Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Gradient information”

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 703 records · Page 39
All sourcesNASA NTRSDOE OSTI
NASA NTRS2009

Atlantic Tropical Cyclogenetic Processes During SOP-3 NAMMA in the GEOS-5 Global Data Assimilation and Forecast System

This article investigates the role of the Saharan air layer (SAL) in tropical cyclogenetic processes associated with a nondeveloping and a developing African easterly wave observed during the Special Observation Period (SOP-3) phase of the 2006 NASA African. Monsoon Multidisciplinary Analyses (NAMMA). The two waves are chosen because they both interact heavily with Saharan air. A glottal data assimilation and forecast system, the NASA Goddard Earth Observing System. version 5 (GEOS-5), is being run to produce a set of high-9 uality global analyses, inclusive of all observations used operationally but with additional satellite information. In particular, following previous works by the same authors, the duality-controlled data from the Atmospheric Infrared Sounder (AIRS) used to produce these analyses have a better coverage than the one adopted by operational centers. From these improved analyses, two sets of 31 five-day high-resolution forecasts, at horizontal resolutions of both half and quarter degrees, are produced. Results indicate that very steep moisture gradients are associated with the SAL in forecasts and analyses, even at great distances from their source over the Sahara. In addition, a thermal dipole in the vertiieat (warm above, cool below) is present in the nondeveloping case. The Moderate Resolution Imaging Spoctroradiometer (MODIS) aboard NASA's Terra and Aqua satellites shows that aerosol optical thickness, indicative of more dust as opposed to other factors, is higher in the nondeveloping case. Altogether, results suggest that the radiative effect of dust may play some role in producing a thermal structure less favorable to cyclogenesis. Results also indicate that only global horizontal resolutions on the order of 20-30 km can capture the large-scale transport and the tine thermal structure of the SAL, inclusive of the sharp moisture gradients, reproducing the effect of tropical cyclone suppression that has been hypothesized by previous authors from observational and regional modeling perspectives. Thcse effects cannot be fully represented at lower resolutions, therefore global resolution of a quarter of a degree is a minimum critical threshold necessary to investigate Atlantic tropical cyclogenesis from a global modeling perspective

Reale, Oreste↗
NASA NTRSDate not supplied

A Notional Artemis Lunar Surface Exploration Package (ArLSEP) based on the Gandalf Staff Platform

