Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “residual network”

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 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Evaluating the location capabilities of a regional infrasonic network in Utah, US, using both ray tracing-derived and empirical-derived celerity-range and backazimuth models

SUMMARY More realistic models for infrasound signal propagation across a region can be used to improve the precision and accuracy of spatial and temporal source localization estimates. Motivated by incomplete infrasound event bulletins in the Western US, the location capabilities of a regional infrasonic network of stations located between 84–458 km from the Utah Test and Training Range, Utah, USA, is assessed using a series of near-surface explosive events with complementary ground truth (GT) information. Signal arrival times and backazimuth estimates are determined with an automatic F-statistic based signal detector and manually refined by an analyst. This study represents the first application of three distinct celerity-range and backazimuth models to an extensive suite of realistic signal detections for event location purposes. A singular celerity and backazimuth deviation model was previously constructed using ray tracing analysis based on an extensive archive of historical atmospheric specifications and is applied within this study to test location capabilities. Similarly, a set of multivariate, season and location specific models for celerity and backazimuth are compared to an empirical model that depends on the observations across the infrasound network and the GT events, which accounts for atmospheric propagation variations from source to receiver. Discrepancies between observed and predicted signal celerities result in locations with poor accuracy. Application of the empirical model improves both spatial localization precision and accuracy; all but one location estimates retain the true GT location within the 90 per cent confidence bounds. Average mislocation of the events is 15.49 km and average 90 per cent error ellipse areas are 4141 km2. The empirical model additionally reduces origin time residuals; origin time residuals from the other location models are in excess of 160 s while residuals produced with the empirical model are within 30 s of the true origin time. We demonstrate that event location accuracy is driven by a combination of signal propagation model and the azimuthal gap of detecting stations. A direct relationship between mislocation, error ellipse area and increased station azimuthal gaps indicate that for sparse networks, detection backazimuths may drive location biases over traveltime estimates.

58 GEOSCIENCES↗
DOE OSTI2024

Challenges in Training PINNs: A Loss Landscape Perspective

This paper explores challenges in training Physics Informed Neural Networks (PINNs), emphasizing the role of the loss landscape in the training process. We examine difficulties in minimizing the PINN loss function, particularly due to ill conditioning caused by differential operators in the residual term. We compare gradient-based optimizers Adam, L-BFGS, and their combination Adam+L-FGS, showing the superiority of Adam+L-BFGS, and introduce a novel secondorder optimizer, NysNewton-CG (NNCG), which significantly improves PINN performance. Theoretically, our work elucidates the connection between ill-conditioned differential operators and ill-conditioning in the PINN loss and shows the benefits of combining first- and second-order optimization methods. Our work presents valuable insights and more powerful optimization strategies for training PINNs, which could improve the utility of PINNs for solving difficult partial differential equations.

Rathore, Pratik↗
DOE OSTI2025

Enhancing Drinking Water Quality Modeling: Leveraging Physics Informed Neural Networks for Learning with Imperfect Reaction Models and Partial Data

Chemical kinetics models, typically formulated as systems of ordinary or partial differential equations, are valuable tools for simulating drinking water quality. However, these models often face inaccuracies due to discrepancies between the laboratory and the real-world conditions, as well as limitations in experimental analytical methods, hindering the accurate representation of the true underlying chemical mechanisms. In this study, we propose a Physics Informed Neural Network (PINN), using the eXtreme Theory of Functional Connections, to improve the prediction of chemical concentrations over time. The PINN method accounts for imperfect chemical models and incorporates partial data to improve predictions. Focusing on reactions describing water disinfection residual and disinfectant byproduct formation, which are crucial for public health and regulatory compliance, we demonstrate that the PINN model is able to accurately predict the concentrations of chemical species across various pH values. Notably, the model extends its accuracy to predict concentrations of chemical species not originally included in its training data. The developed method can be extended to a variety of chemical systems, offering a wide array of potential applications.

