Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “generative diffusion”

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 343 records · Page 19

FORSE+: Simulating non-Gaussian CMB foregrounds at 3 arcmin in a stochastic way based on a generative adversarial network

We present FORSE+, a Python package that produces non-Gaussian diffuse Galactic thermal dust emission maps at arcminute angular scales and that has the capacity to generate random realizations of small scales. This represents an extension of the FORSE (Foreground Scale Extender) package, which was recently proposed to simulate non-Gaussian small scales of thermal dust emission using generative adversarial networks (GANs). With the input of the large-scale polarization maps from observations, FORSE+ has been trained to produce realistic polarized small scales at 3′ following the statistical properties, mainly the non-Gaussianity, of observed intensity small scales, which are evaluated through Minkowski functionals. Furthermore, by adding different realizations of random components to the large-scale foregrounds, we show that FORSE+ is able to generate small scales in a stochastic way. In both cases, the output small scales have a similar level of non-Gaussianity compared with real observations and correct amplitude scaling as a power law. These realistic new maps will be useful, in the future, to understand the impact of non-Gaussian foregrounds on the measurements of the cosmic microwave background (CMB) signal, particularly on the lensing reconstruction, de-lensing, and the detection of cosmological gravitational waves in CMB polarizationB-modes.

Astronomy & Astrophysics↗

Antimony diffusion in CdTe

Group V dopants may be used for next-generation high-voltage cadmium telluride (CdTe) solar photovoltaics, but fundamental defect energetics and kinetics need to be understood. Here, antimony (Sb) diffusion is studied in single-crystal and polycrystalline CdTe under Cd-rich conditions. Diffusion profiles are determined by dynamic secondary ion mass spectroscopy and analyzed with analytical bulk and grain-boundary diffusion models. Slow bulk and fast grain-boundary diffusion are found. Density functional theory is used to understand formation energy and mechanisms. Lastly, the theory and experimental results create new understanding of group V defect kinetics in CdTe.

14 SOLAR ENERGY↗

DiffLense: a conditional diffusion model for super-resolution of gravitational lensing data

Abstract Gravitational lensing data is frequently collected at low resolution due to instrumental limitations and observing conditions. Machine learning-based super-resolution techniques offer a method to enhance the resolution of these images, enabling more precise measurements of lensing effects and a better understanding of the matter distribution in the lensing system. This enhancement can significantly improve our knowledge of the distribution of mass within the lensing galaxy and its environment, as well as the properties of the background source being lensed. Traditional super-resolution techniques typically learn a mapping function from lower-resolution to higher-resolution samples. However, these methods are often constrained by their dependence on optimizing a fixed distance function, which can result in the loss of intricate details crucial for astrophysical analysis. In this work, we introduce DiffLense , a novel super-resolution pipeline based on a conditional diffusion model specifically designed to enhance the resolution of gravitational lensing images obtained from the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP). Our approach adopts a generative model, leveraging the detailed structural information present in Hubble space telescope (HST) counterparts. The diffusion model, trained to generate HST data, is conditioned on HSC data pre-processed with denoising techniques and thresholding to significantly reduce noise and background interference. This process leads to a more distinct and less overlapping conditional distribution during the model’s training phase. We demonstrate that DiffLense outperforms existing state-of-the-art single-image super-resolution techniques, particularly in retaining the fine details necessary for astrophysical analyses.

Computer Science↗

Probabilistic Diffusion Models Advance Extreme Flood Forecasting

Extreme floods pose escalating risks in a changing climate, yet forecasting remains challenging due to peak flow underestimation and high uncertainty. We introduce diffusion-based runoff model (DRUM), a probabilistic deep learning (DL) approach that advances extreme flood forecasting across representative basins in the contiguous United States. DRUM outperforms state-of-the-art benchmarks, enhancing nowcasting skill for the top 1‰ of flows in 72.3% of studied basins. Under operational scenarios, DRUM extends reliable lead times by nearly a full day for 20- and 50-year floods. When evaluated with measured precipitation, an ideal condition, recall improves by 0.3–0.4 and the early warning window extends by 2.3 days for 50-year floods. The enhancement potential varies regionally, with precipitation-driven flood zones in the eastern and northwestern US benefiting most, gaining 3–7 days in lead time. These findings highlight the transformative potential of diffusion models as a cutting-edge generative AI technique for advancing hydrology and broader Earth system sciences.

