Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “model inference”

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 1,711 records · Page 95

Early Alarm: Robust Event Analysis for Power Systems using 1-D Fully Convolutional Network

This work presents a novel deep learning model for early, accurate, and robust detection, recognition, and temporal localization of multi-type events in large-scale power systems. The proposed method develops a unified 1-D fully convolutional network (FCN) model that takes time series of raw frequency signals measured from a power system as input, extracts distinguishing features, and predicts at every temporal point in the time series if an event is happening and what the type of the event is. Compared to existing methods, the proposed model eliminates the necessity for hand-crafted feature extraction or complicated data pre-processing, can flexibly handle input signals of arbitrary length, and precisely infer the event occurrence time. Most importantly, the model is intentionally trained with incomplete patterns, such that it is more robust to partial features of an event which is common in real-world online recognition, resulting in early alarm for power system failures. Extensive experimental results demonstrate that the proposed method achieves superior performance to the state-of-the-art, and also shows strong robustness to noise and system oscillations.

Li, Chengcheng↗

Applications of a Venus thermospheric circulation model

A variety of Pioneer Venus observations suggest a global scale, day-to-night Venus thermospheric circulation. Model studies of the dynamics and energetics of the Venus thermosphere are presented in order to address new driving, mixing and cooling mechanisms for an improved model simulation. The adopted approach was to reexamine the circulation by first using a previous two dimensional code to quantify those physical processes which can be inferred from the Pioneer Venus observations. Specifically, the model was used to perform sensitivity studies to determine the degree to which eddy cooling, eddy or wave drag, eddy diffusion and 15 micrometer radiational cooling are necessary to bring the model temperature and composition fields into agreement with observations. Three EUV heating cases were isolated for study. Global temperature and composition fields in good agreement with Pioneer data were obtained. Large scale horizontal winds 220 m/s were found to be consistent with the observed cold nightside temperatures and dayside bulges of O, CO and CO2. Observed dayside temperatures were obtained by using a 7 to 19% EUV heating efficiency profile. The enhanced 15 micrometer cooling needed for thermal balance is obtained using the best rate coefficient available for atomic O collisional excitation of CO2(0,1,0). Eddy conduction was not found to be a viable cooling mechanism due to the weakened global circulation. The strong 15 micrometer damping and low EUV efficiency imply a very weak dependence of the general circulation to solar cycle variability. The NCAR terrestrial thermospheric general circulation model was adapted for Venus inputs using the above two dimensional model parameters, to give a three dimensional benchmark for future Venus modelling work.

Bougher, S. W.↗

The Crab Nebula and related supernova remnants; Proceedings of the Workshop, George Mason University, Fairfax, VA, October 11, 12, 1984

Papers are presented on the Crab Nebula's composition, helium distribution, outer structure and jet, and evolution. Attention is given to line emission from supernova remnants and charge transfer reactions, a magnetohydrodynamic model of the Crab Nebula and its radiation, inferences made using data on the pulsed flux from the crab pulsar, a new interpretation of the crab pulsar X-ray interpulse radiation, and evolutionary models of the Crab Nebula's progenitor. Other topics include the evolution of the centimeter flux of 3C58 and the Crab Nebula, a search for a shock wave around the Crab Nebula, high resolution radio studies of the Crab Nebula, supernova shell structure, and the nature of the remnant 0540-693 and its implications for the study of crablike remnants. Papers are also presented on X-ray observations of: Crab-like remnants, the Crab Nebula, the Vela X region, W28, and 3C400.2. Other papers include the 50 millisecond pulsar in the Large Magellanic Cloud and the X-ray pulse emission mechanism, optical emission from the plerionic core of CTB 80, and one-arcminute resolution observations of W50.

Kafatos, M. C.↗

Asteroid families - Physical properties and evolution

Asteroid families are considered to be fragments from collisional destruction of precursor bodies. However, results available on the inferred mineralogy, size distributions, and spins of family members do not confirm the expectations of the traditional model. Only a handful of nearly 100 proposed families, most of them populous, have distributions of inferred mineralogies consistent with simple cosmochemical models for parent bodies. It is suggested that most catastrophic collisions may not result in observable families, but rather in a spray of smaller particles, thus accounting for the small number of confirmed and consistent families, despite evidence for extensive collisional evolution of asteroids.

