Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “feature reduction”

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 775 records · Page 43
All sourcesNASA NTRSDOE OSTI
DOE OSTI2012

Tetrahedral Palladium Nanocrystals: A New Support for Platinum Monolayer Electrocatalysts with High Activity and Stability in the Oxygen Reduction Reaction

The recent availability of tetrahedral palladium (PdTH) nanocrystals with cleaned surfaces allowed us to evaluate their facet-specific electrochemical properties as a new support of platinum monolayer (PtML) catalysts. The Pd–PtML core-shell electrocatalyst was examined by combining structural analyses and Density Functional Theory (DFT) with electrochemical techniques. The surfaces of the PdTH core are composed of (111) facets wherein the Pd atoms are highly coordinated and have low surface energy. Our results revealed that in comparison with sphere Pd (PdSP)-supported PtML or pure Pt, the PdTH-supported PtML features more surface contraction and a downshift of d-band relative to the Fermi level. These geometric- and electronic-effects determine the higher activity of PtML/PdTH/C for the oxygen reduction reaction (ORR) compared to that of PtML/PdSP/C. This shape-property interdependence illuminated new approaches to basic- and applied- research on Pt-based ORR electrocatalysts of significant importance to the widespread use of fuel cells.

30 DIRECT ENERGY CONVERSION↗
NASA NTRS2002

Impact Cratering Calculations

Many Martian craters are surrounded by ejecta blankets which appear to have been fluidized forming lobate and layered deposits terminated by one or more continuous distal scarps, or ramparts. One of the first hypotheses for the formation of so-called rampart ejecta features was shock-melting of subsurface ice, entrainment of liquid water into the ejecta blanket, and subsequent fluidized flow. Our work quantifies this concept. Rampart ejecta found on all but the youngest volcanic and polar regions, and the different rampart ejecta morphologies are correlated with crater size and terrain. In addition, the minimum diameter of craters with rampart features decreases with increasing latitude indicating that ice laden crust resides closer to the surface as one goes poleward on Mars. Our second goal in was to determine what strength model(s) reproduce the faults and complex features found in large scale gravity driven craters. Collapse features found in large scale craters require that the rock strength weaken as a result of the shock processing of rock and the later cratering shear flows. In addition to the presence of molten silicate in the intensely shocked region, the presence of water, either ambient, or the result of shock melting of ice weakens rock. There are several other mechanisms for the reduction of strength in geologic materials including dynamic tensile and shear induced fracturing. Fracturing is a mechanism for large reductions in strength. We found that by incorporating damage into the models that we could in a single integrated impact calculation, starting in the atmosphere produce final crater profiles having the major features found in the field measurements (central uplifts, inner ring, terracing and faulting). This was accomplished with undamaged surface strengths (0.1 GPa) and in depth strengths (1.0 GPa).

Ahrens, Thomas J.↗
DOE OSTI2022

Reduced Stochastic Resistive Switching in Organic-Inorganic Hybrid Memristors by Vapor-Phase Infiltration

We report resistive random-access memory (RRAM) is promising for next-generation data storage and non-von Neumann computing hardware. However, tuning device switching characteristics and particularly, controlling their stochastic variation remain as critical challenges. Here, new organic-inorganic hybrid RRAM media are reported whose bipolar switching characteristics and stochasticity can be controlled by vapor-phase infiltration (VPI), an ex situ hybridization technique derived from atomic layer deposition. Hybrid RRAMs based on AlO x -infiltrated SU-8 feature facile tunability of device switching voltages, off-state current, and on-off ratio by adjusting the amount of infiltrated AlO x in the hybrid. Furthermore, a significant reduction in the stochastic, cycle-to-cycle variations of switching parameters is enabled by AlO x infiltration, driven by the infiltration-induced changes in mechanical, dielectric, and chemical properties of organic medium and their influence on the dimension and formation characteristics of conductive filaments. Finally, multi-level analog switching potentially useful for neuromorphic applications are demonstrated, along with direct, one-step device patterning exploiting the negative-tone resist feature of SU-8. With the demonstrated control over switching characteristics and stochastic variation, combined with analog switching and one-step patterning capabilities, the results not only present a novel hybrid medium for RRAM applications but also showcase the utility of VPI for developing new, high-performance hybrid RRAM devices.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2013