54 ENVIRONMENTAL SCIENCES↗

Denoising diffusion models with geometry adaptation for high fidelity calorimeter simulation

Simulation is crucial for all aspects of collider data analysis, but the available computing budget in the High Luminosity LHC era will be severely constrained. Generative machine learning models may act as surrogates to replace physics-based full simulation of particle detectors, and diffusion models have recently emerged as the state of the art for other generative tasks. Here, we introduce CaloDiffusion, a denoising diffusion model trained on the public CaloChallenge datasets to generate calorimeter showers. Our algorithm employs 3D cylindrical convolutions, which take advantage of symmetries of the underlying data representation. To handle irregular detector geometries, we augment the diffusion model with a new geometry latent mapping (GLaM) layer to learn forward and reverse transformations to a regular geometry that is suitable for cylindrical convolutions. The showers generated by our approach are nearly indistinguishable from the full simulation, as measured by several different metrics.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Modeling of the plasma generated in a rarefied hypersonic shock layer

In this study, a rigorous numerical model is developed to simulate the plasma generated in a rarefied, hypersonic shock layer. The model uses the direct simulation Monte Carlo (DSMC) method to treat the particle collisions and the particle-in-cell (PIC) method to simulate the plasma dynamics in a self-consistent manner. The model is applied to compute the flow along the stagnation streamline in front of a blunt body reentering the Earth's atmosphere at very high velocity. Results from the rigorous DSMC-PIC model are compared directly to the standard DSMC modeling approach that uses the ambipolar diffusion approximation to simulate the plasma dynamics. It is demonstrated that the self-consistent computation of the plasma dynamics using the rigorous DSMC-PIC model captures many physical phenomena not accurately predicted by the standard modeling approach. These computations represent the first assessment of the validity of the ambipolar diffusion approximation when predicting the rarefied plasma generated in a hypersonic shock layer.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Hydrogen-related complexes in Li-diffused ZnO single crystals

Zinc oxide (ZnO) is a wide band gap semiconductor and a potential candidate for next generation white solid state lighting applications. In this work, hydrogen-related complexes in lithium diffused ZnO single crystals were studied. In addition to the well-known Li-OH complex, several other hydrogen defects were observed. When a mixture of Li{sub 2}O and ZnO is used as the dopant source, zinc vacancies are suppressed and the bulk Li concentration is very high (>10{sup 19 }cm{sup −3}). In that case, the predominant hydrogen complex has a vibrational frequency of 3677 cm{sup −1}, attributed to surface O-H species. When Li{sub 2}CO{sub 3} is used, a structured blue luminescence band and O-H mode at 3327 cm{sup −1} are observed at 10 K. These observations, along with positron annihilation measurements, suggest a zinc vacancy–hydrogen complex, with an acceptor level ∼0.3 eV above the valence-band maximum. This relatively shallow acceptor could be beneficial for p-type ZnO.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Phase-field Modeling of Void Migration and Growth Kinetics in Materials under Irradiation and Temperature Field

A phase-field model is developed to investigate the migration of vacancies, interstitials, and voids as well as void growth kinetics in materials under radiation and temperature field. The model takes into account the generation of vacancies and interstitials associated with the irradiation damage, the recombination between vacancies and interstitials, defect diffusion, and defect sinks. The effect of void sizes, vacancy concentration, vacancy generation rate, recombination rate, and temperature gradient on a single void migration and growth kinetics is parametrically studied. The results demonstrate that the temperature gradient causes void migration and defect fluxes, i.e., the Soret effect, which affects void stability and growth kinetics. It is found that 1) the void migration mobility is independent of the void size, which is in agreement with the theoretical prediction with the assumption of bulk diffusion controlled migration; 2) the void migration mobility strongly depends on temperature gradient; and 3) the effect of defect concentration, generation rate, and recombination rate on void migration mobility is minor although they strongly influence the void growth kinetics.

36 MATERIALS SCIENCE↗

Comprehensive studies of response characteristics of organic photodetectors based on rubrene and C{sub 60}

We studied the transient response characteristics of organic photodetectors composing of high mobility materials of rubrene and C{sub 60}, respectively, as donor and acceptor. It was found that the response speed was limited by the delay time of both exciton diffusion and transit, and an anomalous phenomenon that the bandwidth decreases as the reverse bias increases was found for the first time. The investigation of frequency dependence at different device structures and light excitations demonstrated that the light absorption of rubrene prevents the photodetector from obtaining a fast response. With the help of magnetic field effect study, it was clearly elucidated that the slow diffusion time of the long lifetime triplet excitons generated from singlet fission in rubrene limited the bandwidth of the device. Moreover, the simulation of the response of photodetector under transient and steady state by exciton transport-diffusion equation showed that the exciton dissociation efficiency in rubrene increases more quickly than that in C{sub 60}, which should account for the negative dependence of bandwidth on the reverse bias in rubrene-based device.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Tropical Cyclone Super Resolution using conditional diffusion denoising probabilistic model from mesoscale simulation to LES

