Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Differentiable physics”

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 829 records · Page 46
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Celestial leaf amplitudes

Celestial amplitudes may be decomposed as weighted integrals of AdS 3 -Witten diagrams associated to each leaf of a hyperbolic foliation of spacetime. We show, for the Kleinian three-point MHV amplitude, that each leaf subamplitude is smooth except for the expected light-cone singularities. Moreover, we find that the full translationally-invariant celestial amplitude is simply the residue of the pole in the leaf amplitude at the point where the total conformal weights of the gluons equals three. This full celestial amplitude vanishes up to light-cone contact terms, as required by spacetime translation invariance, and reduces to the expression previously derived by Mellin transformation of the Parke-Taylor formula.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2026

Remote sensing of planetary boundary layer from ground and space: Structure, thermodynamics, and boundary layer clouds

The planetary boundary layer (PBL) is a critical interface between Earth's surface and atmosphere, influencing atmospheric convection, weather patterns, and air quality. Recognized by the 2017 National Academies of Sciences, Engineering, and Medicine Earth Science Decadal Survey as an Incubation Targeted Observable, high-quality and effective observation of the PBL has become a priority. This paper reviews surface-based and satellite remote sensing techniques for characterizing PBL features, including PBL height (PBLH), boundary layer thermodynamics, turbulence, and boundary layer clouds (BLCs). These elements are treated as interconnected aspects of the PBL system, while recognizing that different instruments retrieve different physical manifestations of the PBL. The review summarizes recent advances and limitations in Micro-Pulse Lidar (MPL), Doppler lidar, Raman lidar, Differential Absorption Lidar (DIAL), ceilometers, wind profilers, GNSS Radio Occultation, radar, and hyperspectral sounders. These observations help trace dynamic processes within the PBL and link PBL structure to broader weather and climate processes. The review also highlights persistent observational gaps over oceans, remote land, and polar regions, where continuous surface-based profiling is sparse. By comparing instrument capabilities, practical limitations, and interpretation issues, this review emphasizes the need for integrated remote sensing approaches and careful definition of the retrieved PBL quantities.

Boundary layer clouds↗
DOE OSTI2025

Increasing Branching and Functionalization Decreases Crystallinity and Crystallization Rate in Linear EVOH

Here, we report the crystallization behavior of a linear poly(ethylene-co-vinyl alcohol) (LEVOH) under isothermal crystallization as a function of both OH incorporation and undercooling, and compare the results to conventional branched EVOH. This LEVOH is synthesized by post-polymerization functionalization of polycyclooctene, and it exhibits a half-crystallization time and primary crystallization rate nearly an order of magnitude faster than branched EVOH at ~ 6 mol% OH incorporation. While the LEVOH obtains approximately twice the extent of crystallization compared to the branched equivalent, as OH incorporation increases, the crystallization kinetics and crystallinity of LEVOH decrease. The crystal structure of LEVOH is orthorhombic at low functionalization (≤ 11 mol% OH), a mixture of orthorhombic and hexagonal at moderate functionalization (17 - 21 mol%), and hexagonal at high functionalization (23 %). The LEVOH crystallizes into plates, with crystallite widths ~ 20 - 60 times the crystallite thickness. Ultimately, we show LEVOH crystallizes faster and to a greater extent than commercial branched EVOH, a potential advantage of post-polymerization functionalization as part of polymer-to-polymer upcycling.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Assessing physics of ion temperature gradient turbulence via hierarchical reduced-model representations

In this work, the saturation physics of ion temperature gradient (ITG) turbulence is probed by studying how amplitudes and scalings with key parameters vary across a hierarchy of reduced models. The models derive from nonlinear fluid equations for toroidal ITG turbulence under approximations to the mode coupling interactions in wavenumber space and the representation of turbulent decorrelation. Mode coupling approximations include local-in-wavenumber treatments like the spectral density of flux in quasilinear theory, a truncation to three nonlinearly interacting waves, and the interactions in a cascade to high radial wavenumber mediated by a single zonal flow. Turbulent decorrelation treatments are based on the triplet correlation time with and without eddy damping. Model fidelity is assessed by the scalings and magnitudes of the squared amplitudes of unstable mode, stable mode, and zonal flow with respect to the flow-damping rate and temperature gradient. It is shown that all models reproduce fundamental scalings, provided they incorporate the coupling of unstable mode, stable mode, and zonal flow. Accurate amplitude prediction requires eddy damping in the triplet correlation time and proper representation of the zonal-flow drive by interactions associated with the radial wavenumber cascade.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2023

