Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Model classification”

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 1,387 records · Page 77
All sourcesNASA NTRSDOE OSTI
NASA NTRS2004

Optical Calibration Process Developed for Neural-Network-Based Optical Nondestructive Evaluation Method

A completely optical calibration process has been developed at Glenn for calibrating a neural-network-based nondestructive evaluation (NDE) method. The NDE method itself detects very small changes in the characteristic patterns or vibration mode shapes of vibrating structures as discussed in many references. The mode shapes or characteristic patterns are recorded using television or electronic holography and change when a structure experiences, for example, cracking, debonds, or variations in fastener properties. An artificial neural network can be trained to be very sensitive to changes in the mode shapes, but quantifying or calibrating that sensitivity in a consistent, meaningful, and deliverable manner has been challenging. The standard calibration approach has been difficult to implement, where the response to damage of the trained neural network is compared with the responses of vibration-measurement sensors. In particular, the vibration-measurement sensors are intrusive, insufficiently sensitive, and not numerous enough. In response to these difficulties, a completely optical alternative to the standard calibration approach was proposed and tested successfully. Specifically, the vibration mode to be monitored for structural damage was intentionally contaminated with known amounts of another mode, and the response of the trained neural network was measured as a function of the peak-to-peak amplitude of the contaminating mode. The neural network calibration technique essentially uses the vibration mode shapes of the undamaged structure as standards against which the changed mode shapes are compared. The published response of the network can be made nearly independent of the contaminating mode, if enough vibration modes are used to train the net. The sensitivity of the neural network can be adjusted for the environment in which the test is to be conducted. The response of a neural network trained with measured vibration patterns for use on a vibration isolation table in the presence of various sources of laboratory noise is shown. The output of the neural network is called the degradable classification index. The curve was generated by a simultaneous comparison of means, and it shows a peak-to-peak sensitivity of about 100 nm. The following graph uses model generated data from a compressor blade to show that much higher sensitivities are possible when the environment can be controlled better. The peak-to-peak sensitivity here is about 20 nm. The training procedure was modified for the second graph, and the data were subjected to an intensity-dependent transformation called folding. All the measurements for this approach to calibration were optical. The peak-to-peak amplitudes of the vibration modes were measured using heterodyne interferometry, and the modes themselves were recorded using television (electronic) holography.

Decker, Arthur J.↗
NASA NTRS2013

An Approach for the Assessment of System Upset Resilience

This report describes an approach for the assessment of upset resilience that is applicable to systems in general, including safety-critical, real-time systems. For this work, resilience is defined as the ability to preserve and restore service availability and integrity under stated conditions of configuration, functional inputs and environmental conditions. To enable a quantitative approach, we define novel system service degradation metrics and propose a new mathematical definition of resilience. These behavioral-level metrics are based on the fundamental service classification criteria of correctness, detectability, symmetry and persistence. This approach consists of a Monte-Carlo-based stimulus injection experiment, on a physical implementation or an error-propagation model of a system, to generate a system response set that can be characterized in terms of dimensional error metrics and integrated to form an overall measure of resilience. We expect this approach to be helpful in gaining insight into the error containment and repair capabilities of systems for a wide range of conditions.

Torres-Pomales, Wilfredo↗
NASA NTRS2014

Search for Cosmic-Ray-Induced Gamma-Ray Emission in Galaxy Clusters

Current theories predict relativistic hadronic particle populations in clusters of galaxies in addition to the already observed relativistic leptons. In these scenarios hadronic interactions give rise to neutral pions which decay into gamma rays that are potentially observable with the Large Area Telescope (LAT) on board the Fermi space telescope. We present a joint likelihood analysis searching for spatially extended gamma-ray emission at the locations of 50 galaxy clusters in four years of Fermi-LAT data under the assumption of the universal cosmic-ray (CR) model proposed by Pinzke & Pfrommer. We find an excess at a significance of 2.7 delta, which upon closer inspection, however, is correlated to individual excess emission toward three galaxy clusters: A400, A1367, and A3112. We discuss these cases in detail and conservatively attribute the emission to unmodeled background systems (for example, radio galaxies within the clusters).Through the combined analysis of 50 clusters, we exclude hadronic injection efficiencies in simple hadronic models above 21% and establish limits on the CR to thermal pressure ratio within the virial radius, R(sub 200), to be below 1.25%-1.4% depending on the morphological classification. In addition, we derive new limits on the gamma-ray flux from individual clusters in our sample.