Accurate modeling of tropical cyclone wind fields is essential for the design, risk assessment, and operational planning of offshore energy infrastructure. While mesoscale simulations are widely used thanks to their computational efficiency, they lack the necessary resolution to capture key features such as wind shear and veer profiles as well as the distribution turbulent kinetic energy (TKE). High-fidelity large-eddy simulation (LES) models on the other hand, can resolve turbulent structures and provide a more accurate representation of the complex wind field, albeit at a higher computational cost. To address this modeling gap, we introduce a two-part generative framework to enhance the resolution and physics-capturing ability of mesoscale simulations. First, a reduced-order model based on Karhunen–Loève (KL) decomposition is used to extract dominant spatial modes from one-dimensional mean wind profiles. A multilayer perceptron (MLP) is trained to map mesoscale mode weights to their LES counterparts, enabling accurate reconstruction of vertical velocity profiles. Second, a conditional Diffusion Denoising Probabilistic Model (DDPM) is developed to super-resolve coarse and low-fidelity mesoscale velocity fields, recovering fine-scale turbulence structures and stress distributions. The framework is evaluated across different tropical cyclone intensity categories defined by the Saffir–Simpson scale and demonstrates strong performance in both interpolation and extrapolation tasks. The generated fields accurately reproduce spatial coherence, stress distributions, and spectral energy characteristics observed in LES data. By bridging the fidelity gap between mesoscale and LES outputs, this approach offers a scalable, data-driven solution for enhancing the representation of tropical cyclone wind fields, enabling more robust offshore energy infrastructure systems design in tropical-cyclone-prone areas.

17 WIND ENERGY↗

Debye scale turbulence within the electron diffusion layer during magnetic reconnection

During collisionless, anti-parallel magnetic reconnection, the electron diffusion layer is the region of both fieldline breaking and plasma mixing. Due to the in-plane electrostatic fields associated with collisionless reconnection, the inflowing plasmas are accelerated towards the X-line and form counter-streaming beams within the unmagnetized diffusion layer. This configuration is inherently unstable to in-plane electrostatic streaming instabilities provided that there is sufficient scale separation between the Debye length λ{sub D} and the electron skin depth c/ω{sub pe}. This scale separation has hitherto not been well resolved in kinetic simulations. Using both 2D fully kinetic simulations and a simple linear model, we demonstrate that these in-plane streaming instabilities generate Debye scale turbulence within the electron diffusion layer at electron temperatures relevant to magnetic reconnection both in the magnetosphere and in laboratory experiments.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

SOLAR PARITY ISSUE WITH FLUX-TRANSPORT DYNAMO

We investigated the dependence of the solar magnetic parity between the hemispheres on two important parameters, the turbulent diffusivity and the meridional flow, by means of axisymmetric kinematic dynamo simulations based on the flux-transport dynamo model. It is known that the coupling of the magnetic field between hemispheres due to turbulent diffusivity is an important factor for the solar parity issue, but the detailed criterion for the generation of the dipole field has not been investigated. Our conclusions are as follows. (1) The stronger diffusivity near the surface is more likely to cause the magnetic field to be a dipole. (2) The thinner layer of the strong diffusivity near the surface is also more apt to generate a dipolar magnetic field. (3) The faster meridional flow is more prone to cause the magnetic field to be a quadrupole, i.e., symmetric about the equator. These results show that turbulent diffusivity and meridional flow are crucial for the configuration of the solar global magnetic field.