13 HYDRO ENERGY↗
DOE OSTI2022

Network analysis uncovers the communication structure of SARS-CoV-2 spike protein identifying sites for immunogen design

The COVID-19 pandemic, caused by the SARS-CoV-2 virus, has triggered myriad efforts to understand the structure and dynamics of this complex pathogen. The spike glycoprotein of SARS-CoV-2 is a significant target for immunogens as it is the means by which the virus enters human cells, while simultaneously sporting mutations responsible for immune escape. These functional and escape processes are regulated by complex molecular-level interactions. Our study presents quantitative insights on domain and residue contributions to allosteric communication, immune evasion, and local- and global-level control of functions through the derivation of a weighted graph representation from all-atom MD simulations. Focusing on the ancestral form and the D614G-variant, we provide evidence of the utility of our approach by guiding the selection of a mutation that alters the spike’s stability. Taken together, the network approach serves as a valuable tool to evaluate communication “hot-spots” in proteins to guide design of stable immunogens.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2024

Multisubstrate specificity shaped the complex evolution of the aminotransferase family across the tree of life

Aminotransferases (ATs) are an ancient enzyme family that play central roles in core nitrogen metabolism, essential to all organisms. However, many of the AT enzyme functions remain poorly defined, limiting our fundamental understanding of the nitrogen metabolic networks that exist in different organisms. Here, we traced the deep evolutionary history of the AT family by analyzing AT enzymes from 90 species spanning the tree of life (ToL). We found that each organism has maintained a relatively small and constant number of ATs. Mapping the distribution of ATs across the ToL uncovered that many essential AT reactions are carried out by taxon-specific AT enzymes due to wide-spread nonorthologous gene displacements. This complex evolutionary history explains the difficulty of homology-based AT functional prediction. Biochemical characterization of diverse aromatic ATs further revealed their broad substrate specificity, unlike other core metabolic enzymes that evolved to catalyze specific reactions today. Interestingly, however, we found that these AT enzymes that diverged over billion years share common signatures of multisubstrate specificity by employing different nonconserved active site residues. These findings illustrate that AT family enzymes had leveraged their inherent substrate promiscuity to maintain a small yet distinct set of multifunctional AT enzymes in different taxa. This evolutionary history of versatile ATs likely contributed to the establishment of robust and diverse nitrogen metabolic networks that exist throughout the ToL. The study provides a critical foundation to systematically determine diverse AT functions and underlying nitrogen metabolic networks across the ToL.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Optimization-Based Azeotropic Distillation System Synthesis Using Geometric Insights

The synthesis of azeotropic distillation systems is challenging due to the existence of compartments in the residue curve map coupled with the combinatorial aspects from numerous possible system configurations. In this work, an optimization-based approach is introduced to synthesize homogeneous azeotropic distillation systems. The approach employs a network-based representation generated via a matrix method. To design the distillation columns, the modified Underwood equations are adopted, in which pseudocomponent-based compositions are used. For cases where separatrices are significantly curved, corrections via piece-wise linear functions and collinearity properties are implemented. Two examples are presented to illustrate the proposed approach.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

B/PS bulges in DESI Legacy edge-on galaxies – I. Sample building

ABSTRACT We present the biggest up-to-date sample of edge-on galaxies with boxy/peanut-shaped (B/PS) bulges and X-structures. The sample was prepared using images from the Dark Energy Spectroscopic Instrument (DESI) Legacy catalogue and contains about 2000 galaxies. To find suitable candidates in catalogue, we made the assumption that the residues (original images minus model) of galaxies with B/PS bulges should exhibit a characteristic X-shape. Galaxies with such features were selected by eye and then used as input data for a neural network training, which was applied to a bigger sample of edge-on galaxies. Using the available data and the photometric models from the literature, we investigated the observational and statistical properties of the sample created. Comparing the B/D ratios for galaxies with and without B/PS bulges, we found that the B/D ratio for galaxies from our sample is statistically higher, with typical values in the range ≈0.2–0.5 depending on the decomposition procedure. We studied how the opening angles φ of the X-structure and the length of its rays are distributed in the formed sample and found them to be consistent with previous measurements and predictions from N-body models, e.g. $\varphi \gtrsim 25{^\circ }$, but measured here for a much larger number of galaxies. We found a sharp increase in the B/PS bulge fraction for stellar masses log M⋆ ≳ 10.4, but for edge-on galaxies, which complements the results of previous works. The sample can be used in future work to test various bar models and their relationship with B/PS bulges, as well as to study their stability and evolution.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Attack on Grid Event Cause Analysis: An Adversarial Machine Learning Approach