Development of Optimal Catalyst Designs and Operating Strategies for Lean NOx Reduction in Coupled LNT-SCR Systems

Oxides of nitrogen in the form of nitric oxide (NO) and nitrogen dioxide (NO 2 ) commonly referred to as NO x , is one of the two chemical precursors that lead to ground-level ozone, a ubiquitous air pollutant in urban areas. A major source of NO x } is generated by equipment and vehicles powered by diesel engines, which have a combustion exhaust that contains NO x in the presence of excess O 2 . Catalytic abatement measures that are effective for gasoline-fueled engines such as the precious metal containing three-way catalytic converter (TWC) cannot be used to treat O 2 -laden exhaust containing NO x . Two catalytic technologies that have emerged as effective for NO x abatement are NO x storage and reduction (NSR) and selective catalytic reduction (SCR). NSR is similar to TWC but requires much larger quantities of expensive precious metals and sophisticated periodic switching operation, while SCR requires an on-board source of ammonia which serves as the chemical reductant of the NO x . The fact that NSR produces ammonia as a byproduct while SCR requires ammonia to work has led to interest in combining the two together to avoid the need for the cumbersome ammonia generation system. In this project a comprehensive study was carried out of the fundamental aspects and application feasibility of combined NSR/SCR. The project team, which included university, industry, and national lab researchers, investigated the kinetics and mechanistic features of the underlying chemistry in the lean NOx trap (LNT) wherein NSR was carried out, with particular focus on identifying the operating conditions such as temperature and catalytic properties which lead to the production of ammonia in the LNT. The performance features of SCR on both model and commercial catalysts focused on the synergy between the LNT and SCR converters in terms of utilizing the upstream-generated ammonia and alternative reductants such as propylene, representing the hydrocarbon component of diesel exhaust. First-principle models of the LNT and SCR converters, which utilized the mechanistic-based kinetics and realistic treatments of the flow and transport processes, in combination with bench-scale reactor experiments helped to identify the best designs for combining the NSR and SCR catalysts over a range of operating conditions encountered in practice. This included catalysts having multiple zones and layers and additives with the focus on determining the minimal precious metal component needed to meet emission abatement targets over a wide range of operating conditions. The findings from this study provide diesel vehicle and catalyst companies valuable information to develop more cost effective diesel emissions catalysts which helps to expand the use of more fuel efficient diesel power. The fundamental modeling and experimental tools and findings from this project can be applied to catalyst technologies used in the energy and chemical industries. Finally, the project also led to training of several doctoral students who were placed in research jobs in industry and academia.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2016

SU-F-I-46: Optimizing Dose Reduction in Adult Head CT Protocols While Maintaining Image Quality in Postmortem Head Scans

Purpose: To optimize adult head CT protocol by reducing dose to an appropriate level while providing CT images of diagnostic quality. Methods: Five cadavers were scanned from the skull base to the vertex using a routine adult head CT protocol (120 kVp, 270 mA, 0.75 s rotation, 0.5 mm × 32 detectors, 70.8 mGy CTDIvol) followed by seven reduced-dose protocols with varying combinations of reduced tube current, reduced rotation time, and increased detectors with CTDIvol ranging from 38.2 to 65.6 mGy. Organ doses were directly measured with 21 OSL dosimeters placed on the surface and implanted in the head by a neurosurgeon. Two neuroradiologists assessed grey-white matter differentiation, fluid space, ventricular size, midline shift, brain mass, edema, ischemia, and skull fractures on a three point scale: (1) Unacceptable, (2) Borderline Acceptable, and (3) Acceptable. Results: For the standard scan, doses to the skin, lens of the eye, salivary glands, thyroid, and brain were 37.55 mGy, 49.65 mGy, 40.67 mGy, 4.63 mGy, and 27.33 mGy, respectively. Two cadavers had cerebral edema due to changing dynamics of postmortem effects, causing the grey-white matter differentiation to appear less distinct. Two cadavers with preserved grey-white matter received acceptable scores for all image quality features for the protocol with a CTDIvol of 57.3 mGy, allowing organ dose savings ranging from 34% to 45%. One cadaver allowed for greater dose reduction for the protocol with a CTDIvol of 42 mGy. Conclusion: Efforts to optimize scan protocol should consider both dose and clinical image quality. This is made possible with postmortem subjects, whose brains are similar to patients, allowing for an investigation of ideal scan parameters. Radiologists at our institution accepted scan protocols acquired with lower scan parameters, with CTDIvol values closer to the American College of Radiology’s (ACR) Achievable Dose level of 57 mGy.