Mutually magnetically insulated two-species Brillouin flow

In this work, we use the Brillouin flow analytic framework to examine the physics of Magnetically Insulated Transmission Lines (MITL). Here, we derive a model applicable to any particle species, including both positive and negative ions, in planar and cylindrical configurations. We then show how to self-consistently solve for two-species simultaneously, using magnetically insulated electrons and positive ions as an example. We require both layers to be spatially separated and magnetically insulated (mutually magnetically insulated); for a 7.5 cm gap with a 2 MV bias voltage, this condition requires magnetic fields in excess of 2.73 T. We see a close match between mutually insulated MITL performance and “superinsulated” (high degree of magnetic insulation) electron-only theory, as may be expected for these high magnetic fields. However, the presence of ions leads to several novel effects: (1) Opposite to electron-only theory, total electron currents increase rather than decrease as the degree of magnetic insulation becomes stronger. The common assumption of neglecting electrons for superinsulated MITL operation must be revisited when ions are present—we calculate up to 20× current enhancement. (2) The electron flow layer thickness increases up to double, due to ion space-charge enhancement. (3) The contributions from both ions and electrons to the MITL flow impedance are calculated. The flow impedance drops by over 50% when ions fill the gap, which can cause significant reflections at the load if not anticipated and degrade performance. Additional effects and results from the inclusion of the ion layer are discussed.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

Symplectic machine learning model for fast simulation of space-charge effects

Symplectic simulation of space-charge effects is crucial for the design and operation of high-intensity particle accelerators. Traditional methods for simulating these effects are often computationally expensive, resulting in significant overhead. In this work, we introduce a generative model based on a U-Net architecture within a generative adversarial network framework to efficiently simulate space-charge effects. The model is trained to predict the transverse multiparticle space-charge Hamiltonian, which can be physically computed using a gridless spectral method. The one-step symplectic transverse transfer map for the particles is then obtained by differentiating the predicted Hamiltonian. Benchmarking results demonstrate that this generative model achieves an order of magnitude higher computational efficiency compared to the spectral method, providing a highly efficient alternative for simulating space-charge effects with a large number of particles. By maintaining symplecticity, the model effectively preserves the phase-space structure and mitigates nonphysical errors in long-term simulations. This model has been integrated into jutrack, a novel autodifferentiable accelerator modeling code developed in the julia programming language.

Beam code development & simulation techniques↗
DOE OSTI2024

Photoproduction of η c , b near threshold

We analyze the photoproduction of η c , b off a proton in the threshold region, in terms of C-odd gluonic correlations in the off-forward proton matrix element. Near threshold, the skewness is large, leading to a production amplitude that is dominated by four C-odd twist-3 gluon GPDs. We use the QCD instanton vacuum to estimate these C-odd contributions in the proton. The results are used to estimate the differential cross sections for coherent photoproduction of η c , b in the threshold region, at current electron facilities. Published by the American Physical Society 2024

Astronomy & Astrophysics↗
DOE OSTI2023

A bacterial sensor taxonomy across earth ecosystems for machine learning applications

Microbial communities have evolved to colonize all ecosystems of the planet, from the deep sea to the human gut. Microbes survive by sensing, responding, and adapting to immediate environmental cues. This process is driven by signal transduction proteins such as histidine kinases, which use their sensing domains to bind or otherwise detect environmental cues and “transduce” signals to adjust internal processes. We hypothesized that an ecosystem’s unique stimuli leave a sensor “fingerprint,” able to identify and shed insight on ecosystem conditions. To test this, we collected 20,712 publicly available metagenomes from Host-associated, Environmental, and Engineered ecosystems across the globe. We extracted and clustered the collection’s nearly 18M unique sensory domains into 113,712 similar groupings with MMseqs2. We built gradient-boosted decision tree machine learning models and found we could classify the ecosystem type (accuracy: 87%) and predict the levels of different physical parameters (R2 score: 83%) using the sensor cluster abundance as features. Feature importance enables identification of the most predictive sensors to differentiate between ecosystems which can lead to mechanistic interpretations if the sensor domains are well annotated. To demonstrate this, a machine learning model was trained to predict patient’s disease state and used to identify domains related to oxygen sensing present in a healthy gut but missing in patients with abnormal conditions. Moreover, since 98.7% of identified sensor domains are uncharacterized, importance ranking can be used to prioritize sensors to determine what ecosystem function they may be sensing. Furthermore, these new predictive sensors can function as targets for novel sensor engineering with applications in biotechnology, ecosystem maintenance, and medicine.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Collaborative Research: Understanding Nanoparticle-Plasma Interactions in Dusty Non-Thermal Plasmas by Nanoparticle Probes and Online Aerosol Characterization (Final Technical Report)