With the ever-increasing reliance on data for data-driven applications in power grids, such as event cause analysis, the authenticity of data streams has become crucially important. The data can be prone to adversarial stealthy attacks aiming to manipulate the data such that residual-based bad data detectors cannot detect them, and the perception of system operators or event classifiers changes about the actual event. This paper investigates the impact of adversarial attacks on convolutional neural network-based event cause analysis frameworks. We have successfully verified the ability of adversaries to maliciously misclassify events through stealthy data manipulations. The vulnerability assessment is studied with respect to the number of compromised measurements. Furthermore, a defense mechanism to robustify the performance of the event cause analysis is proposed. The effectiveness of adversarial attacks on changing the output of the framework is studied using the data generated by real-time digital simulator (RTDS) under different scenarios such as type of attacks and level of access to data.

Niazazari, Iman↗
NASA NTRS2011

An Intercomparison of Tropospheric Ozone Retrievals Derived from Two Aura Instruments and Measurements in Western North America in 2006

Two recently developed methods for quantifying tropospheric ozone abundances based on Aura data, the Trajectory‐enhanced Tropospheric Ozone Residual (TTOR) and an assimilation of Aura data into Goddard Earth Observing System Version 4 (ASM), are compared to ozone measurements from ozonesonde data collected in April-May 2006 during the INTEX Ozonesonde Network Study 2006 (IONS‐06) campaign. Both techniques use Ozone Monitoring Instrument (OMI) and Microwave Limb Sounder (MLS) observations. Statistics on column ozone amounts for both products are presented. In general, the assimilation compares better to sonde integrated ozone to 200 hPa (28.6% difference for TTOR versus 2.7% difference for ASM), and both products are biased low. To better characterize the performance of ASM, ozone profiles based on the assimilation are compared to those from ozonesondes. We noted slight negative biases in the lower troposphere, and slight positive biases in the upper troposphere/lower stratosphere (UT/ LS), where we observed the greatest variability. Case studies were used to further understand ASM performance. We examine one case from 17 April 2006 at Bratt's Lake, Saskatchewan, where geopotential height gradients appear to be related to an underestimation in the ASM in the UT/LS region. A second case, from 21 April 2006 at Trinidad Head, California, is a situation where the overprediction of ozone in the UT/LS region does not appear to be due to current dynamic conditions but seems to be related to uncertainty in the flow pattern and large differences in MLS observations upstream.

UTLS↗
DOE OSTI2023

Enhanced rare-earth separation with a metal-sensitive lanmodulin dimer

Technologically critical rare-earth elements are notoriously difficult to separate, owing to their subtle differences in ionic radius and coordination number. The natural lanthanide-binding protein lanmodulin (LanM) is a sustainable alternative to conventional solvent-extraction-based separation. Here we characterize a new LanM, from Hansschlegelia quercus (Hans-LanM), with an oligomeric state sensitive to rare-earth ionic radius, the lanthanum(III)-induced dimer being >100-fold tighter than the dysprosium(III)-induced dimer. X-ray crystal structures illustrate how picometre-scale differences in radius between lanthanum(III) and dysprosium(III) are propagated to Hans-LanM’s quaternary structure through a carboxylate shift that rearranges a second-sphere hydrogen-bonding network. Comparison to the prototypal LanM from Methylorubrum extorquens reveals distinct metal coordination strategies, rationalizing Hans-LanM’s greater selectivity within the rare-earth elements. Finally, structure-guided mutagenesis of a key residue at the Hans-LanM dimer interface modulates dimerization in solution and enables single-stage, column-based separation of a neodymium(III)/dysprosium(III) mixture to >98% individual element purities. This work showcases the natural diversity of selective lanthanide recognition motifs, and it reveals rare-earth-sensitive dimerization as a biological principle by which to tune the performance of biomolecule-based separation processes.

