialization, it is a new paradigm of training. Instead of making direct updates to the matrix weights, we do an extra step of backpropagation to collect the derivatives of the loss function with respect to the shapes of the triangles, training them to tilt left or right. This process essentially holds the networks hand throughout the loss landscape and forces it to always use depth effectively by producing triangular shapes internally. This can produce several orders of magnitude of improvement on convex one-dimensional regression problems. Much more theoretical work is needed to realize its full potential beyond this context, but the implementation in this repository will still work in arbitrary numbers of dimensions. The file Triangle_Method.py is a generalized form of the method that will build each neuron its own custom 1-d convex activation function (with exponential efficiency). Example usage on one dimensional problems can be found in Example_Usage.ipynb and an example of using this in a real neural network can be found in Example_VGG16_CIFAR10.ipynb.

Milkert, Max [National Renewable Energy Laboratory↗

DOE OSTI2025

Applying Deep Learning for Wildfire Identification: Economical and Accessible Solutions Leveraging Small Datasets

Wildfires significantly impact human health, air quality, visibility, weather, and climate change and cause substantial economic losses. While state and county-operated air quality monitors provide critical insights during wildfires, they are not available in all regions. This highlights the need for affordable, accessible tools that allow the general public to assess air quality impacts. In this study, we apply machine learning with deep neural networks to diagnose air quality rapidly from sky images taken at the Pacific Northwest National Laboratory in Richland, WA, USA. Using a convolutional neural network (CNN) framework, we trained a deep learning model to classify air quality indices based on sky images. By leveraging transfer learning, our approach fine-tunes a pre-trained model on a small dataset of sky images, significantly reducing training time while maintaining high accuracy. Our results demonstrate the potential of deep learning to provide rapid air quality diagnostics during wildfire episodes, offering early warnings to the public and enabling timely mitigation strategies, particularly for vulnerable populations. Additionally, we show that lower respiratory infections pose the highest health risk during acute smoke exposures. Reactive oxygen species (ROS) from wildfire particles further exacerbate health risks by triggering inflammation and other adverse effects.

54 ENVIRONMENTAL SCIENCES↗

NASA NTRS1986

System concepts and design examples for optical communication with planetary spacecraft

Systems concepts for optical communication with future deep-space (planetary) spacecraft are described. These include not only the optical transceiver package aboard the distant spacecraft, but the earth-vicinity optical-communications receiving station as well. Both ground-based, and earth-orbiting receivers are considered. Design examples for a number of proposed or potential deep-space missions are then presented. These include an orbital mission to Saturn, a Lander and Rover mission to Mars, and an astronomical mission to a distance of 1000 astronomical units.

Lesh, James R.↗

NASA NTRS2018

TPSAS-NF1676L-32014-DND

The Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP), on-board the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) is a satellite-borne polarization sensitive lidar. It has been providing the vertical distributions of clouds and aerosols along with their microphysical and optical properties since 2006. One of its important Level 2 products, feature classification, has been determined using the lidar information from 532 nm parallel and perpendicular channels, and 1064 nm channel measurements of layer integrated backscatter. Deep machine learning methods which combine both the channel and texture information to recognize feature patterns is uniquely beneficial when applied to this data. In this study, we will use Convolutional Neural Network (CNN), a deep machine learning method, to classify lidar aerosol subtypes by using the lidar profile observations. This method uses additional information from the vertical texture of the feature instead of using only the layer information. Note that in the integrated layer properties, the texture information has been masked due to averaging. Our results will show how the texture information plays a role in the classification. This preliminary work explores the benefits and potential of deep machine learning methods for lidar retrievals and focuses on the aerosol subtype classification. The broader application extends to the classification of other feature types. Future applications include the developing deep machine learning methods with neural networks to retrieve properties of the features, and studies of indirect effect of cloud-aerosol interaction from lidar measurements.