This report describes results from the University of Minnesota’s contribution to studies performed under the award DE-SC0018202 (principal Investigator: Christopher J. Hogan Jr.. This award is a collaborative research grant with support equally provided to Case Western Reserve University and to the University of Illinois (with principal investigator R. Mohan Sankaran). To date ten peer-reviewed published papers have resulted from the studies performed with University of Minnesota personnel contributing; this report describes the results of these studies as well as several unpublished studies, grouped into seven separate study types. The overarching goal of the carried out studies was to understand nanoparticle behavior and dynamics in non-thermal plasma synthesis reactors, specifically adapting aerosol measurement technology to provide online measurements of particle size, charge, and structure in plasma reactor systems.

42 ENGINEERING↗
NASA NTRS1985

A self-consistent linear-mode model of stellar convection

A normal-mode expansion of the linearized fluid equations in terms of small subset of spherical harmonics can provide a foundation for a physically motivated, self-consistent description of a solar-type convection zone. In the absence of dissipation, a second-order differential equation governs the radial dependence of the modes, so that interpretation of the effects on convection quantities of the normal-form 'potential well' is straightforward. The philosophy is quite different from the more recent work of Narasimha and Antia (1982): all envelopes presented here differ substantially from MLT envelopes, and therefore, from theirs, which are constructed to be consistent with MLT. The amplitude of all modes is set by a Kelvin-Helmholtz-('shear'-) instability argument unrelated to solar observations, with the result that the convection description may be considered to arise from 'first-hueristic-principles'. The thermodynamics modelled vaguely resemble the sun's, and more vigorously convective envelopes show some phenomena qualitatively like solar observations (e.g., atmospheric velocity spectra).

Macauslan, J.↗
NASA NTRS1986

The rotational velocities of low-mass stars

The rotational velocities of stars provide important clues to how stars form and evolve. Yet until recently, studies of stellar rotation were limited to stars more massive than the sun. This is beginning to change, and an observational outline of the rotational velocity evolution of stars less massive than the sun can now be provided. Low-mass stars rotate slowly during the early stages of premain-sequence evolution, and spin up as they contract to the main sequence. This spin-up culminates in a brief period of very rapid rotation at an age of order 50 million years. Physical interpretation of this increase in rotation and the subsequent main-sequence spin-down are complicated by the possibility of differential internal rotation. The observed rapidity of spin-down among G dwarfs suggests that initially only the outer convective envelopes of these stars are slowed. The data suggest an intrinsic spread in angular momentum among young stars of the same mass and age, a spread which is apparently minimized by the angular-momentum loss mechanism in old low-mass stars.

Stauffer, John B.↗
NASA NTRS2004

Mars Express Interplanetary Navigation from Launch to Mars Orbit Insertion: The JPL Experience

The National Aeronautics and Space Administration (NASA) Jet Propulsion Laboratory (JPL) played a significant role in supporting the safe arrival of the European Space Agency (ESA) Mars Express (MEX) orbiter to Mars on 25 December 2003. MEX mission is an international collaboration between member nations of the ESA and NASA, where NASA is supporting partner. JPL's involvement included providing commanding and tracking service with JPL's Deep Space Network (DSN), in addition to navigation assurance. The collaborative navigation effort between European Space Operations Centre (ESOC) and JPL is the first since ESA's last deep space mission, Giotto, and began many years before the MEX launch. This paper discusses the navigational experience during the cruise and final approach phase of the mission from JPL's perspective. Topics include technical challenges such as orbit determination using non-DSN tracking data and media calibrations, and modeling of spacecraft physical properties for accurate representation of non-gravitational dynamics. Also mentioned in this paper is preparation and usage of DSN Delta Differential Oneway Range ((Delta)DOR) measurements, a key element to the accuracy of the orbit determination.