60 APPLIED LIFE SCIENCES↗
NASA NTRS1974

Design of state-feedback controllers including sensitivity reduction, with applications to precision pointing

The design procedure of feedback controllers was described and the considerations for the selection of the design parameters were given. The frequency domain properties of single-input single-output systems using state feedback controllers are analyzed, and desirable phase and gain margin properties are demonstrated. Special consideration is given to the design of controllers for tracking systems, especially those designed to track polynomial commands. As an example, a controller was designed for a tracking telescope with a polynomial tracking requirement and some special features such as actuator saturation and multiple measurements, one of which is sampled. The resulting system has a tracking performance comparing favorably with a much more complicated digital aided tracker. The parameter sensitivity reduction was treated by considering the variable parameters as random variables. A performance index is defined as a weighted sum of the state and control convariances that sum from both the random system disturbances and the parameter uncertainties, and is minimized numerically by adjusting a set of free parameters.

Hadass, Z.↗
DOE OSTI2013

Implementation of hybrid variance reduction methods in a multi group Monte Carlo code for deep shielding problems

In this paper, the work that has been done to implement variance reduction techniques in a three dimensional, multi group Monte Carlo code - Tortilla, that works within the frame work of the commercial deterministic code - Attila, is presented. This project is aimed to develop an integrated Hybrid code that seamlessly takes advantage of the deterministic and Monte Carlo methods for deep shielding radiation detection problems. Tortilla takes advantage of Attila's features for generating the geometric mesh, cross section library and source definitions. Tortilla can also read importance functions (like adjoint scalar flux) generated from deterministic calculations performed in Attila and use them to employ variance reduction schemes in the Monte Carlo simulation. The variance reduction techniques that are implemented in Tortilla are based on the CADIS (Consistent Adjoint Driven Importance Sampling) method and the LIFT (Local Importance Function Transform) method. These methods make use of the results from an adjoint deterministic calculation to bias the particle transport using techniques like source biasing, survival biasing, transport biasing and weight windows. The results obtained so far and the challenges faced in implementing the variance reduction techniques are reported here. (authors)

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Sub-micrometer random-pyramid texturing of silicon solar wafers with excellent surface passivation and low reflectance

Select emerging solar cell technologies, ranging from very narrow screen-printed fingers to perovskite/silicon tandems, would benefit from a reduction of the height of the standard pyramidal texture of monocrystalline silicon wafers to below 1 μm. However, fabricating such small surface features usually compromises the requisite low reflectance and high passivation quality. Through systematic design of experiments, we demonstrate that conventional alkaline-based wet-chemical etching processes can be tuned to produce dense, spatially homogeneous, and uniformly sized sub-micrometer pyramids. Precise texture size can be further controlled by adding potassium silicate to the texturing solution to effectively slow the pyramid growth rate and promote nucleation. With pyramid size distribution data—extracted from multiple atomic force microscopy images across each wafer—we demonstrate 16 nanotextures with average pyramid sizes ranging from 62 to 512 nm, with the largest having only 0.3% of its pyramids taller than 1 μm. Six of the nanotextures exhibit light trapping and surface defect densities comparable to state-of-the-art microtextured wafers. In conclusion, the best nanotexture achieved an AM1.5G-weighted reflectance of 11.8% and a minority-carrier lifetime of 3.4 ms after passivation with amorphous silicon, demonstrating that smaller textures need not deteriorate solar cell performance.

