Search NASA⌕ Search

SEARCH · Search NASA

Results for “data driven model”

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

Urban-scale building energy consumption database: a case study for Wuhan, China

Building energy consumption accounts for 30% of the overall energy end use worldwide. This number is even much higher in urban areas. With rapid urbanization in China, cities are expanding with new constructions. It is essential to create an updated urban-scale building energy consumption database to represent energy use for different types of buildings in China, which could help urban planners, managers and decision makers to understand temporal and spatial building energy consumption distribution and ensure required electricity and/or gas supply. However, such urban-scale database is rarely found in China. This paper creates baseline EnergyPlus models for residential, small office and large office buildings and validates the baseline models using survey data from literature. Parametric simulations are conducted to consider different design factors, such as building enclosure, lighting power density, equipment power density, HVAC schedule, etc. In total 351 EnergyPlus models are generated to cover different energy use intensity scenarios. Data-driven regression analysis is conducted to predict building energy consumption using building price/rent. The prediction results are expected to provide design decision support for urban planning and power distribution for new constructions.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Structural basis of ligand interaction with atypical chemokine receptor 3

Chemokines drive cell migration through their interactions with seven-transmembrane (7TM) chemokine receptors on cell surfaces. The atypical chemokine receptor 3 (ACKR3) binds chemokines CXCL11 and CXCL12 and signals exclusively through β-arrestin-mediated pathways, without activating canonical G-protein signalling. This receptor is upregulated in numerous cancers making it a potential drug target. Here we collected over 100 distinct structural probes from radiolytic footprinting, disulfide trapping, and mutagenesis to map the structures of ACKR3:CXCL12 and ACKR3:small-molecule complexes, including dynamic regions that proved unresolvable by X-ray crystallography in homologous receptors. The data are integrated with molecular modelling to produce complete and cohesive experimentally driven models that confirm and expand on the existing knowledge of the architecture of receptor:chemokine and receptor:small-molecule complexes. Additionally, we detected and characterized ligand-induced conformational changes in the transmembrane and intracellular regions of ACKR3 that elucidate fundamental structural elements of agonism in this atypical receptor.

59 BASIC BIOLOGICAL SCIENCES↗

Coupling physics in machine learning to predict properties of high-temperatures alloys

Abstract High-temperature alloy design requires a concurrent consideration of multiple mechanisms at different length scales. We propose a workflow that couples highly relevant physics into machine learning (ML) to predict properties of complex high-temperature alloys with an example of the 9–12 wt% Cr steels yield strength. We have incorporated synthetic alloy features that capture microstructure and phase transformations into the dataset. Identified high impact features that affect yield strength of 9Cr from correlation analysis agree well with the generally accepted strengthening mechanism. As a part of the verification process, the consistency of sub-datasets has been extensively evaluated with respect to temperature and then refined for the boundary conditions of trained ML models. The predicted yield strength of 9Cr steels using the ML models is in excellent agreement with experiments. The current approach introduces physically meaningful constraints in interrogating the trained ML models to predict properties of hypothetical alloys when applied to data-driven materials.

36 MATERIALS SCIENCE↗

Half-lives and branchings for {beta}-delayed neutron emission for neutron-rich Co-Cu isotopes in the r-process

The {beta} decays of very neutron-rich nuclides in the Co-Zn region were studied experimentally at the National Superconducting Cyclotron Laboratory using the NSCL {beta}-counting station in conjunction with the neutron detector NERO. We measured the branchings for {beta}-delayed neutron emission (P{sub n} values) for {sup 74}Co (18{+-}15%) and {sup 75-77}Ni (10{+-}2.8%, 14{+-}3.6%, and 30{+-}24%, respectively) for the first time, and remeasured the P{sub n} values of {sup 77-79}Cu, {sup 79,81}Zn, and {sup 82}Ga. For {sup 77-79}Cu and for {sup 81}Zn we obtain significantly larger P{sub n} values compared to previous work. While the new half-lives for the Ni isotopes from this experiment had been reported before, we present here in addition the first half-life measurements of {sup 75}Co (30{+-}11 ms) and {sup 80}Cu (170{sub -50}{sup +110} ms). Our results are compared with theoretical predictions, and their impact on various types of models for the astrophysical rapid neutron-capture process (r-process) is explored. We find that with our new data, the classical r-process model is better able to reproduce the A=78-80 abundance pattern inferred from the solar abundances. The new data also influence r-process models based on the neutrino-driven high-entropy winds in core collapse supernovae.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Light and Strange Vector Resonances from Lattice QCD at Physical Quark Masses