Shan Zeng Kowalski↗

NASA NTRSDate not supplied

Towards Surrogate Modeling of Subgrid Turbulent Transport for 3D Radiative Hydrodynamic Simulations of the Quiet Sun

In this work, we investigate the use of deep learn-ing techniques as surrogate models, to enhance the estimationof effects of subgrid turbulent transport for 3D radiatuve hy-drodynamic simulations of the quiet Sun. We develop two dis-tinct 3D Convolutional Neural Networks (3DCNNs) to capturespatio-temporal dependencies in 3D velocity fields, leveragingdifferent activation functions and architectural designs. Thesemodels integrate both averaged velocity vector components andscalar features such as plasma density to enhance predictionaccuracy. Additionally, a Multilayer Perceptron (MLP) modelis employed to approximate complex nonlinear relationships,offering a comparison in performance between convolutionaland fully connected architectures. Logarithmic transformationis applied to the targets to handle heavily skewed data, im-proving model performance. All models are compared againsta physics-based Gradient Model. Results show that the 3DCNNmodels excel at approximating Reynolds stress tensors, makingthem a candidate for assisting in producing reduced resolutionsimulations, and thereby reducing computational overheadwhile maintaining higher accuracy than the baseline. Thesefindings demonstrate the potential of deep learning, particu-larly CNNs, to advance scalable and accurate simulations ofsolar dynamics, offering a promising alternative to traditionalturbulence models.

Heliophysics↗

DOE OSTI2024

Accurate and Data‐Efficient Micro X‐ray Diffraction Phase Identification Using Multitask Learning: Application to Hydrothermal Fluids

Traditional analysis of highly distorted micro X‐ray diffraction (μ‐XRD) patterns from hydrothermal fluid environments is a time‐consuming process, often requiring substantial data preprocessing and labeled experimental data. Herein, the potential of deep learning with a multitask learning (MTL) architecture to overcome these limitations is demonstrated. MTL models are trained to identify phase information in μ‐XRD patterns, minimizing the need for labeled experimental data and masking preprocessing steps. Notably, MTL models show superior accuracy compared to binary classification convolutional neural networks. Additionally, introducing a tailored cross‐entropy loss function improves MTL model performance. Most significantly, MTL models tuned to analyze raw and unmasked XRD patterns achieve close performance to models analyzing preprocessed data, with minimal accuracy differences. This work indicates that advanced deep learning architectures like MTL can automate arduous data handling tasks, streamline the analysis of distorted XRD patterns, and reduce the reliance on labor‐intensive experimental datasets.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

A New Platform for Twistronics: Perovskite Moiré Superlattices Beyond van der Waals

Moiré superlattices have reshaped condensed‐matter physics by enabling twist‐angle engineering of electronic and excitonic states, traditionally in van der Waals (vdW) systems. This Perspective focuses on perovskites as a transformative non‐van der Waals (non‐vdW) platform for twistronics. These materials are distinguished by ionic‐covalent bonding and soft octahedral lattices that support deep moiré potentials (≈100–200 meV), which in turn yield bright, room‐temperature excitons and tunable quantum phenomena. We organize recent advances into two core classes: (i) moiré superlattices constructed from 2D Ruddlesden‐Popper (RP) perovskites via spontaneous or guided twisted stacking; and (ii) moiré architectures in 3D perovskite lamellae, achieved through topotactic 2D RP‐to‐3D conversion and ligand‐induced self‐assembly. Key experimental observations in these systems include twist‐tunable moiré periods, suppressed exciton–exciton annihilation, flat band‐like transport, and ferroelectric vortex lattices in their oxide perovskite counterparts. Here, we outline key bottlenecks, including defects arising during phase conversion, stringent lattice‐matching requirements, and the absence of in situ twist metrology, and highlight opportunities in lattice‐mismatched heterostructures and multifunctional quantum‐optoelectronic devices. Collectively, these advances establish perovskite moiré systems as a critical bridge between fundamental moiré physics and room‐temperature technological applications.

2D Ruddlesden–Popper perovskites↗

DOE OSTI2025

Cu–Ni Oxidation Mechanism Unveiled: A Machine Learning-Accelerated First-Principles and in Situ TEM Study

Here, the development of accurate methods for determining how alloy surfaces spontaneously restructure under reactive and corrosive environments is a key, long-standing, grand challenge in materials science. Using machine learning-accelerated density functional theory and rare-event methods, in conjunction with in situ environmental transmission electron microscopy (ETEM), we examine the interplay between surface reconstructions and preferential segregation tendencies of CuNi(100) surfaces under oxidation conditions. Our modeling approach predicts that oxygen-induced Ni segregation in CuNi alloys favors Cu(100)-O c(2 × 2) reconstruction and destabilizes the Cu(100)-O (2√2 × √2)R45° missing row reconstruction (MRR). In situ ETEM experiments validate these predictions and show Ni segregation followed by NiO nucleation and growth in regions without MRR, with secondary nucleation and growth of Cu 2 O in MRR regions. Our approach based on combining disparate computational components and in situ ETEM provides a holistic description of the oxidation mechanism in CuNi, which applies to other alloy systems.