14 SOLAR ENERGY↗
DOE OSTI2024

A Theoretical Benchmark of the Geometric and Optical Properties for 3d Transition Metal Nanoclusters via Density Functional Theory

Understanding structure–property relationships in atomically precise metal nanoclusters is vital in finding selective and tunable catalysts. Here, in this study, density functional theory (DFT) was used to benchmark seven exchange correlation functionals at different basis sets for 17 atomically precise nanoclusters against experimentally determined geometries, band gaps, and optical gaps. The set contains both monometallic and bimetallic clusters that possess at least two types of 3d transition metals (specifically, Cu, Ni, Fe, or Co). The benchmark highlights that PBE0 is a good functional to use regardless of the basis set, and Minnesota functionals do well with respect to specific metals. Further, while long-range corrected functionals overestimate band and optical gaps, they model absorption features better than the other considered functionals. The study additionally looks at the photoinduced hydrogen evolution reaction (HER) and the CO 2 reduction mechanism on nanoclusters reported from the literature.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Chemical Kinetic Method for Active-Site Quantification in Fe-N-C Catalysts and Correlation with Molecular Probe and Spectroscopic Site-Counting Methods

Mononuclear Fe ions ligated by nitrogen (FeN x ) dispersed on nitrogen-doped carbon (Fe-N-C) serve as active centers for electrocatalytic O 2 reduction and thermocatalytic aerobic oxidations. Despite their promise as replacements for precious metals in a variety of practical applications, such as fuel cells, the discovery of new Fe-N-C catalysts has relied primarily on empirical approaches. In this context, the development of quantitative structure–reactivity relationships and benchmarking of catalysts prepared by different synthetic routes and by different laboratories would be facilitated by the broader adoption of methods to quantify atomically dispersed FeN x active centers. In the present study, we develop a kinetic probe reaction method that uses the aerobic oxidation of a model hydroquinone substrate to quantify the density of FeN x centers in Fe- N-C catalysts. The kinetic method is compared with low-temperature Mössbauer spectroscopy, CO pulse chemisorption, and electrochemical reductive stripping of NO derived from NO 2 – on a suite of Fe-N-C catalysts prepared by diverse routes and featuring either the exclusive presence of Fe as FeN x sites or the coexistence of aggregated Fe species in addition to FeN x . The FeN x site densities derived from the kinetic method correlate well with those obtained from CO pulse chemisorption and Mössbauer spectroscopy. The broad survey of Fe-N-C materials also reveals the presence of outliers and challenges associated with each site quantification approach. Furthermore, the kinetic method developed here does not require pretreatments that may alter active-site distributions nor specialized equipment beyond reaction vessels and analytical instrumentation (e.g., NMR).

36 MATERIALS SCIENCE↗
DOE OSTI2024

Tracking the topology of neural manifolds across populations

Neural manifolds summarize the intrinsic structure of the information encoded by a population of neurons. Advances in experimental techniques have made simultaneous recordings from multiple brain regions increasingly commonplace, raising the possibility of studying how these manifolds relate across populations. However, when the manifolds are nonlinear and possibly code for multiple unknown variables, it is challenging to extract robust and falsifiable information about their relationships. We introduce a framework, called the method of analogous cycles, for matching topological features of neural manifolds using only observed dissimilarity matrices within and between neural populations. We demonstrate via analysis of simulations and in vivo experimental data that this method can be used to correctly identify multiple shared circular coordinate systems across both stimuli and inferred neural manifolds. Conversely, the method rejects matching features that are not intrinsic to one of the systems. Further, as this method is deterministic and does not rely on dimensionality reduction or optimization methods, it is amenable to direct mathematical investigation and interpretation in terms of the underlying neural activity. We thus propose the method of analogous cycles as a suitable foundation for a theory of cross-population analysis via neural manifolds.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Population Count on Intel® CPU, GPU, and FPGA

