Search NASA⌕ Search

SEARCH · Search NASA

Results for “parametric model dependence”

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 55 records · Page 3

The Effects of B/L-Dependent Heating on the Formation and Evolution of a Multi-Threaded Prominence

We have developed a comprehensive, multi-threaded, three-dimensional model of the plasma dynamics and energetics of a prominence and its overlying arcade (Luna et al. 2012). In this model, the basic magnetic structure is that of two interacting sheared arcades, while the cool condensations composing the prominence are formed by the well-studied thermal nonequilibrium mechanism. In a given filament-channel flux tube, the mass is evaporated from the chromosphere by heating localized near the footpoints, and condenses in the form of transient blobs or a persistent thread. Our previous studies of thermal nonequilibrium used steady or impulsive heating functions with no dependence on local physical. However, parametric active-region models with steady heating proportional to B/L, where B is the flux-tube magnetic field strength at each footpoint and L is the flux-tube length, yield the best agreement with observations (e.g., Schrijver et al. 2008). We have determined the effects of this active-region heating function on our model for the formation and evolution of prominence mass. We have also expanded the range of our computational domain to include more ofthe overlying arcade (the so-called "cavity") than in Luna et al. (2012), and have increased the number of selected flux tubes from 125 to 533. We will illustrate the time-dependent plasma behavior produced by the B/L heating function with synthetic images in several ALA passbands, and compare the resulting prominence properties with those predicted by our model with steady heating.

Karpen, Judith T.↗

Reduced basis approximations of parameterized dynamical partial differential equations via neural networks

Projection-based reduced order models are effective at approximating parameter-dependent differential equations that are parametrically separable. When parametric separability is not satisfied, which occurs in both linear and nonlinear problems, projection-based methods fail to adequately reduce the computational complexity. Devising alternative reduced order models is crucial for obtaining efficient and accurate approximations to expensive high-fidelity models. In this work, we develop a timestepping procedure for dynamical parameter-dependent problems, in which a neural-network is trained to propagate the coefficients of a reduced basis expansion. This results in an online stage with a computational cost independent of the size of the underlying problem. Here, we demonstrate our method on several parabolic partial differential equations, including a problem that is not parametrically separable.

97 MATHEMATICS AND COMPUTING↗

Propagation of input parameter uncertainties in transport models

Here, the many sources of uncertainty in validation studies of plasma turbulence in magnetically confined fusion devices are well-known. In this paper, we investigate how to efficiently transform uncertainties in experimentally derived transport model inputs into model prediction uncertainties, using the quasilinear trapped-gyro-Landau-fluid (TGLF) turbulent transport model. We use the rapidly converging and computationally inexpensive non-intrusive probabilistic collocation method (PCM) to propagate input parameter uncertainty probability distribution functions (PDFs) through TGLF, yielding PDFs of predicted transport fluxes. We observe in many cases that the flux PDFs exhibit significant non-normal features such as strong skewness, even when the input distributions were normal. To illustrate the utility of the PCM approach, we apply this methodology to transport predictions for a DIII-D ITER baseline plasma in which the mix of neutral beam injection (NBI) and electron cyclotron heating (ECH) was varied. The model predictions show clear changes in the parametric dependencies and sensitivities of the turbulence between the two heating mixes. Specifically, when only NBI heating was used, the transport fluxes responded significantly only to the ion temperature gradient scale length. However, when both NBI and ECH were applied, the electron transport channels demonstrate a strong sensitivity to the electron temperature and density gradients not observed in the NBI-only case. Additional context for the PCM approach is provided by comparing its predictions with those obtained via a local flux-matching approach. A new set of validation metrics based on the Wasserstein distance is proposed for PDF-based comparisons.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Three-Dimensional Aerodynamic Instabilities In Multi-Stage Axial Compressors