Introduction: The Artemis program is planning to deliver crew and cargo to the lunar surface, but there is no current package for supporting lunar in-struments and experiments similar to the Apollo Lunar Surface Exploration Package (ALSEP). This abstract provides a possible concept for such a package using the Gandalf Staff Platform as a common core. Gandalf Staff: The Gandalf Staff is an early prototype system developed over FY’21/FY’22 using NASA Science Technology Mission Directorate (STMD) Center Information Fund (CIF) grants to de-sign, build and test “proof-of-concept” components. These components include a 24v battery powered monopole that powers a suite of subsystems, including a Graphical User Interface (GUI) for crew, surface voice and data communications, Lunar Search and Rescue (LunaSAR) navigation and communications, LiDAR, field site external lighting, 360-degree camera, and a geothermal instrument for measuring sub-surface temperature gradient. The staff can be carried independently by an Extra-Vehicular Activity (EVA) astronaut, or can be mounted into a tripod for “hands free” support at a surface site being investigated. The staff can be attached to an external solar array and power storage system for long-duration operations. [1,2] ALSEP: An ASLEP flew on each mission Apollo 12 to Apollo 17. For Apollo 11, a simplified packaged called the Early Apollo Scientific Experiments Pack-age (EASEP) was flown. Each package included a “Central Station” that provided the power and communications connected to a variety of instruments and sensors. The power was provided by a Radioisotope Thermoelectric Generator (RTG) fueled by Plutoni-um-238 generating 70 watts of power (initially, decayed over time) [3]. The communications system provide for direct to Earth data transfer from the lunar surface. Each pack-age was stowed externally in the Lunar Module (LM) Scientific Equipment (SEQ) bay with a mass up to 163 kg (Apollo 17). The crew unloaded the ALSEP from the LM and deployed the instruments on the lunar surface. Although designed to operate for only 1 year, many sites operated for up to 8 years successfully [4]. The Active Seismic Experiment (ASE) included 3 geophones for detecting seismic waves created by mortars and thumpers deployed by the crew. Other active experiments measured the lunar atmosphere, the heat flow in the subsurface, the lunar gravity and potential gravity waves, the lunar magnetic field, the solar wind and plasma interactions in cislunar space. Passive experiments included collectors for dust and cosmic rays, and retroreflectors for precise measurements of distance using a laser from Earth. The ALSEP program continues to generate insights into lunar formation and evolution. ArLSEP Concepts: The lunar surface science package for the Artemis program will hopefully exceed the capability of the ALSEP. There are multiple issues for discussion leading to the design of a new ArLSEP, needing requirements definition from the science community, NASA mission architecture, and NASA budget planners. 1. Delivery Mechanism Two possible projects currently provide capability to deliver scientific cargo to the lunar surface: 1) the Commercial Lunar Payload Services (CLPS) [5] and the Human Landing System (HLS) [6, 7]. Each project is controlled by a different organization within NASA and budgeted with different criteria although both support lunar exploration. The HLS system delivers crew (and potentially cargo) to human landing sites. If an ArLSEP is “predeployed” to such a site, the design must include power (either from the vehicle or independently) to keep the electronics functioning until deployed by the crew. If an ArLSEP is delivered on a vehicle after the crew is present on the lunar surface, safety protocols require adequate distance from the humans for impact from descent propelled sur-face regolith ejecta. This distance can not exceed the capability of the crew to walk (if no rover) to the vehicle for ArLSEP deployment. 2. Overall Guidelines The general design of ArLSEP will likely follow the ALSEP with a common system for communications and power; however, significant architecture differences between Apollo and Artemis exist. Power: The RTG will not be available for early Artemis missions nor likely follow-on Lunar Exploration Transportation Services (LETS) missions [8]. Thus, ArLSEP power must be supplied by solar arrays with sufficient battery capability to “keep alive” necessary electronics during any lunar surface eclipse period. Communication: The Artemis program is developing a series of communications satellites for lunar orbit to provide surface transmission of data and voice to Earth. Called “LunaNET”, this network is component useful for ArLSEP since south polar locations may not always have direct “line-of-sight” to Earth [9]. 3. Concept of Operations (ConOps) The general ConOps for ArLSEP is to deliver the package to lunar surface before the crew arrives, and then have the crew deploy the package after some period of time. This requires coordinated design (for power systems) and launch window (for schedule) on both the cargo and crew missions. Once the ArLSEP is deployed, it will operate autonomously for a number of years. It should be designed to be EVA compatible for crew maintenance and upgrade. 4. Notional Design (for discussion purpose only) The landing site near the South Pole is expected to have no eclipse cycle exceeding 5 days, so the “keep alive” power is 144 hours (6 days to include margin). A 12v ArLSEP will use rechargeable LiFePO4 cells, which are common in the Electric Vehicle (EV) industry. With a current of 5 amps and a 125 watt system, the mass is about 90kg. The comm. system and structure adds another 10kg, thus the “Central Station” is approximately 100kg. The solar power is collected on four arrays (each 2m above the surface), and the entire ArLSEP is designed to stow in a 2m x 1m x 1m volume. The experiment and instrument design will vary for each installation and add mass to the total (although they are expected to fit within the 2m3 volume). Seismic wave generation will likely not be provided with mortars, thus an electric “thumper” will be required. Active instruments such as imaging systems and sensing instruments will benefit from the additional power and communication capability provided by ArLSEP. Passive systems such as retroreflectors, witness plates, and cosmic dust collectors can be added to either the landing vehicle and/or the ArLSEP. With repeated HLS missions to the same human site, the ArLSEP can be expanded and easily maintained for long duration science collection on the lunar surface.

ALSEP↗
DOE OSTI2025

Position-Enhanced Gradient Attack (PEGA) on Medical Language Models

Federated Learning (FL) enables collaborative training of language models on sensitive clinical notes without sharing the data. However, this paradigm is vulnerable to gradient inversion attacks that can reconstruct private data from shared gradients. We find that state-of-the-art attacks are less effective in the medical domain, failing to overcome the unique challenges posed by its specialized vocabulary and unstructured format. To address this, we introduce the Position-Enhanced Gradient Attack (PEGA), a novel attack that makes gradients position-aware by optimizing token and position embeddings simultaneously. PEGA employs two key innovations: a periodic sorting of positional embeddings to resolve token order ambiguity and a late-stage embedding replacement strategy to correct hard-to-recover critical tokens. To evaluate the leakage of sensitive data more directly, we also propose the Unified PHI-Recall (UPHI), a new metric measuring the recovery of Protected Health Information. Experiments on the MIMIC-III dataset show that PEGA significantly outperforms leading attacks like TAG and LAMP, particularly in its ability to reconstruct identifiable patient information, exposing a more severe and nuanced privacy risk in federated medical NLP.