Population count is a primitive used in many applications. Commodity processors have dedicated instructions for achieving high-performance population count. Motivated by the productivity of high-level synthesis and the importance of population count, in this paper we investigated the OpenCL implementations of population count algorithms, and evaluated their performance and resource utilizations on an FPGA. Based on the results, we select the most efficient implementation. Then we derived a reduction pattern from a representative application of population count. We parallelized the reduction with atomic functions, and optimized it with vectorized memory accesses, tree reduction, and compute-unit duplication. We evaluated the performance of the reduction kernel on an Intel (R) Xeon (R) CPU and an Intel (R) IrisT Pro integrated GPU, and an FPGA card that features an Intel (R) Arria (R) 10 FPGA. When DRAM memory bandwidth is comparable on the three computing platforms, the FPGA can achieve the highest kernel performance for large workload. On the other hand, we described performance bottlenecks on the FPGA. To make FPGAs more competitive in raw performance compared to high-performant CPU and GPU platforms, it is important to increase external memory bandwidth, minimize data movement between a host and a device, and reduce OpenCL runtime overhead on an FPGA.

heterogeneous computing↗
DOE OSTI2025

Moltensaltpropnet

MoltenSaltPropnet is a physics-informed machine learning framework that aims to predict the thermophysical properties of molten fluoride and chloride salt mixtures, which are crucial for the design and safety of Generation IV molten salt reactors. The code processes data from the Molten-Salt Thermal Properties Database (MSTDB-TP) and the Janz compendium, converting critically evaluated correlations into fast, differentiable surrogate models for density, viscosity, thermal conductivity, and heat capacity across 448 distinct salt systems. The implementation consists of several key components: 1. Data Curation: The code parses and cleans the raw data, normalizing elemental mole fractions and extracting relevant regression coefficients for various thermophysical properties. 2. Feature Engineering: It generates fixed-length numerical descriptors that encapsulate the composition and temperature, incorporating polynomial interaction terms and dimensionality-reduction techniques to optimize model performance. 3. Coefficient Learning: Four different machine learning architectures are employed: a deep residual network (ResNet), a Kolmogorov–Arnold network (KAN), a sparsity-inducing neural network (SNN), and classical regression models. Each model learns to predict coefficients that define the temperature-dependent correlations for the thermophysical properties. 4. Property Reconstruction: The predicted coefficients are used to compute temperature-dependent property values, ensuring positivity and monotonic trends through a composite loss function that enforces physical constraints. 5. User Interface: An open-source web application enables users to filter the database, train task-specific models, and visualize the results, allowing for rapid exploration of candidate salt mixtures. MoltenSaltPropnet bridges the gap between limited experimental data and high-fidelity reactor simulations, providing a powerful tool for researchers in the field of molten salt reactors and advanced nuclear energy systems.

Retamales, Mauricio Eduardo Tano [Idaho National L↗
DOE OSTI2011

An application-level parallel I/O library for Earth system models

We describe the design and implementation of an application-level parallel I/O (PIO) library for the reading and writing of distributed arrays to several common scientific data formats. PIO provides the flexibility to control the number of I/O tasks through data rearrangement to an I/O friendly decomposition. This flexibility enables reductions in per task memory usage and improvements in disk I/O performance versus a serial I/O approach. We illustrate the impact various features within PIO have on memory usage and disk I/O bandwidth on a Cray XT5 system.

Computer Science↗
DOE OSTI2012

Transportation Energy Pathways LDRD.