This thesis presents the conceptualization and development of a computational model for describing three-dimensional non-linear disturbances associated with instability and inlet distortion in multistage compressors. Specifically, the model is aimed at simulating the non-linear aspects of short wavelength stall inception, part span stall cells, and compressor response to three-dimensional inlet distortions. The computed results demonstrated the first-of-a-kind capability for simulating short wavelength stall inception in multistage compressors. The adequacy of the model is demonstrated by its application to reproduce the following phenomena: (1) response of a compressor to a square-wave total pressure inlet distortion; (2) behavior of long wavelength small amplitude disturbances in compressors; (3) short wavelength stall inception in a multistage compressor and the occurrence of rotating stall inception on the negatively sloped portion of the compressor characteristic; (4) progressive stalling behavior in the first stage in a mismatched multistage compressor; (5) change of stall inception type (from modal to spike and vice versa) due to IGV stagger angle variation, and "unique rotor tip incidence" at these points where the compressor stalls through short wavelength disturbances. The model has been applied to determine the parametric dependence of instability inception behavior in terms of amplitude and spatial distribution of initial disturbance, and intra-blade-row gaps. It is found that reducing the inter-blade row gaps suppresses the growth of short wavelength disturbances. It is also concluded from these parametric investigations that each local component group (rotor and its two adjacent stators) has its own instability point (i.e. conditions at which disturbances are sustained) for short wavelength disturbances, with the instability point for the compressor set by the most unstable component group. For completeness, the methodology has been extended to describe finite amplitude disturbances in high-speed compressors. Results are presented for the response of a transonic compressor subjected to inlet distortions.

Tan, Choon S.↗

Attributes of GRB Pulses: Analysis of BATSE TTE Data

Our extensive program of modeling GRB profiles is aimed at elucidating the physical processes responsible for the burst itself, as well as possible extrinsic phenomena (e.g. time dilation) as discussed in another paper in this Symposium (Norris et al., "GRB PROFILES AS COSMIC PROBES"). We have developed special methods to extract the wealth of short time-scale information contained in the BATSE time-tag event (TTE) data. Our algorithm yields a piecewise-constant representation of the light curve -- using only the raw photon arrival times, and based on Bayesian change-point methods. This representation in effect lets the data determine the bin size and location, and avoids unwanted effects due to arbitrary choices of the bin parameters. We have determined widths, separations, and amplitudes of pulses contained in the bursts, without invoking a specific pulse model. The effect of cosmic time dilation can be easily seen in a direct plot of amplitude vs. time scale for individual pulses, without the need to lump the data into a small number of brightness classes. We are also performing noise equalization on these data (to reduce a well-known bias of pulse width as a function of signal-to-noise ratio), as well as fits of parametric pulse-shape models -- including explicit energy dependence of the pulse parameters. Such refinements are expected to improve the quality and physical significance of these results.

Norris, J. P.↗

Affine Approximation of Parametrized Kernels and Model Order Reduction for Nonlocal and Fractional Laplace Models

In this work, we consider parametrized problems driven by spatially nonlocal integral operators with parameter-dependent kernels. In particular, kernels with varying nonlocal interaction radius $\delta > 0$ and fractional Laplace kernels, parametrized by the fractional power $s\in(0,1)$, are studied. Furthermore, in order to provide an efficient and reliable approximation of the solution for different values of the parameters, we develop the reduced basis method as a parametric model order reduction approach. Major difficulties arise since the kernels are not affine in the parameters, singular, and discontinuous. Moreover, the spatial regularity of the solutions depends on the varying fractional power $s$. To address this, we derive regularity and differentiability results with respect to $\delta$ and $s$, which are of independent interest for other applications such as optimization and parameter identification. We then use these results to construct affine approximations of the kernels by local polynomials. Finally, we certify the method by providing reliable a posteriori error estimators, which account for all approximation errors, and support the theoretical findings by numerical experiments.

97 MATHEMATICS AND COMPUTING↗

New Seawater Dielectric Constant Parametrization and Application to SMOS Retrieved Salinity