Deep Space Network (DSN)↗
NASA NTRS2012

Radiation-Hard SpaceWire/Gigabit Ethernet-Compatible Transponder

A radiation-hard transponder was developed utilizing submicron/nanotechnology from IBM. The device consumes low power and has a low fabrication cost. This device utilizes a Plug-and-Play concept, and can be integrated into intra-satellite networks, supporting SpaceWire and Gigabit Ethernet I/O. A space-qualified, 100-pin package also was developed, allowing space-qualified (class K) transponders to be delivered within a six-month time frame. The novel, optical, radiation-tolerant transponder was implemented as a standalone board, containing the transponder ASIC (application specific integrated circuit) and optical module, with an FPGA (field-programmable gate array) friendly parallel interface. It features improved radiation tolerance; high-data-rate, low-power consumption; and advanced functionality. The transponder utilizes a patented current mode logic library of radiation-hardened-by-architecture cells. The transponder was developed, fabricated, and radhard tested up to 1 MRad. It was fabricated using 90-nm CMOS (complementary metal oxide semiconductor) 9 SF process from IBM, and incorporates full BIT circuitry, allowing a loop back test. The low-speed parallel LVCMOS (lowvoltage complementary metal oxide semiconductor) bus is compatible with Actel FPGA. The output LVDS (low-voltage differential signaling) interface operates up to 1.5 Gb/s. Built-in CDR (clock-data recovery) circuitry provides robust synchronization and incorporates two alarm signals such as synch loss and signal loss. The ultra-linear peak detector scheme allows on-line control of the amplitude of the input signal. Power consumption is less than 300 mW. The developed transponder with a 1.25 Gb/s serial data rate incorporates a 10-to-1 serializer with an internal clock multiplication unit and a 10-1 deserializer with internal clock and data recovery block, which can operate with 8B10B encoded signals. Three loop-back test modes are provided to facilitate the built-in-test functionality. The design is based on a proprietary library of differential current switching logic cells implemented in the standard 90-nm CMOS 9SF technology from IBM. The proprietary low-power LVDS physical interface is fully compatible with the SpaceWire standard, and can be directly connected to the SFP MSA (small form factor pluggable Multiple Source Agreement) optical transponder. The low-speed parallel interfaces are fully compatible with the standard 1.8 V CMOS input/output devices. The utilized proprietary annular CMOS layout structures provide TID tolerance above 1.2 MRad. The complete chip consumes less than 150 mW of power from a single 1.8-V positive supply source.

Katzman, Vladimir↗
DOE OSTI2025

DS-TIDE: Harnessing Dynamical Systems for Efficient Time-Independent Differential Equation Solving

Time-Independent Differential Equations (TIDEs) are central to modeling equilibrium behavior across a wide range of scientific and engineering domains, from electrostatics to porous media flow. Conventional numerical solvers offer reliable solutions but incur significant computational costs due to fine-grained discretization and iterative procedures. Machine learning-based approaches address this by replacing iterative solving processes with one-time inference; however, their sophisticated models require extensive training resources that often exceed those of traditional solvers. Consequently, designing a TIDE solver that achieves high accuracy, broad applicability, and exceptional computational efficiency remains a fundamental challenge. In this paper, we propose DS-TIDE, a novel hardware solver that is inspired by, and subsequently leverages, the intrinsic connection between Dynamical Systems (DS) and Differential Equations (DEs) to efficiently and accurately solve TIDEs. DS-TIDE employs a CMOS-compatible DS-based processor, whose physical states evolve under carefully designed DE-driven dynamics and naturally converge to equilibrium -- the solution of the target TIDE -- within ~1µs on a ~1-watt DS-TIDE processor. To enhance expressivity, DS-TIDE incorporates Heterogeneous Dynamics with Temporal Layering (HDTL), which solves TIDEs through a three-stage DS evolution -- conditioning, solving, and decoding -- each governed by specialized dynamics. The entire evolution process is analogous to an infinitely deep neural network temporally unrolled, offering the system the capability of representing complex equations. Furthermore, DS-TIDE is equipped with an on-device DS-DE Auto-Alignment mechanism that dynamically adapts intrinsic hardware dynamics within milliseconds, effectively aligning the system’s dynamics to diverse target DEs. Experimental results across TIDEs from a wide range of scientific and engineering domains demonstrate that DS-TIDE achieves ~10^3× speedup, ~10^5× energy savings, and competitive or superior accuracy compared to state-of-the-art numerical and ML-based solvers.