79 ASTRONOMY AND ASTROPHYSICS↗

Diffusive shock acceleration and reconnection acceleration processes

Shock waves, as shown by simulations and observations, can generate high levels of downstream vortical turbulence, including magnetic islands. We consider a combination of diffusive shock acceleration (DSA) and downstream magnetic-island-reconnection-related processes as an energization mechanism for charged particles. Observations of electron and ion distributions downstream of interplanetary shocks and the heliospheric termination shock (HTS) are frequently inconsistent with the predictions of classical DSA. We utilize a recently developed transport theory for charged particles propagating diffusively in a turbulent region filled with contracting and reconnecting plasmoids and small-scale current sheets. Particle energization associated with the anti-reconnection electric field, a consequence of magnetic island merging, and magnetic island contraction, are considered. For the former only, we find that (i) the spectrum is a hard power law in particle speed, and (ii) the downstream solution is constant. For downstream plasmoid contraction only, (i) the accelerated spectrum is a hard power law in particle speed; (ii) the particle intensity for a given energy peaks downstream of the shock, and the distance to the peak location increases with increasing particle energy, and (iii) the particle intensity amplification for a particular particle energy, f(x,c/c{sub 0})/f(0,c/c{sub 0}), is not 1, as predicted by DSA, but increases with increasing particle energy. The general solution combines both the reconnection-induced electric field and plasmoid contraction. The observed energetic particle intensity profile observed by Voyager 2 downstream of the HTS appears to support a particle acceleration mechanism that combines both DSA and magnetic-island-reconnection-related processes.

79 ASTRONOMY AND ASTROPHYSICS↗

ADCP Upsample

This R script takes reads the csv file of stream-wise velocity data and generates a new time series of with comparable but distinct noise using the Langevin method and library. It simulates a time series measured at 1/10 the sampling rate by calculating a moving average of 10, though other sampling rates can be calculated by adjusting the moving average window. A stationary signal is calculated by taking the difference of the original signal and the moving average, then normalizing by the standard deviation of the moving average. The stationary signal is used to calculate drift and diffusion coefficients. Those coefficients are used to generate a new stationary signal. The new stationary signal is un-normalized and added to the moving average (or new equivalent signal) to create a new up-sampled velocity time-series.

Green, Nathan↗

DOLCE: A Model-Based Probabilistic Diffusion Framework for Limited-Angle CT Reconstruction

Limited-Angle Computed Tomography (LACT) is a non-destructive evaluation technique used in a variety of applications ranging from security to medicine. The limited angle coverage in LACT is often a dominant source of severe artifacts in the reconstructed images, making it a challenging inverse problem. We present DOLCE, a new deep model-based framework for LACT that uses a conditional diffusion model as an image prior. Diffusion models are a recent class of deep generative models that are relatively easy to train due to their implementation as image denoisers. DOLCE can form high-quality images from severely under-sampled data by integrating data-consistency updates with the sampling updates of a diffusion model, which is conditioned on the transformed limited-angle data. We show through extensive experimentation on several challenging real LACT datasets that, the same pre-trained DOLCE model achieves the SOTA performance on drastically different types of images. Additionally, we show that, unlike standard LACT reconstruction methods, DOLCE naturally enables the quantification of the reconstruction uncertainty by generating multiple samples consistent with the measured data.

Kim, Hyojin↗

A potential emerging issue concerning repair welding of out-of-core PWR components involving tritium exposure and 3 He retention

This paper identifies a potential but previously unrecognized risk of helium-induced embrittlement and cracking during repair welding of out-of-core pressurized water reactor (PWR) components exposed to tritium-contaminated coolant. While previous weldability concerns centered on 4 He accumulation in neutron-irradiated alloys located within the in-core or near-core regions, new measurements show that 3 He generated by tritium decay can accumulate in out-of-core components. Because hydrogen isotopes readily diffuse to grain boundaries and become trapped there, significant 3 He generation at grain boundaries may lead to cracking during weld repairs. Initial data from far-below-core PWR flux thimble tubes confirm the presence of 3 He levels above known cracking thresholds for repair welds. In conclusion, these findings indicate that out-of-core regions should be considered when defining safe weld repair windows in reactors operating for 60–100 years.

Extended PWR lifetimes↗

Lower Temperature Annealing of Vapor Diffused Nb₃Sn for Accelerator Cavities

Nb₃Sn is a next-generation superconducting material for the accelerator cavities with higher critical temperature and superheating field, both twice compared to Nb. It promises superior performance and higher operating temperature than Nb, resulting in significant cost reduction. So far, the Sn vapor diffusion method is the most preferred and successful technique to coat niobium cavities with Nb₃Sn. Although several post-coating techniques (chemical, electrochemical, mechanical) have been explored to improve the surface quality of the coated surface, an effective process has yet to be found. Since there are only a few studies on the post-coating heat treatment at lower temperatures, we annealed Nb₃Sn-coated samples at 800 C - 1000 C to study the effect of heat treatments on surface properties, primarily aimed at removing surface Sn residues. This paper discusses the systematic surface analysis of coated samples after annealing at temperatures between 850 C and 950 C.

43 PARTICLE ACCELERATORS↗