The accuracy of the Sea Surface Salinity (SSS) retrieved from L-Band radiometer measurements is strongly dependent on the reliability of the dielectric constant model. Two new parametrizations were recently developed based on one hand on the Soil Moisture and Ocean Salinity (SMOS) satellite multi-angular brightness temperature measurements by Boutin et al. (2021) (BV), and on the other hand, on new George Washington University laboratory measurements by Zhou et al. (2021) (GW2020). These two approaches are fully independent. For most SSS and Sea Surface Temperature (SST) conditions commonly observed over the open ocean, the relative variations of brightness temperatures Tb simulated through the BV and GW2020 parametrizations agree particularly well, and better than with earlier parametrizations previously used in the SMOS, Soil Moisture Active Passive (SMAP) and Aquarius SSS retrievals. Nevertheless, uncertainty remains, especially below 10 °C where a ∼ 0.1 K relative difference between the two models is observed. This motivates the development of a revised parameterization, BVZ, based on a methodology similar to that used to derive BV but using GW2020 instead of SMOS measurements. Compared to the GW2020 parameterization, BVZ is derived with a reduced number of degrees of freedom, it relies on the TEOS10 PSS78 conductivity-salinity relationship, and on the previously derived static permittivity of fresh water. One month per season of SMOS data have been reprocessed in 2018 using BV, GW2020, and BVZ. We find the best overall agreement between SMOS SSS and Argo SSS with BVZ parametrization, with noticeable improvement in the 5 °C–15 °C SST range.

sea measurements↗

Physics-informed latent neural operator for real-time predictions of time-dependent parametric PDEs

Deep operator network (DeepONet) has shown significant promise as surrogate models for systems governed by partial differential equations (PDEs), enabling accurate mappings between infinite-dimensional function spaces. However, when applied to systems with high-dimensional input-output mappings arising from large numbers of spatial and temporal collocation points, these models often require heavily overparameterized networks, leading to long training times. Latent DeepONet addresses some of these challenges by introducing a two-step approach: first learning a reduced latent space using a separate model, followed by operator learning within this latent space. While efficient, this method is inherently data-driven and lacks mechanisms for incorporating physical laws, limiting its robustness and generalizability in data-scarce settings. Here, in this work, we propose PI-Latent-NO, a physics-informed latent neural operator framework that integrates governing physics directly into the learning process. Our architecture features two coupled DeepONets trained end-to-end: a Latent-DeepONet that learns a low-dimensional representation of the solution, and a Reconstruction-DeepONet that maps this latent representation back to the physical space. By embedding PDE constraints into the training via automatic differentiation, our method eliminates the need for labeled training data and ensures physics-consistent predictions. The proposed framework is both memory and compute-efficient, exhibiting near-constant scaling with problem size and demonstrating significant speedups over traditional physics-informed operator models. We validate our approach on a range of parametric PDEs, showcasing its accuracy, scalability, and suitability for real-time prediction in complex physical systems.

Latent representations↗

Convection Parametrization and Multi-Nesting Dependence of a Heavy Rainfall Event over Namibia with Weather Research and Forecasting (WRF) Model

Namibia is considered as one of the countries that are most vulnerable to climate change due to its generally dry climate and the percentage of its population that rely on subsistence agriculture for their livelihoods. Early warning systems are an important aspect of adapting to climate change. Weather forecasting relies on the use of numerical weather prediction models and these need to be configured properly. In this study we investigate the effects of using multi-nests and a convection scheme on the simulation of a heavy rainfall event over the Northwestern region of Kunene, Namibia. The event which was associated with a cut-off low system, was short-lived and resulted in over 45 mm of rainfall in one hour. For the multi-nest, a 9 km grid length parent domain is nested within the Global Forecast System (GFS) simulations, which in turn forces a 3 km grid spacing child domain. A different set of simulations are produced using a single nest of 3 km grid spacing, nested directly inside the GFS data. The simulations are produced with the convection scheme switched on and off. The impact of a single versus multi nest is found to be small in general, with slight differences in the location of high rainfall intensity. Switching off the convection schemes results in high rainfall intensity and increased detail in the simulations, including when a grid spacing of 9 km mis used. Using a grid spacing of 3 km with the convection scheme on, results in a loss of detail in the simulations as well as lower rainfall amounts. The study shows a need for different configurations to be tested before an optimum configuration can be selected for operational forecasting. We recommend further tests with different synoptic forcing to be conducted to identify a suitable configuration for Namibia.