Xu, Nuo [University of Minnesota]↗
DOE OSTI2020

Functional, Hierarchical Colloidal Liquid Crystal Gels and Liquid Crystal Elastomers Nanocomposites. Final Report

Biological systems rely on a hierarchy of dynamic, complex and reconfigurable compartments to carry out functions essential for life. This hierarchical organization serves as an amplifier that allows highly localized, molecular events to propagate into the mesoscale, resulting in dynamic functional properties of biological systems that have not yet been fully realized in synthetic material designs. We pursued a program of research in which we recreated such principles using synthetic liquid crystals (LC) as a versatile platform. We elucidated new hierarchical design strategies that heavily leverage surface anchoring, coexisting phases and interfacial tension gradient to realize equilibrium and non-equilibrium, dynamic mesoscale phenomena in the context of micrometer-sized LC droplets containing multiple compartments. Specifically, we unmasked how equilibrium and dynamical phenomena emerged from hierarchical organizations in reconfigurable, thermotropic liquid crystalline assemblies. The long-term impact of the project is to advance new hierarchical designs of complex soft matter systems in which biomimetic principles, including the propagation of events or information over multiple temporal and spatial scales, are enacted in a facile manner.

36 MATERIALS SCIENCE↗
NASA NTRS1974

Recent advancement of turbulent flow measurement techniques

Advancements of the fluctuating density gradient cross beam laser Schlieren technique, the fluctuating line-reversal temperature measurement and the development of the two-dimensional drag-sensing probe to a three-dimensional drag-sensing probe are discussed. The three-dimensionality of the instantaneous momentum vector can shed some light on the nature of turbulence especially with swirling flow. All three measured fluctuating quantities (density, temperature, and momentum) can provide valuable information for theoreticians.

Battle, T.↗
NASA NTRS1994

Electric field effects on a near-critical fluid in microgravity

The effects of an electric field on a sample of SF6 fluid in the vicinity of the liquid-vapor critical point is studied. The isothermal increase of the density of a near-critical sample as a function of the applied electric field was measured. In agreement with theory, this electrostriction effect diverges near the critical point as the isothermal compressibility diverges. Also as expected, turning on the electric field in the presence of density gradients can induce flow within the fluid, in a way analogous to turning on gravity. These effects were observed in a microgravity environment by using the Critical Point Facility which flew onboard the Space Shuttle Columbia in July 1994 as part of the Second International Microgravity Laboratory Mission. Both visual and interferometric images of two separate sample cells were obtained by means of video downlink. The interferometric images provided quantitative information about the density distribution throughout the sample. The electric field was generated by applying 500 Volts to a fine wire passing through the critical fluid.

Zimmerli, G.↗
DOE OSTI2021

Explicit physics-informed neural networks for nonlinear closure: The case of transport in tissues

In upscaling methods, closures for nonlinear problems present a well-known challenge. While a number of theoretical methods have been proposed for handling such closures, nonlinearities still remain a significant obstacle for many problems. In this work, we use a combination of formal upscaling and data-driven machine learning for explicitly closing a nonlinear transport and reaction process in multiscale tissues. The classical effectiveness factor model is used to formulate the macroscale reaction kinetics. We train a multilayer perceptron network using training data generated by direct numerical simulations over microscale examples. Once trained, the network is used in an algorithm for numerically solving the upscaled (coarse-grained) differential equation describing mass transport and reaction in two example tissues. The network is described as being explicit in the sense that the network is trained using macroscale concentrations and gradients of concentration as components of the feature space rather than incorporating them as part of a constraint in the optimization process. Network training and solutions to the macroscale transport equations were computed for two different tissues. The two tissue types (brain and liver) exhibit markedly different geometrical complexity and spatial scale (cell size and sample size). The upscaled solutions for the average concentration are compared with numerical solutions derived from the microscale concentration fields by a posteriori averaging. There are three outcomes of this work of particular note. 1) Our overall approach results in an upscaled nonlinear PDE. The PDE is closed using a neural network, and our approach results in the definition of the classical effectiveness factor for effecting closure. 2) We identify particular source terms for the closure problem that are important for representing the structure of the closure. These source terms involve macroscale concentrations and their gradients. We adopt these source terms to use as explicit features in the learning algorithm. We find the trained networks that include the macroscale source terms generate models that are able to predict the correction factor with increased fidelity over those that do not. 3) We find that the trained network exhibits good generalizability, and it is able to predict the effectiveness factor with high fidelity for realistically-structured tissues despite the significantly different scale and geometrical complexity of the two example tissue types. This latter result emphasizes our purposeful connection between conventional averaging methods with the use of machine learning for closure; this contrasts with some machine learning methods for upscaling where the exact form of the macroscale equation remains unknown.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Low predictability of energy balance traits and leaf temperature metrics in desert, montane and alpine plant communities