36 MATERIALS SCIENCE↗
DOE OSTI2025

High-quality composite Pati-Salam axion

We present a composite QCD axion model where the Peccei-Quinn symmetry emerges as a high-quality, accidental symmetry. The axion potential is only modified by eight-fermion, dimension-12 operators, which if present at the Planck scale, allow for axion dark matter from misalignment while solving the strong 𝐶⁢𝑃 problem. The model is an SU⁡(𝑁 𝑐 ) gauge theory with ten flavors where the Pati-Salam unified subgroup SO⁡(6) × SO⁡(4) ⊂ SU⁢(10)𝐿 and Sp⁡(10) ⊂ SU⁢(10) 𝑅 are weakly gauged. The dynamics breaks SU⁢(10) 𝐿 × SU⁢(10) 𝑅 → SU⁢(10) 𝑉 and the weakly gauged groups to U⁡(3) × U⁡(2) ⊃ SU⁢(3) 𝑐 × SU⁢(2) 𝐿 × U⁡(1) 𝑌 , with the QCD axion identified as one of the Nambu-Goldstone bosons. This axion has a relatively large coupling to photons while a residual $\bar{𝜃}$ eff may be just below the current limit on the neutron electric dipole moment. If the dimension-12 operators are present near the grand unified theory scale, they can cause domain wall networks to decay, allowing for axion dark matter even for the postinflationary scenario.

Gherghetta, Tony [University of Minnesota, Minneap↗
NASA NTRS1977

Range validation using Kalman filter techniques

Range pseudo-residuals may be improved to the level required for data validation by a measurement updating process which utilizes Bierman's adaptation of the Kalman filter measurement updating algorithms together with process noise compensation to account for model errors. This algorithm involves combining the currently available range predictions and measurements to produce an updated range residual measurement whose accuracy is constrained by the range data quality and by the estimated error in the prediction. The algorithm is compact and fast, and is thus suitable for on-line applications in network control or at the station.

Madrid, G. A.↗
NASA NTRS1990

A statistical study of the differential very long baseline interferometry delay rate error budget

The delta differential one-way Doppler (delta-DOD) error budget is assessed by analyzing 12 years of dual-frequency Deep Space Network very long baseline interferometry data. A typical delta-DOD observation is outlined, and the Magellan error budget along with tropospheric effects are considered. Formation of quasar-quasar observables, RMS delta-DOD residual, angular source separation and time separation between observations are discussed, along with elevation angle and atmospheric pathlength dependence, solar-plasma and ionospheric effects. It is concluded that the weak dependence of the delta-DOD error on angular source separation can mean that quasars with larger angular separation from a spacecraft may be used as navigation reference with little increase in the delta-DOD error. It is also indicated that the delta-DOD errors are dominated by wet troposphere fluctuations, and the results of numerical integrations for the single-station RMS delay rate at different elevation angles are presented.

Edwards, Charles D.↗
NASA NTRS2013

NASA Technology Evaluation for Environmental Risk Mitigation Remediation Technology Collaboration Development

