Search NASA⌕ Search

SEARCH · Search NASA

Results for “Random effect 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 163 records · Page 9

Retention modeling for ultra-thin density of Cu-based conductive bridge random access memory (CBRAM)

We investigate the effect of Cu concentration On-state resistance retention characteristics of W/Cu/Ti/HfO{sub 2}/Pt memory cell. The development of RRAM device for application depends on the understanding of the failure mechanism and the key parameters for device optimization. In this study, we develop analytical expression for cations (Cu{sup +}) diffusion model using Gaussian distribution for detailed analysis of data retention time at high temperature. It is found that the improvement of data retention time depends not only on the conductive filament (CF) size but also on Cu atoms concentration density in the CF. Based on the simulation result, better data retention time is observed for electron wave function associated with Cu{sup +} overlap and an extended state formation. This can be verified by analytical calculation of Cu atom defects inside the filament, based on Cu{sup +} diffusion model. The importance of Cu diffusion for the device reliability and the corresponding local temperature of the filament were analyzed by COMSOL Multiphysics simulation.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

A survey of unsupervised learning methods for high-dimensional uncertainty quantification in black-box-type problems

Constructing surrogate models for uncertainty quantification (UQ) on complex partial differential equations (PDEs) having inherently high-dimensional O(10 n ), n ≥ 2, stochastic inputs (e.g., forcing terms, boundary conditions, initial conditions) poses tremendous challenges. The “curse of dimensionality” can be addressed with suitable unsupervised learning techniques used as a pre-processing tool to encode inputs onto lower-dimensional subspaces while retaining its structural information and meaningful properties. In this work, we review and investigate thirteen dimension reduction methods including linear and nonlinear, spectral, blind source separation, convex and non-convex methods and utilize the resulting embeddings to construct a mapping to quantities of interest via polynomial chaos expansions (PCE). Here, we refer to the general proposed approach as manifold PCE (m-PCE), where manifold corresponds to the latent space resulting from any of the studied dimension reduction methods. To investigate the capabilities and limitations of these methods we conduct numerical tests for three physics-based systems (treated as black-boxes) having high-dimensional stochastic inputs of varying complexity modeled as both Gaussian and non-Gaussian random fields to investigate the effect of the intrinsic dimensionality of input data. We demonstrate both the advantages and limitations of the unsupervised learning methods and we conclude that a suitable m-PCE model provides a cost-effective approach compared to alternative algorithms proposed in the literature, including recently proposed expensive deep neural network-based surrogates and can be readily applied for high-dimensional UQ in stochastic PDEs.

42 ENGINEERING↗

Bottom Fixed Platform Dynamics Models Assessing Surface Ice Interactions for Transitional Depth Structures in the Great Lakes: FAST8 – IceDyn