Online-only material: color figures↗
NASA NTRS2021

Linear and Nonlinear Trend Analyzes in Global Satellite‐Based Precipitation, 1998–2017

Precipitation varies spatio‐temporally in amount, intensity, and frequency. Although, much research has been conducted on analyzing precipitation patterns and variation at the global scale, trend types have still not received much attention. This study developed a new polynomial‐based model for detecting nonlinear and linear trends in a satellite precipitation product (TRMM 3B43) for the 1998–2017 period at a near‐global scale. We used an automated trend classification method that detects significant trends and classifies them into linear and nonlinear (cubic, quadratic, and concealed) trend types in satellite‐based precipitation at near‐global, continental, and climate zone scales. We found that 12.3% of pixel‐based precipitation time series across the globe have significant trend at 0.05 significance level (50% positive and 50% negative trends). In all continents except Asia, decreasing trends were found to cover larger areas than corresponding increasing trends. Regarding climate zone and precipitation trend change, our results indicate that a linear trend is dominant in the warm temperate (77.7%) and equatorial climates (80.4%) while the least linear change was detected in the polar climate (68.9%). The combined results of continental and climate zone scales indicate significant increasing trends in Asia and arid climate over the last 20 yr. Furthermore, positive trends were found to be more significant at the continental scale, particularly, in Asia relative to the climate zone scale. Linear change in precipitation (80%) was the most dominant trend observed as opposed to nonlinear (quadratic [11%] and cubic [9%]) trend types at the global scale.

Majid Kazemzadeh↗
DOE OSTI2025

Neural network-based classification and regression of magnetohydrodynamic modes in tokamaks

We present a machine learning-based magnetohydrodynamic (MHD) classifier and regressor that utilizes real or complex-valued 3D magnetic sensor array data to determine neoclassical tearing mode (NTM) onset times in tokamaks with millisecond accuracy. The input dataset consists of poloidal profiles of complex Fourier amplitudes with an n = 1 toroidal mode number from 144 human-labeled ITER Baseline Scenario discharges in the DIII-D tokamak, spanning both tearing-dominated and sawtooth-dominated regimes. Since m, n = 2,1 NTMs frequently emerge alongside sawteeth at the same frequency in this scenario, the focus is on isolating the m = 1 and m = 2 components of the n = 1 MHD mode near the tearing onset. To improve model regularization and prediction stability, singular value decomposition was applied to balance the sawtooth and tearing datasets. The enriched datasets facilitated training neural networks that learn the key distinguishing features of sawtooth and tearing modes in the poloidal profiles of their magnetic amplitude and phase. When the modes occur independently, the networks achieve perfect classification due to the modes’ distinct characteristics and low measurement noise. In the more experimentally relevant case where both modes coexist, the networks maintain exceptional performance across key metrics. Tests on synthetic data with known ground truth demonstrate the superior accuracy of the neural network trained on complex-valued input compared to models using real amplitude, phase, or pseudo-complex data, achieving both a mean time delay and standard deviation below 1 ms. Notably, standard linear regression methods fitting the dominant singular modes to the data closely match the neural network’s performance. Applying these methods across a broad range of H-mode scenarios will enable future studies to systematically identify dominant NTM triggers as scenario-specific variables, paving the way for more effective tearing mode avoidance strategies in future fusion reactor designs.

machine learning↗
DOE OSTI2023

Digital Modeling on Large Kernel Metamaterial Neural Network

Deep neural networks (DNNs) utilized recently are physically deployed with computational units (e.g., CPUs and GPUs). Such a design might lead to a heavy computational burden, significant latency, and intensive power consumption, which are critical limitations in applications such as Internet of Things (IoT), edge computing, and usage of drones. Recent advances in optical computational units (e.g., metamaterial) have shed light on energy-free and light-speed neural networks. However, the digital design of the metamaterial neural network (MNN) is fundamentally limited by its physical limitations, such as precision, noise, and bandwidth during fabrication. Moreover, the unique advantages of MNN’s (e.g., light-speed computation) are not fully explored via standard 3×3 convolution kernels. In this paper, we propose a novel large kernel metamaterial neural network (LMNN) that maximizes the digital capacity of the state-of-the-art (SOTA) MNN with model re-parametrization and network compression, while also considering the optical limitation explicitly. The new digital learning scheme can maximize the learning capacity of MNN while modeling the physical restrictions of meta-optics. With the proposed LMNN, the computation cost of the convolutional front-end can be offloaded to fabricated optical hardware. The experimental results on two publicly available datasets demonstrate that the optimized hybrid design improved classification accuracy while reducing computational latency. In conclusion, the development of the proposed LMNN is a promising step towards the ultimate goal of energy-free and light-speed AI.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2014