NASA is committed to finding solutions to agency cleanup problems that are better, cheaper, and more effective than the status quo. Unfortunately, some potential solutions involve innovative technologies for which NASA remediation managers may not have a high level of understanding or confidence. Since 2004, NASA's Stennis Space Center (SSC) in Mississippi has been pumping groundwater contaminated with trichloroethylene (TCE) and other halogenated volatile organic compounds (HVOC) from their cleanup location designated "Area G" through extraction wells to an aboveground treatment system. Over time, however, the effectiveness of this treatment strategy has diminished and an alternative approach is needed. In 2012, professionals from NASA's Principal Center for Technology Evaluation for Environmental Risk Mitigation (TEERM) introduced SSC managers to an innovative technology for enhancing the performance of SSC's existing pump and treat system. The technology, generally referred to as in situ chemical oxidation (ISCO), involves slowly and continuously injecting a strong but safe chemical oxidant into the groundwater. Treatment is enhanced by a "surfactant-type effect" which causes residual contamination from saturated soil to be released into the dissolved-phase where it can be readily oxidized. Any dissolved-phase contamination that was not oxidized can be collected by the extraction well network and treated aboveground. SSC was not familiar with the technology so to increase their confidence, TEERM identified a contractor who was willing to demonstrate their product and process at a significantly reduced price. An initial, small-scale demonstration of ISCO began at sse in March 2012 and completed in August 2012. This successful demonstration was followed by three larger-scale ISCO demonstrations between August and December 2012. The contractor's innovative Continuous Injection System (CIS) incorporated "green" and sustainable technologies and practices. A slow injection rate was maintained autonomously by the CIS, eliminating the need for multiple mobilizations of personnel and powered equipment. The CIS was calibrated to deliver only as much reagent as the formation would accept without "short circuiting", minimizing material waste. Public utility water pressure was used to mix, dilute, and inject the reagent. NASA personnel were trained to operate and maintain the system and remote monitoring and injection control capabilities were developed, further reducing the need for contractor mobilizations to the site. An integrated solar photovoltaic panel was used to power the control valves and web monitoring telemetry. SSC provided hydrogen peroxide for the process using their existing supplier for the aboveground pump and treat systems. The only consumable required from the contractor was their proprietary activator to catalyze SSC's peroxide.

Romeo, James↗
DOE OSTI2020

Machine Learning Approaches for Nuclear Material Accounting Data from Irradiation and Reprocessing

We are currently exploring data analysis methods for their ability to strengthen the synthesis and evaluation of information generated within domestic and international safeguard regimes. Safeguard data typically includes rich heterogenous datasets amenable to advanced data analytics methods, such as machine learning. We have converted transactional data containing both numerical and categorical attributes from a domestic nuclear material control and accountability (NMC&A) system into a low-dimensional numerical structure via linear principal component analysis. This data representation allows for global structure discovery via cluster analysis, which can characterize the typical behavior of each of the primary types of transaction events. Furthermore, the structure of the top principal components captures the “typical behavior” of the data and is thus amenable to anomaly detection through statistical hypothesis testing. We explored this capability by generating erroneous permutations of the data and computing the Q-residual quantity increase associated with the information loss when these data are projected into the low-dimensional principal component analysis space representative of typical transactions.Future work will focus on identifying data transformations (e.g., graph networks) that more closely align with the inherent structure of these transactions to explain more salient information and disambiguate the underlying nuclear process—in this case, irradiation and reprocessing—from artifacts of NMC&A system transactional record keeping data.

Drescher, Adam↗
DOE OSTI2026

Study on Application of Distributed Network of Sensors with List Mode for NMAC Literature Review

Nuclear material accounting and control (NMAC) for nuclear security detects, deters, and resolves questions related to unauthorized removal (i.e. theft) or misuse of nuclear material. NMAC also serves as a key insider threat mitigation measure and aids in recovery of nuclear material that is missing. Effective nuclear security depends on NMAC for timely and accurate information about nuclear material types, quantities, and locations. Bulk nuclear material processing facilities, however, present unique challenges for effective NMAC due to the presence of large quantities of material in-process and the accumulation of residual material holdup within process equipment. These holdup accumulations can obscure accurate physical inventory taking and complicate efforts to resolve NMAC irregularities at the facility level. Bulk material monitoring systems often rely on material balance calculations and indirect measurement techniques, which may mask protracted theft of smaller amounts of nuclear material. These monitoring limitations have generated increased interest in continuous monitoring technologies, including distributed non-destructive assay (NDA) sensor networks capable of providing real-time or near-real-time measurement of material movement and accumulation within bulk processing environments. Recent advancements in distributed networks of NDA radiation detectors and sensing technologies provide an opportunity to address these limitations. Although such distributed sensor networks have been implemented in select facilities for IAEA Safeguards applications, their potential for supporting NMAC functions specifically tailored to nuclear security objectives remains largely unexplored. Furthermore, emerging list-mode data acquisition technologies have reached high technology readiness levels, enabling time-correlated detection of nuclear events across multiple temporal scales. These capabilities provide enhanced opportunities for accurate holdup measurement, continuous process monitoring, and improved detection of material theft or misuse over time. The increasing global expansion of civil nuclear power and development of related bulk material processing facilities, including those supporting high-assay low-enriched uranium (HALEU) and other advanced reactor fuel fabrication, further increases the need for advanced measurement and monitoring strategies for NMAC.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
NASA NTRS1994