To create long-term solutions for offshore wind turbines in a variety of environmental conditions, CAE tools are needed to model the design-driving loads that interact with an offshore wind turbine system during operation. This report describes our efforts in augmenting existing CAE tools used for offshore wind turbine analysis with a new module that can provide simulation capabilities for ice loading on the system. This augmentation was accomplished by creating an ice-loading module coupled to FAST8, the CAE tool maintained by the NREL for simulating land-based and offshore wind turbine dynamics. The new module includes both static and dynamic ice loading that can be applied during a dynamic simulation of the response of an offshore wind turbine. The ice forces can be prescribed, or influenced by the structure’s compliant response, or by the dynamics of both the structure and the ice floe. The new module covers ice failure modes of spalling, buckling, crushing, splitting, and bending. The supporting structure of wind turbines can be modeled as a vertical or sloping form at the waterline. The Inward Battered Guide Structure (IBGS) foundation designed by Keystone Engineering for the Great Lakes was used to study the ice models coupled to FAST8. The IBGS foundation ice loading simulations in FAST8 were compared to the baseline simulation case without ice loading. The ice conditions reflecting those from Lake Huron at Port Huron and Lake Michigan at North Manitou were studied under near rated wind speed of 12 m/s for the NREL 5-MW reference turbine. Simulations were performed on ice loading models 1 through 4 and ice model 6 with their respective sub-models. The purpose of ice model 5 is to investigate ice loading on sloping structures such as ice-cones on a monopile and is not suitable for multi-membered jacketed structures like the IBGS foundation. The key response parameters from the simulations, shear forces and moments from the tower base and IBGS foundation base, were compared. Ice models 1 and 6 do not significantly affect the tower fore-aft shear and moment. However, ice model 2 (dynamic analyses), model 3 (random ice loading), and model 4 (multiple ice failure zone loading) show increased effect on the tower fore-aft shear and moment with significant effect from ice model 3.1. In general ice loading creates large reaction forces and moments at the base of the IBGS foundation; the largest occurred in model 1.1 (steady creep ice indentation loading) followed by model 3.1 (random creep ice indentation loading). In general the power production from the ice loading cases had little deviation from the baseline case without ice loading. For ultimate limit state (ULS), ice model 1.1 ice and 3.1 appear to be the ice most critical models to consider at an early stage of design. Ice model 4 is an important tool for assessing structural fatigue.

17 WIND ENERGY↗

Uncertainty quantification in reacting flow.

Chemically reacting flow models generally involve inputs and parameters that are determined from empirical measurements, and therefore exhibit a certain degree of uncertainty. Estimating the propagation of this uncertainty into computational model output predictions is crucial for purposes of reacting flow model validation, model exploration, as well as design optimization. Recent years have seen great developments in probabilistic methods and tools for efficient uncertainty quantification (UQ) in computational models. These tools are grounded in the use of Polynomial Chaos (PC) expansions for representation of random variables. The utility and effectiveness of PC methods have been demonstrated in a range of physical models, including structural mechanics, transport in porous media, fluid dynamics, aeronautics, heat transfer, and chemically reacting flow. While high-dimensionality remains nominally an ongoing challenge, great strides have been made in dealing with moderate dimensionality along with non-linearity and oscillatory dynamics. In this talk, I will give an overview of UQ in chemical systems. I will cover both: (1) the estimation of uncertain input parameters from empirical data, and (2) the forward propagation of parametric uncertainty to model outputs. I will cover the basics of forward PC UQ methods with examples of their use. I will also highlight the need for accurate estimation of the joint probability density over the uncertain parameters, in order to arrive at meaningful estimates of model output uncertainties. Finally, I will discuss recent developments on the inference of this density given partial information from legacy experiments, in the absence of raw data.

99 GENERAL AND MISCELLANEOUS↗

Uncertainty quantification in reacting flow.

Chemically reacting flow models generally involve inputs and parameters that are determined from empirical measurements, and therefore exhibit a certain degree of uncertainty. Estimating the propagation of this uncertainty into computational model output predictions is crucial for purposes of reacting flow model validation, model exploration, as well as design optimization. Recent years have seen great developments in probabilistic methods and tools for efficient uncertainty quantification (UQ) in computational models. These tools are grounded in the use of Polynomial Chaos (PC) expansions for representation of random variables. The utility and effectiveness of PC methods have been demonstrated in a range of physical models, including structural mechanics, transport in porous media, fluid dynamics, aeronautics, heat transfer, and chemically reacting flow. While high-dimensionality remains nominally an ongoing challenge, great strides have been made in dealing with moderate dimensionality along with non-linearity and oscillatory dynamics. In this talk, I will give an overview of UQ in chemical systems. I will cover both: (1) the estimation of uncertain input parameters from empirical data, and (2) the forward propagation of parametric uncertainty to model outputs. I will cover the basics of forward PC UQ methods with examples of their use. I will also highlight the need for accurate estimation of the joint probability density over the uncertain parameters, in order to arrive at meaningful estimates of model output uncertainties. Finally, I will discuss recent developments on the inference of this density given partial information from legacy experiments, in the absence of raw data.