Introducing Multisensor Satellite Radiance-Based Evaluation for Regional Earth System Modeling

Earth System modeling has become more complex, and its evaluation using satellite data has also become more difficult due to model and data diversity. Therefore, the fundamental methodology of using satellite direct measurements with instrumental simulators should be addressed especially for modeling community members lacking a solid background of radiative transfer and scattering theory. This manuscript introduces principles of multisatellite, multisensor radiance-based evaluation methods for a fully coupled regional Earth System model: NASA-Unified Weather Research and Forecasting (NU-WRF) model. We use a NU-WRF case study simulation over West Africa as an example of evaluating aerosol-cloud-precipitation-land processes with various satellite observations. NU-WRF-simulated geophysical parameters are converted to the satellite-observable raw radiance and backscatter under nearly consistent physics assumptions via the multisensor satellite simulator, the Goddard Satellite Data Simulator Unit. We present varied examples of simple yet robust methods that characterize forecast errors and model physics biases through the spatial and statistical interpretation of various satellite raw signals: infrared brightness temperature (Tb) for surface skin temperature and cloud top temperature, microwave Tb for precipitation ice and surface flooding, and radar and lidar backscatter for aerosol-cloud profiling simultaneously. Because raw satellite signals integrate many sources of geophysical information, we demonstrate user-defined thresholds and a simple statistical process to facilitate evaluations, including the infrared-microwave-based cloud types and lidar/radar-based profile classifications.

Planetary Boundary Layer↗
DOE OSTI2020

Loops and trees in generic EFTs

We consider aspects of tree and one-loop behavior in a generic 4d EFT of massless scalars, fermions, and vectors, with a particular eye to the high-energy limit of the Standard Model EFT at operator dimensions 6 and 8. First, we classify the possible Lorentz structures of operators and the subset of these that can arise at tree-level in a weakly coupled UV completion, extending the tree/loop classification through dimension 8 using functional methods. Second, we investigate how operators contribute to tree and one-loop helicity amplitudes, exploring the impact of non-renormalization theorems through dimension 8. We further observe that many dimension 6 contributions to helicity amplitudes, including rational parts, vanish exactly at one-loop level. This suggests the impact of helicity selection rules extends beyond one loop in non-supersymmetric EFTs.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

One-shot gas detection with transformer paired neural networks in Mako collected longwave infrared hyperspectral imagery

To date, careful data treatment workflows and statistical detectors are used to perform hyperspectral image (HSI) detection of any gas contained in a spectral library, which is often expanded with physics models to incorporate different spectral characteristics. In general, surrounding evidence or known gas-release parameters are used to provide confidence in or confirm detection capability, respectively. This makes quantifying detection performance difficult as it is nearly impossible to develop an absolute ground truth for gas target pixel presence in collected HSI. Consequently, developing and comparing new detection methods, especially machine learning (ML)-based methods, is susceptible to subjectivity in derived detection map quality. Here, in this work, we demonstrate the first use of transformer-based paired neural networks (PNNs) for one-shot gas target detection for multiple gases while providing quantitative classification and detection metrics for their use on labeled data. Terabytes of training data are generated from a database of long-wave infrared HSI obtained from historical Mako sensor campaigns over Los Angeles. By incorporating labels, singular signature representations, and a model development pipeline, we can tune and select PNNs to detect multiple gas targets that are not seen in training on a quantitative basis. We additionally assess our test set detections using interpretability techniques widely employed with ML-based predictors, but less common with detection methods relying on learned latent spaces.

Hyperspectral imaging↗
DOE OSTI2023

Life history strategies among soil bacteria—dichotomy for few, continuum for many

Abstract Study of life history strategies may help predict the performance of microorganisms in nature by organizing the complexity of microbial communities into groups of organisms with similar strategies. Here, we tested the extent that one common application of life history theory, the copiotroph-oligotroph framework, could predict the relative population growth rate of bacterial taxa in soils from four different ecosystems. We measured the change of in situ relative growth rate to added glucose and ammonium using both 18O–H2O and 13C quantitative stable isotope probing to test whether bacterial taxa sorted into copiotrophic and oligotrophic groups. We saw considerable overlap in nutrient responses across most bacteria regardless of phyla, with many taxa growing slowly and few taxa that grew quickly. To define plausible life history boundaries based on in situ relative growth rates, we applied Gaussian mixture models to organisms’ joint 18O–13C signatures and found that across experimental replicates, few taxa could consistently be assigned as copiotrophs, despite their potential for fast growth. When life history classifications were assigned based on average relative growth rate at varying taxonomic levels, finer resolutions (e.g., genus level) were significantly more effective in capturing changes in nutrient response than broad taxonomic resolution (e.g., phylum level). Our results demonstrate the difficulty in generalizing bacterial life history strategies to broad lineages, and even to single organisms across a range of soils and experimental conditions. We conclude that there is a continued need for the direct measurement of microbial communities in soil to advance ecologically realistic frameworks.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Effects of Substance Use and Antisocial Personality on Neuroimaging-Based Machine Learning Prediction of Schizophrenia