36 MATERIALS SCIENCE↗

DOE OSTI2025

Tuning water dissociation at oxide–electrolyte interfaces with electric fields

Understanding how electric fields influence water dissociation at heterogeneous interfaces is crucial for controlling interfacial chemical reactions and advancing next-generation energy technologies. Herein, ab initio–based machine learning simulations show that even small electric field changes can significantly alter the water dissociation fraction at planar TiO 2 –electrolyte interfaces. The resulting free energy difference between undissociated and dissociated interfacial water exhibits a linear dependence on the field change with a slope of 1.97 eÅ, which far exceeds the dissociation-induced dipole change of a water molecule. Employing a machine-learned collective variable to investigate the reaction statistics of thousands of water dissociation/recombination events, we find that small electric field changes exert minor effects on individual reaction energy barriers but significantly influence the populations of local configurations associated with initial states that are most favorable for reactions. These findings elucidate the pronounced impact of electric fields on interfacial water dissociation and reveal a mechanism for electric-field-controlled chemical reactions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2025

Diffusion of acceptor dopants in monoclinic 𝛽−Ga 2⁢ O 3

𝛽−Ga 2 ⁢O 3 is a promising material for next-generation power electronics because of its ultrawide band gap and high critical breakdown voltage. However, realizing its full potential requires precise control over dopant incorporation and stability. In this work, we use first-principles calculations to systematically assess the diffusion behavior of eight potential deep-level substitutional acceptors (Au, Ca, Co, Cu, Fe, Mg, Mn, and Ni) in 𝛽−Ga 2 ⁢O 3 . We consider two key diffusion mechanisms: (i) interstitial diffusion under nonequilibrium conditions relevant to ion implantation, and (ii) trap-limited diffusion (TLD) under near-equilibrium thermal annealing conditions. Our results reveal a strong diffusion anisotropy along the 𝑏 and 𝑐 axes, with dopant behavior governed by competition between diffusion and incorporation (or dissociation) activation energies. Under interstitial diffusion, Ca$^{2+}_{i}$ and Mg$^{2+}_{i}$ show the most favorable combination of low migration and incorporation barriers, making them promising candidates for efficient doping along the 𝑏 and 𝑐 axes, respectively. In contrast, Au$^{+}_{i}$ diffuses readily, but exhibits an incorporation barrier that exceeds 5 eV, rendering it ineffective as a dopant. From a thermal stability perspective, Co$^{2+}_{i}$ shows poor activation but high diffusion barriers, which may suppress undesirable migration at elevated temperatures. Under trap-limited diffusion, the dissociation of dopant-host complexes controls mobility. Mg$^{2+}_{i}$ again emerges as a leading candidate, exhibiting the lowest dissociation barriers along both axes, whereas Co$^{2+}_{i}$ and Fe$^{2+}_{i}$ display the highest barriers, suggesting improved dopant retention under thermal stress. In conclusion, our findings guide dopant selection by balancing activation and thermal stability, essential for robust semi-insulating substrates.

Defects↗

DOE OSTI2025

Solving key challenges in collider physics with foundation models

Foundation models are neural networks that are capable of simultaneously solving many problems. Large language foundation models like ChatGPT have revolutionized many aspects of daily life, but their impact for science is not yet clear. In this paper, we use a new foundation model for hadronic jets to solve three key challenges in collider physics. In particular, we show how experiments can (1) save significant computing power when developing reconstruction algorithms, (2) perform a complete uncertainty quantification for high-dimensional measurements, and (3) search for new physics with model agnostic methods using low-level inputs. In each case, there are significant computational or methodological challenges with current methods that limit the science potential of deep learning algorithms. By solving each problem, we take jet foundation models beyond proof-of-principle studies and into the toolkit of practitioners.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2025

Variational AutoEncoders Reveal Intensifying GPP Extremes in Continental United States based on CESM2 Simulations