99 GENERAL AND MISCELLANEOUS↗

Quantum fragmentation in the extended quantum breakdown model

We introduce a one-dimensional (1D) extended quantum breakdown model comprising a fermionic and a spin degree of freedom per site, and featuring a spatially asymmetric breakdown-type interaction between the fermions and spins. Furthermore, our model resembles the breakdown process of particles incident into a cloud chamber with nonzero quantum amplitudes of both exciting and not exciting the local vapor atoms. We analytically show that, in the absence of any magnetic field for the spins, the model exhibits Hilbert space fragmentation within each symmetry sector into exponentially many Krylov subspaces and hence displays nonthermal dynamics. Here, we demonstrate that the fragmentation naturally occurs in an entangled basis and thus provides an example of “quantum fragmentation.” Besides establishing the nature of fragmentation analytically, we also study the long-time behavior of the entanglement entropy and its deviation from the expected Page value as a probe of ergodicity in the system. Upon introducing a magnetic field for the spins, most of the Krylov subspaces merge and the model becomes chaotic. Finally, we study the effects of strong randomness on the system and observe behavior similar to that of many-body localized systems.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Analytical homogenization techniques applied to the Fickian diffusion: Effective diffusivity coefficient

For multiple applications in nuclear energy, the ability to accurately represent material behavior with a simplified model is important to facilitate practical engineering-scale simulations. In this work, we focus on the homogenized thermal response of a medium containing spherical inclusions, similar to a fuel form (compact or pebble) containing TRISO particles. An extensive survey on effective thermal conductivity modeling was performed in our previous study, considering a random distribution of mono-sized spherical inclusions in a continuous matrix. Using the analogy between heat conduction and the simplified Fickian diffusion (or fission product species conservation), we can use the same analytical homogenization methods to obtain ETC as for the effective diffusivity coefficient (EDC). We performed several numerical experiments at varying conditions to assess the validity of our hypothesis for EDC calculations.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Advancements on Multi-Fidelity Random Fourier Neural Networks: Application to Hurricane Modeling for Wind Energy

Multi-fidelity approaches are emerging as effective strategies in computational science to handle otherwise intractable tasks like Uncertainty Quantification (UQ), training of Machine Learning (ML) models, and optimization, for expensive high-fidelity applications in which the amount of available simulations or data is limited. The main idea is simple: large datasets generated for low-fidelity approximations of the problem at hand are fused with a much sparser dataset for the target (high-fidelity) system. In this paper, we build on our recent success in designing random Fourier Neural Networks (rFNNs) [1] to target problems arising in wind energy applications and in particular problems of interest for hurricane modeling. In this context, data for the high-fidelity models are limited and lower fidelity alternatives are needed. In this work, we introduce a novel multi-fidelity training approach for our rFNNs and demonstrate its use on a simple verification problem and on a hurricane modeling problem in which high-fidelity data are generated via Large-Eddy Simulations (LES), while low-fidelity data are given by a mesoscale model. Initial results demonstrate how the multi-fidelity training approach can improve the quality of the resulting surrogate.

Fourier Neural Networks↗

Physical mechanisms of copper-copper wafer bonding

The study of the physical mechanisms driving Cu-Cu wafer bonding allowed for reducing the bonding temperatures below 200 °C. Metal thermo-compression Cu-Cu wafer bonding results obtained at such low temperatures are very encouraging and suggest that the process is possible even at room temperature if some boundary conditions are fulfilled. Sputtered (PVD) and electroplated Cu thin layers were investigated, and the analysis of both metallization techniques demonstrated the importance of decreasing Cu surface roughness. For an equal surface roughness, the bonding temperature of PVD Cu wafers could be even further reduced due to the favorable microstructure. Their smaller grain size enhances the length of the grain boundaries (observed on the surface prior bonding), acting as efficient mass transfer channels across the interface, and hence the grains are able to grow over the initial bonding interface. Due to the higher concentration of random high-angle grain boundaries, this effect is intensified. The model presented is explaining the microstructural changes based on atomic migration, taking into account that the reduction of the grain boundary area is the major driving force to reduce the Gibbs free energy, and predicts the subsequent microstructure evolution (grain growth) during thermal annealing.