Abstract Background and hypothesis Neuroimaging-based machine learning (ML) algorithms have the potential to aid the clinical diagnosis of schizophrenia. However, literature on the effect of prevalent comorbidities such as substance use disorder (SUD) and antisocial personality (ASPD) on these models’ performance has remained unexplored. We investigated whether the presence of SUD or ASPD affects the performance of neuroimaging-based ML models trained to discern patients with schizophrenia (SCH) from controls. Study design We trained an ML model on structural MRI data from public datasets to distinguish between SCH and controls (SCH = 347, controls = 341). We then investigated the model’s performance in two independent samples of individuals undergoing forensic psychiatric examination: sample 1 was used for sensitivity analysis to discern ASPD (N = 52) from SCH (N = 66), and sample 2 was used for specificity analysis to discern ASPD (N = 26) from controls (N = 25). Both samples included individuals with SUD. Study results In sample 1, 94.4% of SCH with comorbid ASPD and SUD were classified as SCH, followed by patients with SCH + SUD (78.8% classified as SCH) and patients with SCH (60.0% classified as SCH). The model failed to discern SCH without comorbidities from ASPD + SUD (AUC = 0.562, 95%CI = 0.400–0.723). In sample 2, the model’s specificity to predict controls was 84.0%. In both samples, about half of the ASPD + SUD were misclassified as SCH. Data-driven functional characterization revealed associations between the classification as SCH and cognition-related brain regions. Conclusion Altogether, ASPD and SUD appear to have effects on ML prediction performance, which potentially results from converging cognition-related brain abnormalities between SCH, ASPD, and SUD.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2020

How Do Visual Explanations Foster End Users' Appropriate Trust in Machine Learning?

We investigated the effects of different visual explanations on users' trust in machine learning classification. We proposed three forms of visual explanations of a classification based on identifying relevant training instances. We conducted a user study to evaluate these visual explanations as well as a no explanation condition. We measured users' trust of a classifier, quantified the effects of these three forms of explanations, and assessed the changes in users' trust. We found that participants trust a classifier appropriately when an explanation is available. The combination of human, classification algorithm and understandable explanation makes better decisions than the classifier and human alone.This work advances the state-of-the-art closer to building trust-able machine learning models and informs the design and appropriate use of automated systems.

Yang, Fumeng↗
DOE OSTI2023

Discriminative analysis of schizophrenia patients using graph convolutional networks: A combined multimodal MRI and connectomics analysis

Introduction Recent studies in human brain connectomics with multimodal magnetic resonance imaging (MRI) data have widely reported abnormalities in brain structure, function and connectivity associated with schizophrenia (SZ). However, most previous discriminative studies of SZ patients were based on MRI features of brain regions, ignoring the complex relationships within brain networks. Methods We applied a graph convolutional network (GCN) to discriminating SZ patients using the features of brain region and connectivity derived from a combined multimodal MRI and connectomics analysis. Structural magnetic resonance imaging (sMRI) and resting-state functional magnetic resonance imaging (rs-fMRI) data were acquired from 140 SZ patients and 205 normal controls. Eighteen types of brain graphs were constructed for each subject using 3 types of node features, 3 types of edge features, and 2 brain atlases. We investigated the performance of 18 brain graphs and used the TopK pooling layers to highlight salient brain regions (nodes in the graph). Results The GCN model, which used functional connectivity as edge features and multimodal features (sMRI + fMRI) of brain regions as node features, obtained the highest average accuracy of 95.8%, and outperformed other existing classification studies in SZ patients. In the explainability analysis, we reported that the top 10 salient brain regions, predominantly distributed in the prefrontal and occipital cortices, were mainly involved in the systems of emotion and visual processing. Discussion Our findings demonstrated that GCN with a combined multimodal MRI and connectomics analysis can effectively improve the classification of SZ at an individual level, indicating a promising direction for the diagnosis of SZ patients. The code is available at https://github.com/CXY-scut/GCN-SZ.git .