Climate extremes significantly impact terrestrial carbon cycle dynamics, necessitating robust methods for detecting and analyzing anomalous behavior in plant productivity. This study presents a novel application of variational autoencoders (VAE) for identifying extreme events in gross primary productivity (GPP) from Community Earth System Model version 2 simulations across four AR6 regions in the Continental United States. We compare VAE-based anomaly detection with traditional singular spectral analysis (SSA) methods across three time periods: 1850-80, 1950-80, and 2050-80 under SSP5-8.5 scenario. The VAE architecture employs three dense layers and a latent space with input sequence length of 12 months, training on normalized GPP time series to reconstruct the GPP and identify anomalies based on reconstruction errors. Extreme events are defined using 5th percentile thresholds applied to both VAE and SSA anomalies. Results demonstrate strong regional agreement between VAE and SSA methods in spatial patterns of extreme event frequencies, despite VAE consistently producing higher threshold values (179-756 GgC for VAE vs. 100-784 GgC for SSA across regions and periods). Both methods reveal increasing magnitudes and frequencies of negative carbon cycle extremes toward 2050-80, particularly in Western and Central North America. The VAE approach shows comparable performance to established SSA techniques while offering computational advantages and enhanced capability for capturing non-linear temporal dependencies in carbon cycle variability. This research demonstrates the potential of deep learning approaches for extremes detection and provides a foundation for improved understanding of future carbon cycle risks under future conditions.

Sharma, Bharat [ORNL] (ORCID:0000000266982487)↗

DOE OSTI2025

Surrogate modeling of Cellular-Potts agent-based models as a segmentation task using the U-Net neural network architecture

The Cellular-Potts model is a powerful and ubiquitous framework for developing computational models for simulating complex multicellular biological systems. Cellular-Potts models (CPMs) are often computationally expensive due to the explicit modeling of interactions among large numbers of individual model agents and diffusive fields described by partial differential equations (PDEs). In this work, we develop a convolutional neural network (CNN) surrogate model using a U-Net architecture that accounts for periodic boundary conditions. We use this model to accelerate the evaluation of a mechanistic CPM previously used to investigate in vitro vasculogenesis. The surrogate model was trained to predict 100 computational steps ahead (Monte-Carlo steps, MCS), accelerating simulation evaluations by a factor of 562 times compared to single-core CPM code execution on CPU. Over short timescales of up to 3 recursive evaluations, or 300 MCS, our model captures the emergent behaviors demonstrated by the original Cellular-Potts model such as vessel sprouting, extension and anastomosis, and contraction of vascular lacunae. This approach demonstrates the potential for deep learning to serve as a step toward efficient surrogate models for CPM simulations, enabling faster evaluation of computationally expensive CPM simulations of biological processes.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS1983

X-rays from radio pulsars - The portable supernova remnants

Neutron stars are the longest-lived remnants of supernova explosions. As a reservoir of thermal energy remaining from the explosion and generated by frictional coupling between core and crust, as a storehouse of magnetic and rotational kinetic energy which allows the star to act as a high energy particle accelerator, and as the source of a deep gravitational potential which can generate heat from infalling matter, neutron stars remain capable of producing high energy radiation for a Hubble time. The results of an extensive survey of supernova remnants and radio pulsars performed with the imaging instruments on board the Einstein Observatory are reviewed and the implications of these results for pulsar physics and for the origin and evolution of galactic neutron stars are discussed.

Helfand, D. J.↗

NASA NTRS1992

Observation of the Coma cluster of galaxies with ROSAT during the all-sky survey

The Coma cluster of galaxies was observed with the position sensitive proportional counter (PSPC) during the ROSAT all sky survey. We find evidence for substructure in this cluster. Diffuse X-ray emission is detected from the regions of the NGC 4839 and 4911 subgroups at 6 percent and 1 percent of the total cluster emission respectively. There may be emission associated with the NGC 4874 and 4889 subgroups as well. The NGC 4839 group appears to be in the process of merging with the cluster. These X-ray data show that at least some of the groups previously found in projection are in fact physical objects possessing potential wells deep enough to trap their own X-ray gas. Because of the unlimited field of view of the all sky survey and the low background of the PSPC, we were able to measure the azimuthally averaged surface brightness of Coma out to approximately 100 arcmin, twice as far as was previously possible. Given the validity of our mass models, these new X-ray data imply that within 5/h(50) Mpc the binding mass of the Coma cluster is 1.8 +/- 0.6 x 10 exp 15/h(50) solar mass, and the fraction of cluster mass contained in hot gas is 0.30 +/- 0.14h(50) exp -3/2. Furthermore, the binding mass is more centrally concentrated than is the X-ray gas.