36 MATERIALS SCIENCE↗

Using Machine Learning to Analyze Factors Determining Cycle-to-Cycle Variation in a Spark-Ignited Gasoline Engine

In this work, we have applied a machine learning (ML) technique to provide insights into the causes of cycle-to-cycle variation (CCV) in a gasoline spark-ignited (SI) engine. The analysis was performed on a set of large eddy simulation (LES) calculations of a single cylinder of a four-cylinder port-fueled SI engine. The operating condition was stoichiometric, without significant knock, at a load of 16 bar brake mean effective pressure (BMEP), at an engine speed of 2500 revolutions per minute. A total of 123 cycles was simulated. Of these, 49 were run in sequence, while 74 were run in parallel. For the parallel approach, each cycle is initialized with its own synthetic turbulent field to generate CCV, as part of another work performed by us. In the current work, we used 3D information from all 123 cycles to compute flame topology and pre-ignition flow-field metrics. We then evaluated correlations between these metrics, and peak cylinder pressure (PCP) employing an ML technique called random forest. The computed metrics form the inputs to the random forest model, and PCP is the output. This model captures the effect of all inputs, as well as interactions between them owing to its decision-tree structure. The goal of this work is to demonstrate (as a first step) that ML models can implicitly learn complex relationships between pre-ignition flow-fields, flame shapes, and the eventual outcome of the cycle (whether a cycle will be a high or a low cycle).

42 ENGINEERING↗

Aided Active Learning (AAL) for Enhanced Critical Heat Flux Prediction

Accurate prediction of critical heat flux (CHF) is crucial for the safe and efficient operation of nuclear reactors. Traditional CHF modeling methods often require extensive experimental data, which are hard to obtain. This study introduces the Aided Active Learning (AAL) framework, which strategically minimizes data requirements without sacrificing model accuracy. Unlike conventional Active Learning (AL), AAL introduces an additional step of randomly selecting a subset from the sample pool before applying the query strategy. To evaluate the performance of AAL, two query strategies—uncertainty-based sampling and error-reduction sampling—were evaluated across the following models: random forest (RF), feedforward neural network (FNN), and variational feedforward neural network (vFNN). The proposed framework demonstrated that AAL effectively reduces the number of training samples needed to achieve comparable predictive accuracy. For the RF model, AL required only 710 samples to achieve an R2 score of 0.98, as compared to the 4,785 samples needed by random sampling. Similarly, the FNN model achieved the same R2 score with just 355 samples when using AL, a significant improvement over the 825 samples required by random sampling. In case of uncertainty-based sampling strategy, vFNN attained an R2 of 0.98 with 3,420 samples, reducing the sample requirement by 47% relative to the 6,440 samples needed for random sampling. Its performance suggests that larger training data are required to fully leverage its uncertainty quantification capabilities.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗

Fluctuations destroying long-range order in SU(2) Yang-Mills theory

We study lattice SU(2) Yang-Mills theory with dimension d{>=}4. The model can be expressed as a (d-1)-dimensional O(4) nonlinear {sigma} model in a d-dimensional heat bath. As is well known, the nonlinear {sigma} model alone shows a phase transition. If the quark confinement is a consequence of the absence of a phase transition for the Yang-Mills theory, then the fluctuations of the heat bath must destroy the long-range order of the nonlinear {sigma} model. In order to clarify whether this is true, we replace the fluctuations of the heat bath with Gaussian random variables, and obtain a Langevin equation which yields the effective action of the nonlinear {sigma} model by analyzing the Fokker-Planck equation. It turns out that the fluctuations indeed destroy the long-range order of the nonlinear {sigma} model within a mean-field approximation estimating a critical point, whereas for the corresponding U(1) gauge theory, the phase transition to the massless phase remains against the fluctuations.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