Abstract Leaf energy balance may influence plant performance and community composition. While biophysical theory can link leaf energy balance to many traits and environment variables, predicting leaf temperature and key driver traits with incomplete parameterizations remains challenging. Predicting thermal offsets ( δ , T leaf − T air difference) or thermal coupling strengths ( β , T leaf vs. T air slope) is challenging. We ask: (a) whether environmental gradients predict variation in energy balance traits (absorptance, leaf angle, stomatal distribution, maximum stomatal conductance, leaf area, leaf height); (b) whether commonly measured leaf functional traits (dry matter content, mass per area, nitrogen fraction, δ 13 C, height above ground) predict energy balance traits; and (c) how traits and environmental variables predict δ and β among species. We address these questions with diurnal measurements of 41 species co‐occurring along a 1,100 m elevation gradient spanning desert to alpine biomes. We show that (a) energy balance traits are only weakly associated with environmental gradients and (b) are not well predicted by common functional traits. We also show that (c) δ and β can be partially approximated using interactions among site environment and traits, with a much larger role for environment than traits. The heterogeneity in leaf temperature metrics and energy balance traits challenges larger‐scale predictive models of plant performance under environmental change. A free Plain Language Summary can be found within the Supporting Information of this article.

Blonder, Benjamin↗
NASA NTRS1991

Cascading a systolic array and a feedforward neural network for navigation and obstacle avoidance using potential fields

A technique is developed for vehicle navigation and control in the presence of obstacles. A potential function was devised that peaks at the surface of obstacles and has its minimum at the proper vehicle destination. This function is computed using a systolic array and is guaranteed not to have local minima. A feedfoward neural network is then used to control the steering of the vehicle using local potential field information. In this case, the vehicle is a trailer truck backing up. Previous work has demonstrated the capability of a neural network to control steering of such a trailer truck backing to a loading platform, but without obstacles. Now, the neural network was able to learn to navigate a trailer truck around obstacles while backing toward its destination. The network is trained in an obstacle free space to follow the negative gradient of the field, after which the network is able to control and navigate the truck to its target destination in a space of obstacles which may be stationary or movable.

Plumer, Edward S.↗
NASA NTRS2016

Towards a Viscous Wall Model for Immersed Boundary Methods

Immersed boundary methods are frequently employed for simulating flows at low Reynolds numbers or for applications where viscous boundary layer effects can be neglected. The primary shortcoming of Cartesian mesh immersed boundary methods is the inability of efficiently resolving thin turbulent boundary layers in high-Reynolds number flow application. The inefficiency of resolving the thin boundary is associated with the use of constant aspect ratio Cartesian grid cells. Conventional CFD approaches can efficiently resolve the large wall normal gradients by utilizing large aspect ratio cells near the wall. This paper presents different approaches for immersed boundary methods to account for the viscous boundary layer interaction with the flow-field away from the walls. Different wall modeling approaches proposed in previous research studies are addressed and compared to a new integral boundary layer based approach. In contrast to common wall-modeling approaches that usually only utilize local flow information, the integral boundary layer based approach keeps the streamwise history of the boundary layer. This allows the method to remain effective at much larger y+ values than local wall modeling approaches. After a theoretical discussion of the different approaches, the method is applied to increasingly more challenging flow fields including fully attached, separated, and shock-induced separated (laminar and turbulent) flows.