Chapman, Clark R.↗

AK112: Full Waveform Inversion Tomography of Alaska Improves Waveform Fits While Imaging Crustal, Mantle, and Slab Structure

We report a full waveform inversion tomography model of Alaska and the surrounding regions, inferring radially anisotropic shear and isotropic compressional wavespeeds by fitting complete waveforms from 120 regional earthquakes. Our multiscale approach inverted time–frequency phase misfits (maximum period of 100 s), starting with a minimum period of 40 s and ending at 20 s in 7 stages and 112 total iterations. The model (AK112) was evaluated by computing the misfits for 36 independent validation events. We find that misfit reductions were large and equal (∼55%) for both the inversion and validation data sets, providing confidence in the model. AK112 also provides much better waveform fits compared to other reported models for the region, including an isotropic version of itself, highlighting the importance of anisotropy. The model resolves known crustal, upper mantle, and slab structure to depths of 100 km with new detail: sedimentary basins in the Alaskan Shelf, Cook Inlet, and Colville basins, among others; discontinuous lithospheric structure across major terrane boundaries; and subducting slab geometry and back‐arc volcanic sources. In addition to tectonic interpretations, the model enables full waveform simulations for long‐period earthquake ground motions and source characterization (e.g., moment tensor and finite‐fault inversion).

Rodgers, Arthur [Lawrence Livermore National Labor↗

Predicting Lightning Initiation using Deep Learning

Lightning occurrence presents safety challenges to people and property. The main challenge with lightning safety is that the majority of guidance is reactive. In other words, lightning has to have already occurred nearby before a person will respond and take shelter. Further, most injuries or fatalities occur as the storm approaches, or as it's moving away, when rainfall may not be present at the time of the flash. Thus, this project develops a physically-based deep learning model to produce lightning probabilities out to 15 minutes. The deep learning model combines a Convolutional Neural Network (CNN) with a Long Short-Term Memory (LSTM) network to capture both the spatial and temporal evolution of storms to predict the probability that lightning initiation will occur in the next 15 minutes. The model combines radar reflectivity, correlation coefficient and differential reflectivity to inferred storm hydrometer type and precipitation phase, which aids in the identification of electrification processes. The model is trained with data from the Geostationary Lightning Mapper (GLM), which is a near infrared sensor onboard the GOES-R series of satellites that measures optical brightness from lightning. This presentation will provide an overview of the project.

Andrew T White↗

Parsimonious models of in-host viral dynamics and immune response

Mathematical models of in-host viral dynamics and immune response are a vital tool for patient-specific estimation of the initial viral load, prediction of the course of an infection, etc. The COVID-19 pandemics has given impetus to the development of models with an ever-increasing degree of complexity. We show that one of the most popular models---the Target Cell Limited model---fails the identifiability test, i.e., its parameters cannot be uniquely inferred from readily available data such as viral load measurements. Here, we present a model that is both identifiable and parsimonious according to information criteria. Our model's predictions match both reported observations of COVID-19 patients and predictions of its more complex counterparts.

60 APPLIED LIFE SCIENCES↗

CAHS: Context-Aware Homology Search

Protein homology search is foundational to bioinformatics: it supports annotation transfer, structure/function inference, and evolutionary analysis over rapidly expanding sequence repositories (e.g., UniProtKB). Profile hidden Markov models (pHMMs), as implemented in HMMER, remain the most widely trusted approach because they provide statistically calibrated E-values; however, their gap behavior is fixed once a profile is trained, despite biological evidence that insertion/deletion tolerance varies across flexible loops and intrinsically disordered regions. We present CAHS (Context-Aware Homology Search), a lightweight query-time adapter for pHMM search that incorporates learned and biologically motivated signals without changing HMMER's downstream search pipeline or its calibrated E-value reporting. Given a query sequence, CAHS computes per-residue representations from a protein language model and a disorder predictor, maps these to profile coordinates, and modulates only match-state transition rows (gap-open and gap-extension probabilities) while preserving Plan7 constraints. We comprehensively evaluate CAHS across six structurally diverse protein families and multi-domain architectures against a 570k-sequence target corpus. CAHS expands detection capability, retrieving thousands of additional remote homologs at relaxed thresholds by maintaining alignment quality through flexible regions. For multi-domain proteins, context-aware modulation resolves 94% of fragmented alignments. Crucially, CAHS preserves hit-set invariance at stringent operating points (E<10-10), demonstrating increased statistical confidence without inflating false positives. Furthermore, sharper statistical distinction between homologs and background noise during early filter stages yields up to a 3.87× acceleration in end-to-end wall-clock time on high-performance computing clusters. Overall, CAHS illustrates a practical AI-for-science design pattern: augmenting a trusted probabilistic model with query-specific learned signals to improve interpretable, reproducible inference in data-rich biology.