We present the first ab initio calculation at physical quark masses of scattering amplitudes describing the lightest pseudoscalar mesons interacting via the strong force in the vector channel. Using lattice quantum chromodynamics, we postdict the defining parameters for two short-lived resonances, the ρ(770) and K*(892) which manifest as complex energy poles in ππ and Kπ scattering amplitudes, respectively. The calculation proceeds by first computing the finite-volume energy spectrum of the two-hadron systems and then determining the amplitudes from the energies using the Lüscher formalism. The error budget includes a data-driven systematic error, obtained by scanning possible fit ranges and fit models to extract the spectrum from Euclidean correlators, as well as the scattering amplitudes from the latter. The final results, obtained by analytically continuing multiple parametrizations into the complex energy plane, are M ρ = 796(5)(50)MeV, Γ ρ = 192(10)(31) MeV, M K* = 893(2)(54) MeV, and Γ K = 51(2)(11) MeV, where the subscript indicates the resonance and M and Γ stand for the mass and width, respectively, and where the first bracket indicates the statistical and the second bracket the systematic uncertainty.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Fault Detection via Occupation Kernel Principal Component Analysis

Reliable operation of automatic systems is heavily dependent on the ability to detect faults in the underlying dynamics. While traditional model-based methods have been widely used for fault detection, data-driven approaches have garnered increasing attention due to their ease of deployment and minimal need for expert knowledge. In this letter, we present a novel principal component analysis (PCA) method that uses occupation kernels. Occupation kernels result in feature maps that are tailored to the measured data, have inherent noise-robustness due to the use of integration, and can utilize irregularly sampled system trajectories of variable lengths for PCA. The occupation kernel PCA method is used to develop a reconstruction error approach to fault detection and its efficacy is validated using numerical simulations.

42 ENGINEERING↗

First Measurement of σ(gg → t$\bar{t}$)/σ(p$\bar{p}$ → t$\bar{t}$)

The work presented here is the first measurement of the fraction of top quark pair production through gluon-gluon fusion. We use an integrated luminosity of 0.96 ± 0.06 fb -1 of p{bar p} collisions at √s of 1.96 TeV collected by the CDF II detector. We select t$\bar{t}$ candidates by identifying a high-p T lepton candidate, a large missing E T as evidence for a neutrino candidate and at least four high E T jets, one of which has to be identified as originating from a b quark. The challenge is to discriminate between the two production processes with the identical final state, gg → t$\bar{t}$ and q$\bar{p}$ → t$\bar{t}$. We take advantage of the fact that compared to a quark, a gluon is more likely to radiate a low momentum gluon and therefore, one expects a larger number of charged particles with low p T in a process involving more gluons. Given the large uncertainties associated with the modeling of the low p T charged particle multiplicity, a data-driven technique was employed. Using calibration data samples, we show there exists a clear correlation between the observed average number of low p T charged particles and the average number of gluons involved in the production process predicted by Monte Carlo calculations. Given the correlation, one can identify low p T charged particle multiplicity distributions associated with specific average number of gluons. The W + 0 jet sample and dijets sample with leading jet E T in the range of 80-100 GeV are used to find no-gluon and gluon-rich low p{sub T} charged particle multiplicity distributions, respectively. Using these no-gluon and gluon-rich distributions in a likelihood fit, we find the fraction of gluon-rich events in t{bar t} candidates. This fraction has contributions from the signal and background events. Taking into account these contributions and the gg → t$\bar{t}$ and q$\bar{q}$ → t$\bar{t}$ production channel acceptances, we find σ(gg → t$\bar{t}$)/σ(p$\bar{p}$ → t$\bar{t}$) = 0.07 ± 0.14(stat) ± 0.07(syst) in agreement with the theoretical predictions of 0.15 ± 0.05. This measurement is a stepping stone towards a better understanding of the production mechanism of top quark pairs and implies that at least 67% of the t$\bar{t}$ events are consistent with q{bar q} production at 95% C.L, leaving little room for non SM processes which are not similar to the q$\bar{q}$ → t$\bar{t}$ in their gluon radiation. The result confirms our current understanding of the SM high p T production mechanism and the relative gluon PDFs at relatively high Q 2 and x.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Investigating lab-scaled offshore wind aerodynamic testing failure and developing solutions for early anomaly detections