The Impact of White Dwarf Luminosity Profiles on Oscillation Frequencies

KIC 08626021 is a pulsating DB white dwarf (WD) of considerable recent interest, and the first of its class to be extensively monitored by Kepler for its pulsation properties. Fitting the observed oscillation frequencies of KIC 08626021 to a model can yield insights into its otherwise-hidden internal structure. Template-based WD models choose a luminosity profile where the luminosity is proportional to the enclosed mass, ${L}_{r}\,\propto \,{M}_{r}$, independent of the effective temperature T eff . Evolutionary models of young WDs with T eff ≳ 25,000 K suggest that neutrino emission gives rise to luminosity profiles with L r ${/}\!\!\!\!\!\!{\propto }$ M r . We explore this contrast by comparing the oscillation frequencies between two nearly identical WD models: one with an enforced ${L}_{r}\propto {M}_{r}$ luminosity profile, and the other with a luminosity profile determined by the star's previous evolution history. We find that the low-order g-mode frequencies differ by up to ≃70 μHz over the range of Kepler observations for KIC 08626021. This suggests that by neglecting the proper thermal structure of the star (e.g., accounting for the effect of plasmon neutrino losses), the model frequencies calculated by using an ${L}_{r}\propto {M}_{r}$ profile may have uncorrected, effectively random errors at the level of tens of μHz. A mean frequency difference of 30 μHz, based on linearly extrapolating published results, suggests a template model uncertainty in the fit precision of ≃12% in WD mass, ≃9% in the radius, and ≃3% in the central oxygen mass fraction.

79 ASTRONOMY AND ASTROPHYSICS↗

Engineering atomic quantum reservoirs for photons

We present protocols for creating entangled states of two modes of the electromagnetic field by using a beam of atoms crossing microwave resonators. The atoms are driven by a transverse, classical field and pump correlated photons into (i) two modes of a cavity and (ii) the modes of two distant cavities. The protocols are based on a stochastic dynamics, characterized by random arrival times of the atoms and by random interaction times between atoms and cavity modes. The resulting effective model yields a master equation, whose steady state is an entangled state of the cavity modes. In this respect, the atoms act like a quantum reservoir, pulling the cavity modes into an entangled, Einstein-Podolski-Rosen state, whose degree of entanglement is controlled by the intensity and the frequency of the transverse field. This scheme is robust against stochastic fluctuations in the atomic beam, and it does not require atomic detection nor velocity selection.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Thermo-Mechanical Phase-Field Modeling of Fracture in High-Burnup UO2 Fuels Under Transient Conditions

This study presents a novel multiphysics phase-field fracture model to analyze high-burnup uranium dioxide (UO2) fuel behavior under transient reactor conditions. Fracture is treated as a stochastic phase transition, which inherently accounts for the random microstructural effects that lead to variations in the value of fracture strength. Moreover, the model takes into consideration the effects of temperature and burnup on thermal conductivity. Therefore, the model is able to predict crack initiation, propagation, and complex morphologies in response to thermal gradients and stress distributions. Several simulations were conducted to investigate the effects of operational and transient conditions on fracture behavior and the resulting cracking patterns. High-burnup fuels exhibit reduced thermal conductivity, elevating temperature gradients and resulting in extensive radial and circumferential cracks. Transient heating rates and temperatures significantly affect fracture patterns, with higher heating rates generating steeper gradients and more irregular crack trajectories. This approach provides critical insights into fuel integrity during accident scenarios and supports the safety evaluation of extended burnup limits.