Liu, Chuan↗
DOE OSTI2023

Local lattice distortions and electronic orders in strongly correlated systems by resonant total x-ray scattering: A case study of A Pt 2 X 2 intermetallics ( A = U , Ce, or La and X = Si or Ge)

Revealing local lattice distortions in physical systems is nontrivial, largely because such structural features often are not necessarily amenable to traditional crystallographic investigations. This poses limits to our understanding of the underlying physic, particularly in the case of strongly correlated systems where structural, electronic and magnetic degrees of freedom are intertwined. Using resonant total x-ray scattering coupled to differential pair distribution function analysis, we reveal the presence of pronounced local lattice distortions in the square net Pt planes in the prototypical strongly correlated APt 2 X 2 intermetallics (A = U, Ce or La and X = Si or Ge). The distortions are present before charge density wave, magnetic, Kondo lattice coherence and/or superconducting orders emerge in these materials and, as density functional calculations suggest, are likely to affect them. Finally, our study sheds light on the poorly known interactions between electronic orders and structural disorder in strongly correlated systems and also demonstrates an advanced experimental approach to determine it that is relevant to any physical system showing deviations from perfect crystallinity.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2022

From light neutrino decay phenomenology to muon neutrino cross-section measurement at MINERvA experiment

This thesis tackles two essential topics of neutrino physics: neutrino decay and neutrino cross-section measurement. First, the invisible and visible neutrino decay is analyzed through a phenomenological approach, considering future long-baseline neutrino experiments such as DUNE and a hypothetical neutrino beam toward the ANDES laboratory. The study takes into account the ν μ and ν e disappearance and appearance, respectively, for both FHC and RHC flux modes. The results showed a negligible matter effect for DUNE but significantly more notable at ANDES. At 90% C.L., the sensitivity to the decay parameter α3 can be as small as 2 × 10 -6 eV 2 for a chosen coupling. The impact of neutrino decay in the determination of θ 23 and δ CP were also shown. Second, the double-differential cross-section measurement for ν μ -carbon interactions with three-momentum transfer |$\vec{q}$| < 1.2 GeV obtained with medium energy exposures in the NuMI beam at MINERvA experiment are reported. The measurement is presented as a function of |$\vec{q}$| and E avail and reviews different interaction models and nuclear effects along quasi-elastic to resonance processes to define a new model for a better agreement. The double differential cross sections are compared to the MnvTunes, GENIE, and NuWro predictions.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

Strictly Enforcing Invertibility and Conservation in CNN-Based Super Resolution for Scientific Datasets

Abstract Recently, deep convolutional neural networks (CNNs) have revolutionized image “super resolution” (SR), dramatically outperforming past methods for enhancing image resolution. They could be a boon for the many scientific fields that involve imaging or any regularly gridded datasets: satellite remote sensing, radar meteorology, medical imaging, numerical modeling, and so on. Unfortunately, while SR-CNNs produce visually compelling results, they do not necessarily conserve physical quantities between their low-resolution inputs and high-resolution outputs when applied to scientific datasets. Here, a method for “downsampling enforcement” in SR-CNNs is proposed. A differentiable operator is derived that, when applied as the final transfer function of a CNN, ensures the high-resolution outputs exactly reproduce the low-resolution inputs under 2D-average downsampling while improving performance of the SR schemes. The method is demonstrated across seven modern CNN-based SR schemes on several benchmark image datasets, and applications to weather radar, satellite imager, and climate model data are shown. The approach improves training time and performance while ensuring physical consistency between the super-resolved and low-resolution data. Significance Statement Recent advancements in using deep learning to increase the resolution of images have substantial potential across the many scientific fields that use images and image-like data. Most image super-resolution research has focused on the visual quality of outputs, however, and is not necessarily well suited for use with scientific data where known physics constraints may need to be enforced. Here, we introduce a method to modify existing deep neural network architectures so that they strictly conserve physical quantities in the input field when “super resolving” scientific data and find that the method can improve performance across a wide range of datasets and neural networks. Integration of known physics and adherence to established physical constraints into deep neural networks will be a critical step before their potential can be fully realized in the physical sciences.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1988

Ureilites are not igneous differentiates

Although most all meteorites are as old as the solar system (4.5 billion years), they can be subdivided into primitive and evolved groups, depending on the extent of their chemical and physical processing. Primitive meteorites, most of which are chondrites, are assemblages of dust and millimeter-sized pellets from the presolar nebula, which were not extensively heated and processed since their assembly. Thus they provide information about the conditions in the nebular cloud. Many of the evolved meteorites are achondrites, which are igneous rocks produced by melting on or within an asteroidal object known as the parent body. A major unsolved problem in solar system studies is identification of the source of heat which led to melting of the achondrites. The role of oxygen isotopes in establishing genetic relationships among meteorites is examined.

Clayton, Robert N.↗