As offshore wind systems become more complex, the risk of human error or equipment malfunction increases during experimental testing. This study investigates a lab-scale incident involving a 1 : 50 scale 5 MW wind turbine, where a generator failure led to rotor overspeed and a blade–tower strike. To improve early fault detection, we propose a data-driven method based on multivariate long short-term memory (LSTM) models. High-frequency measurements are projected onto principal components, and anomalies are identified using reconstruction error and its time derivative. Two models are trained on different healthy datasets and tested using single- and multi-principal component (1PC and MPC) variations. Results show that combining both error and error derivative improves detection accuracy. The 1PC model detects faults faster, has a higher recall rate, and achieves a 43 % improvement in anomaly detection accuracy, while the MPC model yields higher precision. This approach provides a simple and effective tool for early anomaly detection in lab-scale experiments, helping to reduce the risk of future failures during the testing of new technologies.

17 WIND ENERGY↗

Contextual Compression of Large-Scale Wind Turbine Array Simulations: Preprint

Data sizes are becoming a critical issue particularly for HPC applications. We have developed a user-driven lossy wavelet-based storage model to facilitate the analysis and visualization of large-scale wind turbine array simulations. The model stores data as heterogeneous blocks of wavelet coefficients, providing high-fidelity access to user-defined data regions believed the most salient, while providing lower-fidelity access to less salient regions on a block-by-block basis. In practice, by retaining the wavelet coefficients as a function of feature saliency, we have seen data reductions in excess of 94 percent, while retaining lossless information in the turbine-wake regions most critical to analysis and providing enough (low-fidelity) contextual information in the upper atmosphere to track incoming coherent turbulent structures. Our contextual wavelet compression approach has allowed us to deliver interactive visual analysis while providing the user control over where data loss, and thus reduction in accuracy, in the analysis occurs. We argue this reduced but contexualized representation is a valid approach and encourages contextual data management.

17 WIND ENERGY↗

Contextual Compression of Large-Scale Wind Turbine Array Simulations

Data sizes are becoming a critical issue particularly for HPC applications. We have developed a user-driven lossy wavelet-based storage model to facilitate the analysis and visualization of large-scale wind turbine array simulations. The model stores data as heterogeneous blocks of wavelet coefficients, providing high-fidelity access to user-defined data regions believed the most salient, while providing lower-fidelity access to less salient regions on a block-by-block basis. In practice, by retaining the wavelet coefficients as a function of feature saliency, we have seen data reductions in excess of 94 percent, while retaining lossless information in the turbine-wake regions most critical to analysis and providing enough (low-fidelity) contextual information in the upper atmosphere to track incoming coherent turbulent structures. Our contextual wavelet compression approach has allowed us to deliver interative visual analysis while providing the user control over where data loss, and thus reduction in accuracy, in the analysis occurs. We argue this reduced but contextualized representation is a valid approach and encourages contextual data management.

97 MATHEMATICS AND COMPUTING↗

Developing a Disaster-Ready Power Grid Agent Through Geophysically-Informed Fault Event Scenarios

Management of the nation’s power grid over the coming decades will need to factor multiple climate-driven threats to the power system that can be stressful or even detrimental to its operations. During disasters, grid operators suffer from cognitive overload where the electric grid is severely impacted by the weather, yet grid operators have limited awareness of these factors. Meanwhile, the electric grid is facing an explosion of data coming from a variety of sources which can enable the operator to evaluate the risks and develop mitigation strategies against hazardous events. In this work, we processed heterogeneous environmental and power grid data to learn and model grid behavior caused by extreme weather events. In this study, we focused on two weather-driven hazards, hurricanes and wildfires, which were analysed for the electric grid of Texas. We the used this data to train a Reinforcement Learning (RL) agent by analysing and predicting future behaviour of the grid during those hazard events. All these parts are incorporated in one framework to have a geophysically informed power simulators that can be used to train RL agents.