Bhattaram, Swethasree [Georgia Institute of Techno↗

Electrostatic emissions between electron gyroharmonics in the outer magnetosphere

Electrostatic emissions are often observed by spacecraft in the outer magnetosphere at frequencies between low electron gyroharmonics. The emissions include the well-known odd half-harmonic emissions as well as weaker emissions which occur near the electron plasma frequency. A scheme is constructed for classifying the emissions, and it is shown that a theoretical model which has been previously used to explain '3/2' emissions can be extended to the other classes of emissions. All of the emissions therefore appear to be generated by the same basic mechanism: an unstable electron plasma distribution consisting of cold electrons (below 100 eV) and hot loss cone electrons (about 1 keV). Each emission class is associated with a particular range of model parameters; the wide-band electric field data can thus be used to infer the density and temperature of the cold plasma component. The model predicts that gyroharmonic emissions near the plasma frequency require large cold plasma densities.

Hubbard, R. F.↗

An evidential reasoning extension to quantitative model-based failure diagnosis

The detection and diagnosis of failures in physical systems characterized by continuous-time operation are studied. A quantitative diagnostic methodology has been developed that utilizes the mathematical model of the physical system. On the basis of the latter, diagnostic models are derived each of which comprises a set of orthogonal parity equations. To improve the robustness of the algorithm, several models may be used in parallel, providing potentially incomplete and/or conflicting inferences. Dempster's rule of combination is used to integrate evidence from the different models. The basic probability measures are assigned utilizing quantitative information extracted from the mathematical model and from online computation performed therewith.

Gertler, Janos J.↗

An experimental and theoretical study of the ice accretion process during artificial and natural icing conditions

Real-time measurements of ice growth during artificial and natural icing conditions were conducted using an ultrasonic pulse-echo technique. This technique allows ice thickness to be measured with an accuracy of + or - 0.5 mm; in addition, the ultrasonic signal characteristics may be used to detect the presence of liquid on the ice surface and hence discern wet and dry ice growth behavior. Ice growth was measured on the stagnation line of a cylinder exposed to artificial icing conditions in the NASA Lewis Icing Research Tunnel (IRT), and similarly for a cylinder exposed in flight to natural icing conditions. Ice thickness was observed to increase approximately linearly with exposure time during the initial icing period. The ice accretion rate was found to vary with cloud temperature during wet ice growth, and liquid runback from the stagnation region was inferred. A steady-state energy balance model for the icing surface was used to compare heat transfer characteristics for IRT and natural icing conditions. Ultrasonic measurements of wet and dry ice growth observed in the IRT and in flight were compared with icing regimes predicted by a series of heat transfer coefficients. The heat transfer magnitude was generally inferred to be higher for the IRT than for the natural icing conditions encountered in flight. An apparent variation in the heat transfer magnitude was also observed for flights conducted through different natural icing-cloud formations.

Kirby, Mark S.↗

Cloud drop number concentrations over the western North Atlantic Ocean: seasonal cycle, aerosol interrelationships, and other influential factors