Immersed↗
DOE OSTI2023

Flow friction in a wide rectangular duct with a beveled inlet: Informal Notes of Fundamental Fluid Physics Studies

Fundamental streamwise pressure distributions were measured at Idaho State University to address a primary objective of obtaining insight into the apparent evolution of air flow regimes at moderate Reynolds numbers in thin, wide rectangular ducts with converging inlets. Ducts of 3.030 and 4.077 mm spacings s (called three and four mm), having non-dimensional lengths of (L/s) ≈ 500.8 and 372.2 and widths (W/s) ≈ 114.6 and 83.2, respectively, were compared. Flow rates ranged from Reynolds numbers based on hydraulic diameter (Re Dh ) of about 980 to 10,000 and corresponding non-dimensional downstream pressure gradient parameters (-K p ) covered the range 0.038 to 0.0061, respectively.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2022

Feedback Power Cost Optimization in Power Distribution Networks with Prosumers

Here, we propose a feedback control approach for solving optimal power flow (OPF) problems in power distribution networks (DNs) based on a projected gradient scheme, where the cost is given by the sum of functions of local power injections. This approach does not require detailed knowledge of the grid model and enables real-time tracking of the optimal generation configuration. Moreover, the communication burden is minimal: only a broadcast from a central entity, called the Network Supervisor (NS), to the agents in the network is needed. Hence, agents do not have to share local information that can be sensitive, e.g., their power injections or the value of the local costs. Our approach is then used to solve a particular OPF problem targeting DNs with high penetration of distributed energy resources (DERs) in which the energy cost is given by the remuneration that the DERs owners receive for the service they provide.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2025

Reproductive and leaf litterfall fluxes in forest ecosystem sites globally (1950-2022)

Forest allocation of net primary productivity (NPP) to reproduction is poorly quantified globally, despite its critical role in forest regeneration and a well-supported trade-off with allocation to growth. Although field measurements of total NPP are rare, our work finds that a proxy for reproductive carbon allocation constructed from leaf (L) and reproductive (R) litterfall fluxes, R/(R+L), is strongly correlated with R/NPP, facilitating analysis across a wide range of sites where biometric estimates of NPP are not available (R² = 0.85; Hanbury-Brown et al., 2022, Ward et al., in prep). To investigate relationships between ecosystem-scale reproductive allocation (RA) and climate, soil fertility, and stand age gradients, we conducted a literature search and synthesized 824 observations of annual average leaf and reproductive litterfall fluxes across forest sites globally. The zip file includes 1) a folder Data/ containing the litterfall data ("GlobalForestRA_data.csv") and metadata ("GlobalForestRA_metadata.doc") files. The data file includes geographic coordinates, long-term mean annual temperature and precipitation (1970-2000, extracted from WorldClim2.1), leaf and reproductive litterfall fluxes, sampling interval and protocols, forest characteristics (dominant leaf morphology, information pertaining to forest age and successional stage, and disturbance history) and soil properties (% sand, %silt, %clay, total phosphorus (P), nitrogen (N), cation exchange capacity (CEC) and pH) extracted from SoilGrids250 and from on-site measurements, where available. The metadata file contains information about each variable reported in the data file, including data sources, processing methods, and all references. The Data folder contains two additional files used to create Figure 1; these are described in greater detail in the README.2) R scripts GloalForestRA_analysis.r and GlobalForestRA_SI.r and a folder /Functions used to produce results, figures, and tables in the manuscript Ward et al. (in press)3) a README file describing how the data and R scripts can be used to reproduce statistical results, figures, and tables found in the manuscript. Ward et al. (in press)This repository can also be found at: https://github.com/r-ward/Global_Analysis_ForestRA.Ward, R.E., Zhang-Zheng, H. Aernethy, K., Adu-Bredu, S., Arroyo, L., Bailey, A. et al. (in press). Forest age rivals climate to explain reproductive allocation patterns in forest ecosystems globally. Ecology Letters. Hanbury-Brown, A.R., Ward, R.E. & Kueppers, L.M. (2022). Forest regeneration within Earth system models: current process representations and ways forward. New Phytol., 235, 20–40.Ward et al. (2025), Forest age rivals climate to explain reproductive allocation patterns in forest ecosystems globally, in prep.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1973

Correlation of heat transfer for the zero pressure gradient hypersonic laminar boundary layer for several gases

A theoretical study of heat transfer for zero pressure gradient hypersonic laminar boundary layers for various gases with particular application to the flows produced in an expansion tube facility was conducted. A correlation based on results obtained from solutions to the governing equations for five gases was formulated. Particular attention was directed toward the laminar boundary layer shock tube splitter plates in carbon dioxide flows generated by high speed shock waves. Computer analysis of the splitter plate boundary layer flow provided information that is useful in interpreting experimental data obtained in shock tube gas radiation studies.