Chemistry↗

A practical method to test the validity of the standard Gumbel distribution in logit-based multinomial choice models of travel behavior

Most multinomial choice models (e.g., the multinomial logit model) adopted in practice assume an extreme-value Gumbel distribution for the random components (error terms) of utility functions. This distributional assumption offers a closed-form likelihood expression when the utility maximization principle is applied to model choice behaviors. As a result, model coefficients can be easily estimated using the standard maximum likelihood estimation method. However, maximum likelihood estimators are consistent and efficient only if distributional assumptions on the random error terms are valid. It is therefore critical to test the validity of underlying distributional assumptions on the error terms that form the basis of parameter estimation and policy evaluation. In this paper, a practical yet statistically rigorous method is proposed to test the validity of the distributional assumption on the random components of utility functions in both the multinomial logit (MNL) model and multiple discrete-continuous extreme value (MDCEV) model. Based on a semi-nonparametric approach, a closed-form likelihood function that nests the MNL or MDCEV model being tested is derived. The proposed method allows traditional likelihood ratio tests to be used to test violations of the standard Gumbel distribution assumption. Simulation experiments are conducted to demonstrate that the proposed test yields acceptable Type-I and Type-II error probabilities at commonly available sample sizes. The test is then applied to three real-world discrete and discrete-continuous choice models. For all three models, the proposed test rejects the validity of the standard Gumbel distribution in most utility functions, calling for the development of robust choice models that overcome adverse effects of violations of distributional assumptions on the error terms in random utility functions.

30 DIRECT ENERGY CONVERSION↗

Cosmology with anisotropic galaxy clustering from the combination of power spectrum and bispectrum

The apparent anisotropies of the galaxy clustering in observable redshift space provide a unique opportunity to simultaneously probe cosmic expansion and gravity on cosmological scales via the Alcock-Paczynski effect and redshift-space distortions. While the improved theoretical models have been proposed and developed to describe the apparent anisotropic clustering at weakly non-linear scales, the applicability of these models is still limited in the presence of the non-perturbative smearing effect caused by the randomness of the relative velocities. Although the cosmological constraint from the anisotropic clustering will be improved with a more elaborate theoretical model, here we consider an alternative approach using the statistical power of both the power spectrum and bispectrum at large scales. Based on the Fisher matrix analysis, we estimate the benefit of combining the power spectrum and bispectrum, finding that for the future spectroscopy survey DESI (Dark Energy Spectroscopy Instrument), the constraints on the cosmic expansion and growth of structure will be improved by a factor of two. This approach compensates for the loss of constraining power, using the power spectrum alone, due to the randomness of the relative velocities.

79 ASTRONOMY AND ASTROPHYSICS↗

Fast estimation of the look-elsewhere effect using Gaussian random fields

Abstract We discuss the use of Gaussian random fields to estimate the look-elsewhere effect correction. We show that Gaussian random fields can be used to model the null-hypothesis significance maps from a large set of statistical problems commonly encountered in physics, such as template matching and likelihood ratio tests. Some specific examples are searches for dark matter using pixel arrays, searches for astronomical transients, and searches for fast-radio bursts. Gaussian random fields can be sampled efficiently in the frequency domain, and the excursion probability can be fitted with these samples to extend any estimation of the look-elsewhere effect to lower p values. In addition, in cases where the Gaussian random field is stationary and the parameter space is Euclidean, the look-elsewhere effect correction can be computed analytically. We demonstrate these methods using two example template matching problems. Finally, we apply these methods to estimate the trial factor of a $$4^3$$ 4 3 accelerometer array for the detection of dark matter tracks in the Windchime project. When a global significance of $$3\sigma $$ 3 σ is required, the estimated trial factor for such an accelerometer array is $$10^{14}$$ 10 14 for a one-second search, and $$10^{22}$$ 10 22 for a 1-year search.

Qin, Juehang (ORCID:0000000182288949)↗