Briel, U. G.↗

NASA NTRS2000

Beamed and Unbeamed X-Ray Emission in FR1 Radio Galaxies

The research exploited ROSAT's sensitivity, together with its spatial and spectral resolution, to separate X-ray emission components in the sources. Prior to ROSAT, the dominant X-ray emission mechanism in radio galaxies as a class was unclear, with correlations between the X-ray and radio emission used on one hand to argue for a nuclear origin for the X-rays, and on the other hand for a thermal origin. Our observations (normally between 10 and 25 ks in length) routinely detected the target sources, and demonstrated that both resolved (thermal) and unresolved X-ray emission are typically present. Highlights of our work included two of the first detections of high-power radio galaxies at high redshift, 3C 280 and 3C 220.1. When combined with the work of two other groups, we find that of the 38 radio galaxies at z > 0.6 in the 3CRR sample, 12 were observed in ROSAT pointed observations and 9 were detected with the four most significant detections exhibiting source extent, including 3C 280 and 3C 220.1. Moreover, we discovered extended emission around five 3CRR quasars at redshift greater than about 0.4, one of which is at z > 0.6. Unification predicts that the X-ray environments of powerful radio galaxies and quasars should be similar, and our results show that powerful radio sources are finding some of the highest-redshift X-ray clusters known to date, pointing to deep gravitational potential wells early in the Universe.

Worrall, Diana M.↗

NASA NTRS2012

Microchip Non-Aqueous Capillary Electrophoresis (MicronNACE) Method to Analyze Long-Chain Primary Amines

A protocol was developed as a first step in analyzing the complex organic aerosols present on Saturn's moon Titan, as well as the analogues of these aerosols (tholins) made on Earth. Labeling of primary amines using Pacific Blue succinimidyl ester is effected in ethanol with 25 mM triethylamine to maintain basic conditions. This reaction is allowed to equilibrate for at least one hour. Separation of the labeled primary amines is performed in ethanol with 1.05 M acetic acid, and 50 mM ammonium acetate in a commercial two-layer glass device with a standard crossmicrochannel measuring 50 microns wide by 20 microns deep. Injection potentials are optimized at 2 kV from the sample (negative) to the waste well (positive), with slight bias applied to the other two wells ( 0.4 and 0.8 V) to pinch the injection plug for the 30-s injection. Separation is performed at a potential of 5 kV along the channel, which has an effective separation distance of 7 cm. The use of ethanol in this method means that long-chain primary amines can be dissolved. Due to the low pH of the separation buffer, electro-osmotic flow (EOF) is minimized to allow for separation of both short-chain and longchain amines. As the freezing point of ethanol is much lower than water, this protocol can perform separations at temperatures lower than 0 C, which would not be possible in aqueous phase. This is of particular importance when considering in situ sampling of Titan aerosols, where unnecessary heating of the sample (even to room temperature) would lead to decomposition or unpredictable side reactions, which would make it difficult to characterize the sample appropriately.

Willis, Peter A.↗

NASA NTRS2013

Advanced Stirling Radioisotope Generator Life Certification Plan

An Advanced Stirling Radioisotope Generator (ASRG) power supply is being developed by the Department of Energy (DOE) in partnership with NASA for potential future deep space science missions. Unlike previous radioisotope power supplies for space exploration, such as the passive MMRTG used recently on the Mars Curiosity rover, the ASRG is an active dynamic power supply with moving Stirling engine mechanical components. Due to the long life requirement of 17 years and the dynamic nature of the Stirling engine, the ASRG project faced some unique challenges trying to establish full confidence that the power supply will function reliably over the mission life. These unique challenges resulted in the development of an overall life certification plan that emphasizes long-term Stirling engine test and inspection when analysis is not practical. The ASRG life certification plan developed is described.

Rusick, Jeffrey J.↗