Cook, W. J.↗
NASA NTRS1996

(abstract) Three Dimensional Ice-Flow Velocity Estimation Using Satellite Radar Interferometry

An understanding of the flow dynamics of an ice sheet's outlet glaciers and ice streams requires knowledge of their flow velocity and strain rates (i.e., velocity gradients). Prior to the recent advent of satellite radar interferometry, it was not possible to measure detailed ice-flow velocity over the vast featureless areas that comprise most of the ice sheets. Since the launch of ERS-1, the use of satellite radar interferometry data for making densly sampled ice-flow velocity measurements has been firmly established by several studies. We have combined data from nonparallel orbits with surface slope information to make vector ice-flow measurements for the Ryder Glacier, Greenland. Our results for the Ryder are promising and indicate that repeat-pass interferometric data can be used to make vector measurements of ice velocity.

ice sheets radar interferometry glaciers flow dyna↗
DOE OSTI2021

Magnetic Field Measurements of First Pre-series Full -Length 4.2 m Quadrupole MQXFA03 Using PCB Rotating Coils for the Hi Hi-Lumi LHC Project

The U.S. Hi-Lumi LHC Accelerator Upgrade Project (AUP) and CERN have joined efforts to develop high field quad-rupoles for the Hi-Lumi LHC upgrade. The US national laborato-ries in the AUP project will deliver 10 magnets and each cryostat has two 4.2 m high gradient quadrupoles in it. These magnets are made of Nb3Sn conductors, with large aperture (150 mm) and in-tegrated gradient of 556.9 T. This paper reports on magnetic measurements performed during the vertical test at Brookhaven National Laboratory (BNL) in 2019-2020. A warm measurement Z-Scan (+/- 15 A) with 42 Z-positions before cool-down was per-formed at BNL. The results were directly compared to field data measured at LBNL during magnet assembly. Measured harmonics and magnetic center offsets (ΔX and ΔY) have provided timely and informative diagnostics on the magnet structure’s shape at both warm and cold temperatures. A new centering fixture was de-signed and added to better center the warm bore tube which con-tains the rotating coil probe. After the quench training to 16.47 kA was achieved, a complete set of cold measurements (Z-Scan at 16.47 kA and I-Scan from 960 A to 16.47 kA and back to 960 A) was made. Periodic axial variation of allowed and nonallowed harmonics was observed which is related to the coil radial and/or mid-plane variations along the magnet axis. Overall, the average harmonics in the straight section are within the required field boundaries.

43 PARTICLE ACCELERATORS↗
DOE OSTI2026

Search for HH → bbτ⁺τ⁻ Using Run 3 Scouting Data Analyze b-tagging and tau-tagging Performance with Unified Particle Transformer

B-tagging and tau-tagging performances play an important role in the search for the rare event HH → bbτ⁺τ⁻. A transformer-based neural network, Unified Particle Transformer, is applied for both tagging tasks, and Run 3 proton–proton collision scouting data at center-of-mass energy of 13.6 TeV is used. The scouting data stream accepts events at a much higher rate compared to traditional triggers, but stores only the objects reconstructed in the trigger, no low-level detector information. Therefore, existing taggers trained for the offline event reconstruction cannot be used. Analysis of the SoftMax plots, ROC/AUC curves, confusion matrix, accuracy and losses are used to evaluate model performance. Specifically, the tagging efficiency of the signal and misidentification probability across multiple background processes are compared for varying working points. Different training samples with distinct distributions of jet flavors are utilized and related model performances are analyzed. Interpretability methods, such as Integrated Gradients, may further be applied to study the input features’ influence on the model’s decisions, providing insights into potential improvements.

Chen, Blair [Purdue U., West Lafayette; Fermilab]↗
NASA NTRS1971

Spectra-spectroheliograph observations.

This paper describes a photographic technique for recording solar spectra. The technique, which we call spectra-spectroheliography, requires the same amount of observing time as conventional spectroheliography but yields much more information. Data reduction techniques have been developed to use this information for the construction of spectroheliograms and other contour maps. Parameters investigated by this method include the vector magnetic field, line-of-sight velocity, continuum intensity, and line strength. Results indicate (1) good correlation between bright regions in spectroheliograms and small vertical magnetic fields; (2) poorer correlation between larger magnetic fields and bright regions; (3) a feature in the continuum resembling a pore but having no vertical magnetic field; and (4) a gradient in the vertical magnetic field of 12kG/arc sec.

Title, A. M.↗