flooding↗

Double L{sub 3}M ionization of Pd induced by impact with medium-energy electrons

The electron-induced L{sub 3}M two-step double ionization cross sections of metallic Pd were determined experimentally for incident electron beam energies ranging from the double ionization threshold up to 18 keV. The double L{sub 3}M ionization cross sections were derived from the intensity ratios (I{sub L{alpha}M}:I{sub L{alpha}}) of the resolved M satellites to the parent diagram lines. The sample was bombarded with monoenergetic electrons from an energy-tunable 20-kV electron gun. The diagram and M-satellite x-ray lines were measured by means of high-resolution x-ray spectroscopy, using a reflection-type von Hamos bent crystal spectrometer. The two-step partial cross sections were determined by subtracting from the measured total double ionization cross sections the contributions due to the shake process and L{sub 1}-L{sub 3}M{sub 4,5} Coster-Kronig transitions. Despite the thick target employed in the present study, the dependence of the two-step cross sections on the incoming electron energy could be derived using a target slice decomposition method. It is shown that the obtained energy dependence can be well reproduced by the semiempirical parametrization model of Pattard and Rost.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

The ν -process with Fully Time-dependent Supernova Neutrino Emission Spectra

The neutrino process that occurs in the outer stellar shells during a supernova explosion and involves neutrino–nucleus reactions produces a range of rare, stable, and radioactive isotopes. We improve previous ν-process studies by using, for the first time, the time-dependent neutrino emission spectra, as predicted from supernova simulations, rather than a simplified parametric description modeled after the neutron star cooling phase. In particular, our calculations use time-dependent neutrino spectra for all neutrino species, consider their deviation from a Fermi–Dirac distribution, and account for the neutrino emission from the neutrino burst and accretion phases. We find that the time-dependent treatment of the neutrino emission spectra results in higher yields for the selected nuclei produced by the ν process as compared to previous studies and to the approximation of assuming constant neutrino energies corresponding to the time-averaged mean energy radiated in each species. The effect is largest for nuclides produced by charged-current reactions. Our results reflect the dynamical competition between neutrino-induced reactions and the effect of the shock passage through the star. By varying the neutrino burst luminosity and the duration of the accretion phase, we study the impact of these early emission phases and their uncertainties on the ν-process nucleosynthesis. We find that the deviation of the neutrino spectra from a Fermi–Dirac distribution calculated in supernova simulations has a negligible effect on the ν-process yields.

79 ASTRONOMY AND ASTROPHYSICS↗

Model selection for pion photoproduction

Partial-wave analysis of meson and photon-induced reactions is needed to enable the comparison of many theoretical approaches to data. In both energy-dependent and independent parametrizations of partial waves, the selection of the model amplitude is crucial. Principles of the S matrix are implemented to a different degree in different approaches; but a many times overlooked aspect concerns the selection of undetermined coefficients and functional forms for fitting, leading to a minimal yet sufficient parametrization. We present an analysis of low-energy neutral pion photoproduction using the least absolute shrinkage and selection operator (LASSO) in combination with criteria from information theory and K-fold cross validation. These methods are not yet widely known in the analysis of excited hadrons but will become relevant in the era of precision spectroscopy. As a result, the principle is first illustrated with synthetic data; then, its feasibility for real data is demonstrated by analyzing the latest available measurements of differential cross sections (dσ/dΩ), photon-beam asymmetries (Σ), and target asymmetry differential cross sections (dσ T /d≡Tdσ/dΩ) in the low-energy regime.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Parametric Study of a YAV-8B Harrier in Ground Effect Using Time-Dependent Navier-Stokes Computations

Flow simulations using the time-dependent Navier-Stokes equations remain a challenge for several reasons. Principal among them are the difficulty to accurately model complex flows, and the time needed to perform the computations. A parametric study of such complex problems is not considered practical due to the large cost associated with computing many time-dependent solutions. The computation time for each solution must be reduced in order to make a parametric study possible. With successful reduction of computation time, the issue of accuracy, and appropriateness of turbulence models will become more tractable.

Shishir, Pandya↗

A parametric study of dissociation and ionization models at 12 km/sec

Thermochemical nonequilibrium-solution dependence on available models for the chemical reaction rates is examined. Solutions from the Kang and Dunn (1973) reaction-rate set, the Park rate set of 1987, and the Park rate set of 1991 are compared. The blunt-nosed, axisymmetric geometry considered is a 60-deg sphere cone with nose radius of 1.07 m and cicular aft skirt. The nonequilibrium test case is 12 km/sec entry into the earth's atmosphere at 80 km altitude. The model variations are implemented into the Langley aerothermodynamics upwind relaxation algorithm code. While variations in the reaction rates have no effect on the surface pressure distribution and little effect on the convective heating, the effect on degree of ionization and radiative heating can be a factor of three.