Chen, Xiaoyi↗
DOE OSTI2022

Deep Learning for In-Situ Layer Quality Monitoring during Laser-Based Directed Energy Deposition (LB-DED) Additive Manufacturing Process

Defects are a leading issue for the rejection of parts manufactured through the Directed Energy Deposition (DED) Additive Manufacturing (AM) process. In an attempt to illuminate and advance in situ quality monitoring and control of workpieces, we present an innovative data-driven method that synchronously collects sensing data and AM process parameters with a low sampling rate during the DED process. The proposed data-driven technique determines the important influences that individual printing parameters and sensing features have on prediction at the inter-layer qualification to perform feature selection. Three Machine Learning (ML) algorithms including Random Forest (RF), Support Vector Machine (SVM), and Convolutional Neural Network (CNN) are used. During post-production, a threshold is applied to detect low-density occurrences such as porosity sizes and quantities from CT scans that render individual layers acceptable or unacceptable. This information is fed to the ML models for training. Training/testing are completed offline on samples deemed “high-quality” and “low-quality”, utilizing only features recorded from the build process. CNN results show that the classification of acceptable/unacceptable layers can reach between 90% accuracy while training/testing on a “high-quality” sample and dip to 65% accuracy when trained/tested on “low-quality”/“high-quality” (respectively), indicating over-fitting but showing CNN as a promising inter-layer classifier.

36 MATERIALS SCIENCE↗
NASA NTRS1983

Integration of Landsat data into the Saginaw River Basin geographic information system

One of the objectives of the Saginaw River Basin study is related to the development of a computer model to predict flood damages. The computer model is to be operational in June 1985. In order to achieve this objective, the input of land-use data into a data base consisting of 198,000 grid cells will be required. A planning technique using Spatial Analysis Methodology (SAM) was developed by the Corps of Engineers Hydrologic Engineering Center (HEC) to systematically handle these data. The HEC-SAM system uses the spatially oriented map data in a series of data management and analysis software programs for input to the Corps hydrologic and environmental models. Attention is given to data base development, Landsat digital data, the placement of the Landsat data into the grid cell data base, and the development of the land cover classification. The Landsat-2 MSS scene covering 85 percent of the Saginaw River Basin was geometrically corrected to a UTM coordinate system.

Mckim, H. L.↗
NASA NTRS2000

Bayesian Fusion of Color and Texture Segmentations

In many applications one would like to use information from both color and texture features in order to segment an image. We propose a novel technique to combine "soft" segmentations computed for two or more features independently. Our algorithm merges models according to a mean entropy criterion, and allows to choose the appropriate number of classes for the final grouping. This technique also allows to improve the quality of supervised classification based on one feature (e.g. color) by merging information from unsupervised segmentation based on another feature (e.g., texture.)

Manduchi, Roberto↗
NASA NTRS2016

How We Can Constrain Aerosol Type Globally

In addition to aerosol number concentration, aerosol size and composition are essential attributes needed to adequately represent aerosol-cloud interactions (ACI) in models. As the nature of ACI varies enormously with environmental conditions, global-scale constraints on particle properties are indicated. And although advanced satellite remote-sensing instruments can provide categorical aerosol-type classification globally, detailed particle microphysical properties are unobtainable from space with currently available or planned technologies. For the foreseeable future, only in situ measurements can constrain particle properties at the level-of-detail required for ACI, as well as to reduce uncertainties in regional-to-global-scale direct aerosol radiative forcing (DARF). The limitation of in situ measurements for this application is sampling. However, there is a simplifying factor: for a given aerosol source, in a given season, particle microphysical properties tend to be repeatable, even if the amount varies from day-to-day and year-to-year, because the physical nature of the particles is determined primarily by the regional environment. So, if the PDFs of particle properties from major aerosol sources can be adequately characterized, they can be used to add the missing microphysical detail the better sampled satellite aerosol-type maps. This calls for Systematic Aircraft Measurements to Characterize Aerosol Air Masses (SAM-CAAM). We are defining a relatively modest and readily deployable, operational aircraft payload capable of measuring key aerosol absorption, scattering, and chemical properties in situ, and a program for characterizing statistically these properties for the major aerosol air mass types, at a level-of-detail unobtainable from space. It is aimed at: (1) enhancing satellite aerosol-type retrieval products with better aerosol climatology assumptions, and (2) improving the translation between satellite-retrieved aerosol optical properties and species-specific aerosol mass and size that are tracked in aerosol transport and climate models used to assess aerosol forcing and its anthropogenic components. This presentation will summarize where we stand on the development of the SAM-CAAM concept.

DARF↗