electic grid, reinforcement learning, climate adap↗

Uncertainty quantification of a deep learning fuel property prediction model

Deep learning models are being widely used in the field of combustion. Given the black-box nature of typical neural network based models, uncertainty quantification (UQ) is critical to ensure the reliability of predictions as well as the training datasets, and for a principled quantification of noise and its various sources. Deep learning surrogate models for predicting properties of chemical compounds and mixtures have been recently shown to be promising for enabling data-driven fuel design and optimization, with the ultimate goal of improving efficiency and lowering emissions from combustion engines. In this study, UQ is performed for a multi-task deep learning model that simultaneously predicts the research octane number (RON), Motor Octane Number (MON), and Yield Sooting Index (YSI) of pure components and multicomponent blends. The deep learning model is comprised of three smaller networks: Extractor 1, Extractor 2, and Predictor, and a mixing operator. The molecular fingerprints of individual components are encoded via Extractor 1 and Extractor 2, the mixing operator generates fingerprints for mixtures/blends based on linear mixing operation, and the predictor maps the fingerprint to the target properties. Two different classes of UQ methods, Monte Carlo ensemble methods and Bayesian neural networks (BNNs), are employed for quantifying the epistemic uncertainty. Combinations of Bernoulli and Gaussian distributions with DropConnect and DropOut techniques are explored as ensemble methods. All the DropConnect, DropOut and Bayesian layers are applied to the predictor network. Aleatoric uncertainty is modeled by assuming that each data point has an independent uncertainty associated with it. The results of the UQ study are further analyzed to compare the performance of BNN and ensemble methods. Although this study is confined to UQ of fuel property prediction, the methodologies are applicable to other deep learning frameworks that are being widely used in the combustion community.

33 ADVANCED PROPULSION SYSTEMS↗

Late Holocene Cliff Retreat in Del Mar, CA, Revealed From Shore Platform 10 Be Concentrations and Numerical Modeling

Rocky coast cliff retreat presents a hazard to coastal communities and infrastructure that is potentially amplified under rising sea level conditions, among other factors. Unfortunately, constraints on retreat rates are typically limited to those derived from imagery and maps spanning the last ~100 years. Here, we use a newly developed coupled model of shore platform profile evolution and cosmogenic radionuclide production that considers the influence of relative sea level (RSL) rise, weathering, material resistance, and wave height decay to model cliff retreat over millennial timescales and its potential drivers in Del Mar, California. We demonstrate the ability to use topographic and bathymetric measurements from a narrow shore platform along with a limited data set of nine cosmogenic 10Be concentrations extending ~125 m from the cliff base to obtain modeled cliff retreat rates that steadily range from 5.0 to 12.5 cm yr -1 over the last two millennia until 100 years before present. These rates are consistent with modern retreat rates of about 2–19 cm yr -1 here. RSL rise in Southern California remained relatively constant during the late Holocene, potentially explaining the relatively stable modeled cliff retreat rate over this time. We also explore the relative influence of weathering, material resistance, and wave erosion efficacy and find that both weathering and wave-driven erosion are necessary to replicate the measured data at this location, with the latter exerting a stronger control on model acceptance, suggesting that waves may provide a possible mechanism by which RSL rise may influence coastal cliff erosion in southern California.

58 GEOSCIENCES↗

Real-Time Automated Hazard Detection Framework for Health Information Technology Systems

An increase in the reliability of Health Information Technology (HIT) will facilitate institutional trust and credibility of the systems. Here, we present an end-to-end framework for improving the reliability and performance of HIT systems. Specifically, we describe the system model, present some of the methods that drive the model, and discuss an initial implementation of two of the proposed methods using data from the Veterans Affairs HIT and Corporate Data Warehouse systems. The contributions of this work, thus, include (1) the design of a system model for monitoring and detecting hazards in HIT systems, (2) a data-driven approach for analysing the health care data warehouse, (3) analytical methods for characterising and analysing failures in HIT systems, and (4) a tool architecture for generating and reporting hazards in HIT systems. Our goal is to work towards an automated system that will help identify opportunities for improvements in HIT systems.

97 MATHEMATICS AND COMPUTING↗

Quantifying the Effects of Snowpack on Soil Thermal and Carbon Dynamics of the Arctic Terrestrial Ecosystems