Cloud drop number concentrations (N d ) over the western North Atlantic Ocean (WNAO) are generally highest during the winter (DJF) and lowest in summer (JJA), in contrast to aerosol proxy variables (aerosoloptical depth, aerosol index, surface aerosol mass concentrations, surface cloud condensation nuclei (CCN) concentrations) that generally peak inspring (MAM) and JJA with minima in DJF. Using aircraft, satellite remote sensing, ground-based in situ measurement data, and reanalysis data, we characterize factors explaining the divergent seasonal cycles and furthermore probe into factors influencing N d on seasonal timescales. The results can be summarized well by features most pronounced in DJF, including features associated with cold-air outbreak (CAO) conditions such as enhanced values of CAO index, planetary boundary layer height (PBLH),low-level liquid cloud fraction, and cloud-top height, in addition to winds aligned with continental outflow. Data sorted into high- and low-N d days in each season, especially in DJF, revealed that all of these conditions were enhanced on the high-N d days, including reduced sea level pressure and stronger wind speeds. Although aerosols may be more abundant in MAM and JJA, the conditions needed to activate those particles into cloud droplets are weaker than in colder months, which is demonstrated by calculations of the strongest (weakest) aerosol indirect effects in DJF (JJA) based on comparing N d to perturbations in four different aerosol proxy variables (total and sulfate aerosol optical depth, aerosol index, surface mass concentration of sulfate). We used three machine learning models and up to 14 input variables to infer about most influential factors related to N d for DJF and JJA, with the best performance obtained with gradient-boosted regression tree (GBRT) analysis. The model results indicated that cloud fraction was the most important input variable, followed by some combination (depending on season) of CAO index and surface mass concentrations of sulfate and organic carbon. Future work is recommended to further understand aspects uncovered here such as impacts of free tropospheric aerosol entrainment on clouds, degree of boundary layer coupling, wet scavenging, and giant CCN effects on aerosol–N d relationships, updraft velocity, and vertical structure of cloud properties such as adiabaticity that impact the satellite estimation of N d .

54 ENVIRONMENTAL SCIENCES↗

Detailed Modeling and Analysis of the CPFM Dataset

A quantitative understanding of photolysis rate coefficients (or "j-values") is essential to determining the photochemical reaction rates that define ozone loss and other crucial processes in the atmosphere. j-Values can be calculated with radiative transfer models, derived from actinic flux observations, or inferred from trace gas measurements. The principal objective of this study is to cross-validate j-values from the Composition and Photodissociative Flux Measurement (CPFM) instrument during the Photochemistry of Ozone Loss in the Arctic Region In Summer (POLARIS) and SAGE I11 Ozone Loss and Validation Experiment (SOLVE) field campaigns with model calculations and other measurements and to use this detailed analysis to improve our ability to determine j-values. Another objective is to analyze the spectral flux from the CPFM (not just the j-values) and, using a multi-wavelength/multi-species spectral fitting technique, determine atmospheric composition.

Swartz, William H.↗

Intra-seasonal Oscillations Inferred from SABER (TIMED) and MLS (UARS) Temperature Measurements

In the zonal mean meridional winds of the upper mesosphere, intra-seasonal oscillations with periods between 1 and 4 months have been inferred from UARS measurements and independently predicted with the Numerical Spectral Model WSM). The wind oscillations tend to be confined to low latitudes and appear to be driven, at least in part, by small-scale gravity waves propagating in the meridional direction. Winds across the equator should generate, due to dynamical heating and cooling, temperature oscillations with opposite phase in the two hemispheres. Investigating this phenomenon, we have analyzed SABER temperatures from TIMED in the altitude range between 55 and 95 km to delineate with an empirical model, the year-long variability of the migrating tides and zonal mean components. The inferred seasonal variations of the diurnal tide, characterized by amplitude maxima near equinox, are in substantial agreement with UARS observations and results from the NSM. For the zonal mean, the dominant seasonal variations in the SABER temperatures, with annual (12 months) and semiannual (6 months) periodicities, agree well with those derived from UARS measurements. The intra-seasonal variations with periods between 2 and 4 months have amplitudes close to 2 K, almost half as large as those for the dominant seasonal variations. Their amplitudes are in qualitative agreement with the corresponding values inferred from UARS during different years. The SABER and UARS temperature variations reveal pronounced hemispherical asymmetries, consistent with meridional wind oscillations across the equator. The phase of the semi-annual temperature oscillations from the NSM agrees with the observations from UARS and SABER. But the amplitudes are systematically smaller, which may indicate that planetary waves are more important than is allowed for in the model. For the shorter-period intra-seasonal variations, which can be generated by gravity wave drag, the model results are generally in better agreement with the observations.