This report presents a system dynamics based model of the supply-demand interactions between the US light-duty vehicle (LDV) fleet, its fuels, and the corresponding primary energy sources through the year 2050. An important capability of our model is the ability to conduct parametric analyses. Others have relied upon scenario-based analysis, where one discrete set of values is assigned to the input variables and used to generate one possible realization of the future. While these scenarios can be illustrative of dominant trends and tradeoffs under certain circumstances, changes in input values or assumptions can have a significant impact on results, especially when output metrics are associated with projections far into the future. This type of uncertainty can be addressed by using a parametric study to examine a range of values for the input variables, offering a richer source of data to an analyst.The parametric analysis featured here focuses on a trade space exploration, with emphasis on factors that influence the adoption rates of electric vehicles (EVs), the reduction of GHG emissions, and the reduction of petroleum consumption within the US LDV fleet. The underlying model emphasizes competition between 13 different types of powertrains, including conventional internal combustion engine (ICE) vehicles, flex-fuel vehicles (FFVs), conventional hybrids(HEVs), plug-in hybrids (PHEVs), and battery electric vehicles(BEVs).We find that many factors contribute to the adoption rates of EVs. These include the pace of technological development for the electric powertrain, battery performance, as well as the efficiency improvements in conventional vehicles. Policy initiatives can also have a dramatic impact on the degree of EV adoption. The consumer effective payback period, in particular, can significantly increase the market penetration rates if extended towards the vehicle lifetime.Widespread EV adoption can have noticeable impact on petroleum consumption and greenhouse gas(GHG) emission by the LDV fleet. However, EVs alone cannot drive compliance with the most aggressive GHG emission reduction targets, even as the current electricity source mix shifts away from coal and towards natural gas. Since ICEs will comprise the majority of the LDV fleet for up to forty years, conventional vehicle efficiency improvements have the greatest potential for reductions in LDV GHG emissions over this time.These findings seem robust even if global oil prices rise to two to three times current projections. Thus,investment in improving the internal combustion engine might be the cheapest, lowest risk avenue towards meeting ambitious GHG emission and petroleum consumption reduction targets out to 2050.3 Acknowledgment The authors would like to thank Dr. Andrew Lutz, Dr. Benjamin Wu, Prof. Joan Ogden and Dr. Christopher Yang for their suggestions over the course of this project. This work was funded by the Laboratory Directed Research and Development program at Sandia National Laboratories.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2019

Fall 2019 Dissertation Update [Slides]

Summary of Presentation: Motivation - Problem Description; Feature Identification. Discussion of Features - Flat Flux Region Near Cask Centerline; Periodic Depressions of Flux in Fuel Region; Abrupt "Level-Off" Region; Non-exponential Reduction in Overpack; Asymmetric Flux. Future Work.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2024

Deep Learning-based Parameterization of Complex 3D CO2 Saturation Data in Large-scale Geological Carbon Storage

In deep learning (DL), dimension reduction plays a pivotal role in improving training efficiency and minimizing overfitting, especially when working with complex datasets like three-dimensional (3D) saturation data. In the context of geological carbon storage (GCS), 3D saturation data introduces unique challenges due to its sparse nature and sharp transitions at plume boundaries, known as shock fronts. To tackle these challenges, we developed a novel DL framework that combines dimension reduction with advanced 3D reconstruction techniques. Our approach utilizes latent variables derived from 2D average saturation fields to efficiently capture the essential features of high-dimensional data while reducing the number of variables. This enhances both the robustness and accuracy of DL models, making the framework more practical for real-world applications. By offering a tailored solution for modeling complex 3D saturation dynamics, this framework holds significant potential for environmental monitoring, energy storage, and other geological applications.

Wang, Hongsheng [University of Texas at Austin]↗
NASA NTRS1976

Lunar cartography with the Apollo 17 ALSE radar imagery

Lunar position differences between thirteen craters in Mare Serenitatis were computed from VHF radar-imagery obtained by the Lunar Sounder instrument flown on the Apollo 17 command module. The radar-derived position differences agree with those obtained by conventional photogrammetric reductions of Apollo metric photography. This demonstrates the feasibility of using the Apollo Lunar Sounder data to determine the positions of lunar features along the Apollo 17 orbital tracks. This will be particularly useful for western limb and farside areas, where no Apollo metric camera pictures are available.

Tiernan, M.↗