Snow insulation effects modify soil and carbon dynamics in northern middle to high latitudes (45°–90°N). This study incorporates these effects by introducing a snow model into an existing soil thermal model in a biogeochemistry modeling framework, the Terrestrial Ecosystem Model. The coupled model is used to quantify snow insulation effects on carbon and soil thermal dynamics in 45°–90°N region for the historical period (2003–2010) and the future period (2017–2099) under two climate scenarios. The revised model captures the snow insulation effects and improves the estimates of soil thermal dynamics and the land freeze-thaw as well as terrestrial ecosystem carbon dynamics. Historical mean cold-season soil temperature at 5 cm depth driven with satellite-based snow data is 6.4°C warmer in comparison with the original model simulation. Frozen area in late spring is estimated to shrink mainly over eastern Siberia, in central to eastern Europe, and along southern Canada in November. During each nongrowing season in the historical period, 0.41 Pg more soil C is released due to warmer soil temperature estimated using the new model. During 2003–2010, the revised model estimates that the region accumulated 0.86 Pg less C due to weaker gross primary production, leading to a regional C loss at 0.19 PgC/year. The revised model projects that the region will lose 38–51% permafrost area by 2100 and continue to be a C source under the low-emission scenario (Representative Concentration Pathway 2.6) but to be gradually transitioning into a weak sink in the latter half of the 21st century under the high-emission scenario (Representative Concentration Pathway 8.5).

54 ENVIRONMENTAL SCIENCES↗

Advancing the Frontiers of Deep Learning for Low-Dose 3D Cone-Beam CT Reconstruction

X-ray computed tomography (CT) is an important noninvasive medical imaging modality for studying the structural details of internal organs. Image reconstruction in CT is an inverse problem of recovering an object's internal structure from the absorption profile of X-ray beams (sinogram) measured using a detector. The classical variational approach for CT reconstruction minimizes an energy functional using an appropriate iterative algorithm. Motivated by the success of deep learning (DL), researchers have begun to leverage training data and enhanced computing capabilities in recent years to produce high-fidelity reconstructed images. Nonetheless, much of the academic research in DL algorithms for CT has focused primarily on the two-dimensional setting (with simplified forward operators and noise model) for proofs-of-concept, and a comprehensive benchmarking of various classical and data-driven CT reconstruction approaches has not beenundertaken. The key objective of our CT reconstruction grand challenge was to promote methodological advancements for both classical and DL-based approaches for clinical CT with a reasonably accurately simulated 3D CT forward operator and noise model. We have utilized the publicly available LIDC-IDRI dataset and simulated sinograms and FDK images corresponding to two dose levels (clinical- and low-dose, constituting two tracks of the challenge) starting from the normal-dose images as the ground truth. In this paper, we summarize the motivation, context, and results of our challenge, and highlight the future research directions in DL for clinical CT.

X-ray tomography↗

Evaluation of Machine Learning Approaches to Estimate Aerosol Mixing State Metrics in Atmospheric Models

Aerosol mixing state describes how aerosol compositions are distributed among atmospheric aerosol particles in a population. Oversimplified assumptions of aerosol mixing state in atmospheric modeling can introduce errors in estimations of weather and climate-relevant aerosol microphysical properties. A more comprehensive representation of the aerosol mixing state can be achieved in principle with a Particle-resolved Monte Carlo (PartMC) model but at added computational cost that may be prohibitive for direct invocation in operational numerical weather prediction or multi-year climate simulations. The aim of our research is to explore the machine learning (ML) methodologies for estimating aerosol mixing state metrics, which we define here in three different ways: with respect to (a) hygroscopicity; (b) optical properties; and (c) chemical species abundance. We adopted a data-driven approach, leveraging deep learning and statistical learning techniques, to take advantage of massive PartMC model simulations. First, we performed particle-resolved simulations by PartMC to create a series of scenarios considering a range of global environmental conditions. Each scenario consists of aerosol populations and corresponding mixing state metrics. The gas concentration, aerosol mass concentration, environmental variables and mixing state metrics of each population constitute the datasets for machine learning implementations. We have adopted and evaluated various configurations of machine learning methodologies in this investigation, embracing deep learning, Extreme Gradient Boosting (XGBoost) algorithm, and the ensemble approaches. After a rigorous model selection process, we identified an appropriate model to derive estimates of aerosol mixing state metrics. We used the computational and data resources of the Oak Ridge Leadership Computing Facility (OLCF) and the ORNL Compute and Data Environment for Science (CADES). The NVIDIA DGX-1 hardware was used for the prototyping of the ML models. Our approach has allowed us to gain a new understanding of how machine learning methodologies can be applied to improve the representation of aerosol mixing state in atmospheric models and benefit the atmospheric research community. Next, we plan to extend our research and methodology to quantify some of the aerosol-related uncertainties in the E3SM Atmospheric Model (EAM) Version 1.

54 ENVIRONMENTAL SCIENCES↗

Particle-in-cell (hPIC) data

Data generated from the hPIC particle-in-cell code to construct a surrogate ion energy-angle distribution (IEAD) model.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