Mitcheltree, R. A.↗

Data-driven models of nonautonomous systems

Nonautonomous dynamical systems are characterized by time-dependent inputs, which complicates the discovery of predictive models describing the spatiotemporal evolution of the state variables of quantities of interest from their temporal snapshots. When dynamic mode decomposition (DMD) is used to infer a linear model, this difficulty manifests itself in the need to approximate the time-dependent Koopman operators. Our approach is to approximate the original nonautonomous system with a modified system derived via a local parameterization of the time-dependent inputs. The modified system comprises a sequence of local parametric systems, which are subsequently approximated by a parametric surrogate model using the DRIPS (dimension reduction and interpolation in parameter space) framework. The offline step of DRIPS relies on DMD to build a linear surrogate model, endowed with reduced-order bases for the observables mapped from training data. The online step interpolates on suitable manifolds to construct a sequence of iterative parametric surrogate models; the target/test parameter points on these manifolds are specified by a local parameterization of the test time-dependent inputs. Here, we use numerical experimentation to demonstrate the robustness of our method and compare its performance with that of deep neural networks.

97 MATHEMATICS AND COMPUTING↗

Molecular and Kinetic Models for High-rate Thermal Degradation of Polyethylene

Thermal degradation of polyethylene is studied under the extremely high rate temperature ramps expected in laser-driven and X-ray ablation experiments—from 10 10 to 10 14 K/s in isochoric, condensed phases. The molecular evolution and macroscopic state variables are extracted as a function of density from reactive molecular dynamics simulations using the ReaxFF potential. The enthalpy, dissociation onset temperature, bond evolution, and observed cross-linking are shown to be rate dependent. These results are used to parametrize a kinetic rate model for the decomposition and coalescence of hydrocarbons as a function of temperature, temperature ramp rate, and density. In conclusion, the results are contrasted to first-order random-scission macrokinetic models often assumed for pyrolysis of linear polyethylene under ambient conditions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Survival analysis of localized prostate cancer with deep learning

In recent years, data-driven, deep-learning-based models have shown great promise in medical risk prediction. By utilizing the large-scale Electronic Health Record data found in the U.S. Department of Veterans Affairs, the largest integrated healthcare system in the United States, we have developed an automated, personalized risk prediction model to support the clinical decision-making process for localized prostate cancer patients. This method combines the representative power of deep learning and the analytical interpretability of parametric regression models and can implement both time-dependent and static input data. To collect a comprehensive evaluation of model performances, we calculate time-dependent C-statistics C td over 2-, 5-, and 10-year time horizons using either a composite outcome or prostate cancer mortality as the target event. The composite outcome combines the Prostate-Specific Antigen (PSA) test, metastasis, and prostate cancer mortality. Our longitudinal model Recurrent Deep Survival Machine (RDSM) achieved C td 0.85 (0.83), 0.80 (0.83), and 0.76 (0.81), while the cross-sectional model Deep Survival Machine (DSM) attained C td 0.85 (0.82), 0.80 (0.82), and 0.76 (0.79) for the 2-, 5-, and 10-year composite (mortality) outcomes, respectively. In addition to estimating the survival probability, our method can quantify the uncertainty associated with the prediction. The uncertainty scores show a consistent correlation with the prediction accuracy. We find PSA and prostate cancer stage information are the most important indicators in risk prediction. Our work demonstrates the utility of the data-driven machine learning model in prostate cancer risk prediction, which can play a critical role in the clinical decision system.

60 APPLIED LIFE SCIENCES↗

Analytical Formulation for Sizing and Estimating the Dimensions and Weight of Wind Turbine Hub and Drivetrain Components

This report summarizes the theory, verification, and validation of a new sizing tool for wind turbine drivetrain components, the Drivetrain Systems Engineering (DriveSE) tool. DriveSE calculates the dimensions and mass properties of the hub, main shaft, main bearing(s), gearbox, bedplate, transformer if up-tower, and yaw system. The level of fi¬ delity for each component varies depending on whether semiempirical parametric or physics-based models are used. The physics-based models have internal iteration schemes based on system constraints and design criteria. Every model is validated against available industry data or finite-element analysis. The verification and validation results show that the models reasonably capture primary drivers for the sizing and design of major drivetrain components.

17 WIND ENERGY↗