Huang, F. T.↗

Spike-and-Slab Shrinkage Priors for Structurally Sparse Bayesian Neural Networks

Network complexity and computational efficiency have become increasingly significant aspects of deep learning. Sparse deep learning addresses these challenges by recovering a sparse representation of the underlying target function by reducing heavily overparameterized deep neural networks. Specifically, deep neural architectures compressed via structured sparsity (e.g., node sparsity) provide low-latency inference, higher data throughput, and reduced energy consumption. In this article, we explore two well-established shrinkage techniques, Lasso and Horseshoe, for model compression in Bayesian neural networks (BNNs). To this end, we propose structurally sparse BNNs, which systematically prune excessive nodes with the following: 1) spike-and-slab group Lasso (SS-GL) and 2) SS group Horseshoe (SS-GHS) priors, and develop computationally tractable variational inference, including continuous relaxation of Bernoulli variables. We establish the contraction rates of the variational posterior of our proposed models as a function of the network topology, layerwise node cardinalities, and bounds on the network weights. Furthermore, we empirically demonstrate the competitive performance of our models compared with the baseline models in prediction accuracy, model compression, and inference latency.

97 MATHEMATICS AND COMPUTING↗

Final project report

The proposed overarching goal for this project was the following: Data integration, simulation and visualization will facilitate metabolic and regulatory network prediction, exploration, and formulation of hypotheses. We stated three specific aims to achieve the overarching goal of this project: (1) Integration of multiple levels of information such as mRNA and protein levels, predicted protein-protein interactions/associations and gene function will enable construction of models describing environmental response and dynamic behavior. (2) Flexible tools for network inference will accelerate our understanding of biological systems. (3) Flexible exploration and queries of model hypotheses will provide focus and reveal novel dependencies. The underlying philosophy of these proposed aims is that an iterative cycle of experiments, experimental design, and verification will lead to a comprehensive and predictive model that will shed light on systems level mechanisms involved in responses elicited by living systems upon sensing a change in their environment. In the previous years report we demonstrated considerable progress in development of data standards, regulatory network inference and data visualization and exploration. We are pleased to report that several manuscripts describing these procedures have been published in top international peer reviewed journals including Genome Biology, PNAS, and Cell. The abstracts of these manuscripts are given and they summarize our accomplishments in this project.

59 BASIC BIOLOGICAL SCIENCES↗

Data Analysis with Graphical Models: Software Tools

Probabilistic graphical models (directed and undirected Markov fields, and combined in chain graphs) are used widely in expert systems, image processing and other areas as a framework for representing and reasoning with probabilities. They come with corresponding algorithms for performing probabilistic inference. This paper discusses an extension to these models by Spiegelhalter and Gilks, plates, used to graphically model the notion of a sample. This offers a graphical specification language for representing data analysis problems. When combined with general methods for statistical inference, this also offers a unifying framework for prototyping and/or generating data analysis algorithms from graphical specifications. This paper outlines the framework and then presents some basic tools for the task: a graphical version of the Pitman-Koopman Theorem for the exponential family, problem decomposition, and the calculation of exact Bayes factors. Other tools already developed, such as automatic differentiation, Gibbs sampling, and use of the EM algorithm, make this a broad basis for the generation of data analysis software.

Buntine, Wray L.↗

The Livingstone Model of a Main Propulsion System

Livingstone is a discrete, propositional logic-based inference engine that has been used for diagnosis of physical systems. We present a component-based model of a Main Propulsion System (MPS) and say how it is used with Livingstone (L2) in order to implement a diagnostic system for integrated vehicle health management (IVHM) for the Propulsion IVHM Technology Experiment (PITEX). We start by discussing the process of conceptualizing such a model. We describe graphical tools that facilitated the generation of the model. The model is composed of components (which map onto physical components), connections between components and constraints. A component is specified by variables, with a set of discrete, qualitative values for each variable in its local nominal and failure modes. For each mode, the model specifies the component's behavior and transitions. We describe the MPS components' nominal and fault modes and associated Livingstone variables and data structures. Given this model, and observed external commands and observations from the system, Livingstone tracks the state of the MPS over discrete time-steps by choosing trajectories that are consistent with observations. We briefly discuss how the compiled model fits into the overall PITEX architecture. Finally we summarize our modeling experience, discuss advantages and disadvantages of our approach, and suggest enhancements to the modeling process.

Bajwa, Anupa↗