Gravity model improvement using the DORIS tracking system on the SPOT 2 satellite

A high-precision radiometric satellite tracking system, Doppler Orbitography and Radio-positioning Integrated by Satellite system (DORIS), has recently been developed by the French space agency, Centre National d'Etudes Spatiales (CNES). DORIS was designed to provide tracking support for missions such as the joint United States/French TOPEX/Poseidon. As part of the flight testing process, a DORIS package was flown on the French SPOT 2 satellite. A substantial quantity of geodetic quality tracking data was obtained on SPOT 2 from an extensive international DORIS tracking network. These data were analyzed to assess their accuracy and to evaluate the gravitational modeling enhancements provided by these data in combination with the Goddard Earth Model-T3 (GEM-T3) gravitational model. These observations have noise levels of 0.4 to 0.5 mm/s, with few residual systematic effects. Although the SPOT 2 satellite experiences high atmospheric drag forces, the precision and global coverage of the DORIS tracking data have enabled more extensive orbit parameterization to mitigate these effects. As a result, the SPOT 2 orbital errors have been reduced to an estimated radial accuracy in the 10-20 cm RMS range. The addition of these data, which encompass many regions heretofore lacking in precision satellite tracking, has significantly improved GEM-T3 and allowed greatly improved orbit accuracies for Sun-synchronous satellites like SPOT 2 (such as ERS 1 and EOS). Comparison of the ensuing gravity model with other contemporary fields (GRIM-4C2, TEG2B, and OSU91A) provides a means to assess the current state of knowledge of the Earth's gravity field. Thus, the DORIS experiment on SPOT 2 has provided a strong basis for evaluating this new orbit tracking technology and has demonstrated the important contribution of the DORIS network to the success of the TOPEX/Poseidon mission.

Nerem, R. S.↗
DOE OSTI2021

Physics-Informed Neural Network Method for Forward and Backward Advection-Dispersion Equations

Advection-dispersion equations (ADEs) are commonly used to describe transport phenomena in porous media. Even though mature discretization-based numerical methods for ADEs exist, some challenges still remain, especially when it comes to solving advection-dominated forward ADEs and diffusion-dominated backward ADEs. The latter problem usually arises in the source identification context and leads to numerically unstable grid-based solutions that require a form of regularization or should be treated as an inverse problem that is computationally more expensive because it requires solving the forward problem multiple times. In this study, we propose a discretization-free approach based on the physics-informed neural network (PINN) method for solving coupled ADE and Darcy flow equations with space-dependent hydraulic conductivity. In this approach, the hydraulic conductivity, hydraulic head, and concentration fields are approximated with deep neural networks (DNNs). We assume that the conductivity field is given by its values on a grid, and we use these values to train the conductivity DNN. The head and concentration DNNs are trained by minimizing the residuals of the flow equation and ADE and using the initial and boundary conditions as additional constraints. The PINN method is applied to one- and two-dimensional forward ADE problems, where its performance for various P\'{e}clet numbers ($Pe$) is compared with the analytical and numerical solutions. We find that the PINN method is accurate with errors of less than 1\% and outperforms some conventional discretization-based methods for $Pe$ larger that 100. Next, we demonstrate that the PINN method remains accurate for the backward ADEs, with the relative errors in most cases staying under 5\% compared to the reference concentration field. Finally, we show that when available, the concentration measurements can be easily incorporated in the PINN method and significantly improve (by more than 50\% in the considered cases) the accuracy of the PINN solution of the backward ADE.

He, Qizhi↗