Search NASA⌕ Search

SEARCH · Search NASA

Results for “complex modeling”

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 991 records · Page 55

Basis set consistent revision of the S22 test set of noncovalent interaction energies

The S22 test set of interaction energies for small model complexes has been very valuable for benchmarking new and existing methods for noncovalent interactions. However, the basis sets utilized to compute the CCSD(T) interaction energies for some of the dimers are insufficient to obtain converged results. Here we consistently extrapolate all CCSD(T)/complete basis set (CBS) interaction energies using larger basis sets for the CCSD(T) component of the computation. The revised values, which we designate S22A, represent the most accurate results to date for this set of dimers. The new values appear to be within a few hundredths of 1 kcal mol –1 of the true CCSD(T)/CBS limit at the given geometries, but the former S22 values are off by as much as 0.6 kcal mol –1 compared to the revised values. Because some of the most promising methods for noncovalent interactions are already achieving this level of agreement (or better) compared to the S22 data, more accurate benchmark values would clearly be helpful. The MP2, SCS-MP2, SCS-CCSD, SCS(MI)-MP2, and B2PLYP-D methods have been tested against the more accurate benchmark set. The B2PLYP-D method outperforms all other methods tested here, with a mean average deviation of only 0.12 kcal mol –1 . Furthermore, the consistent, slight underestimation of the interaction energies computed by the SCS-CCSD method (an overall mean absolute deviation and mean deviation of 0.24 and –0.23 kcal mol –1 , respectively) suggests that the SCS-CCSD method has the potential to become even more accurate with a reoptimization of its parameters for noncovalent interactions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Bayesian soft X-ray tomography using non-stationary Gaussian Processes

In this study, a Bayesian based non-stationary Gaussian Process (GP) method for the inference of soft X-ray emissivity distribution along with its associated uncertainties has been developed. For the investigation of equilibrium condition and fast magnetohydrodynamic behaviors in nuclear fusion plasmas, it is of importance to infer, especially in the plasma center, spatially resolved soft X-ray profiles from a limited number of noisy line integral measurements. For this ill-posed inversion problem, Bayesian probability theory can provide a posterior probability distribution over all possible solutions under given model assumptions. Specifically, the use of a non-stationary GP to model the emission allows the model to adapt to the varying length scales of the underlying diffusion process. In contrast to other conventional methods, the prior regularization is realized in a probability form which enhances the capability of uncertainty analysis, in consequence, scientists who concern the reliability of their results will benefit from it. Under the assumption of normally distributed noise, the posterior distribution evaluated at a discrete number of points becomes a multivariate normal distribution whose mean and covariance are analytically available, making inversions and calculation of uncertainty fast. Additionally, the hyper-parameters embedded in the model assumption can be optimized through a Bayesian Occam's Razor formalism and thereby automatically adjust the model complexity. This method is shown to produce convincing reconstructions and good agreements with independently calculated results from the Maximum Entropy and Equilibrium-Based Iterative Tomography Algorithm methods.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Photolysis of CH{sub 3}CHO at 248 nm: Evidence of triple fragmentation from primary quantum yield of CH{sub 3} and HCO radicals and H atoms

Radical quantum yields have been measured following the 248 nm photolysis of acetaldehyde, CH{sub 3}CHO. HCO radical and H atom yields have been quantified by time resolved continuous wave Cavity Ring Down Spectroscopy in the near infrared following their conversion to HO{sub 2} radicals by reaction with O{sub 2}. The CH{sub 3} radical yield has been determined using the same technique following their conversion into CH{sub 3}O{sub 2}. Absolute yields have been deduced for HCO radicals and H atoms through fitting of time resolved HO{sub 2} profiles, obtained under various O{sub 2} concentrations, to a complex model, while the CH{sub 3} yield has been determined relative to the CH{sub 3} yield from 248 nm photolysis of CH{sub 3}I. Time resolved HO{sub 2} profiles under very low O{sub 2} concentrations suggest that another unknown HO{sub 2} forming reaction path exists in this reaction system besides the conversion of HCO radicals and H atoms by reaction with O{sub 2}. HO{sub 2} profiles can be well reproduced under a large range of experimental conditions with the following quantum yields: CH{sub 3}CHO + hν{sub 248nm} → CH{sub 3}CHO{sup *}, CH{sub 3}CHO{sup *} → CH{sub 3} + HCO ϕ{sub 1a} = 0.125 ± 0.03, CH{sub 3}CHO{sup *} → CH{sub 3} + H + CO ϕ{sub 1e} = 0.205 ± 0.04, CH{sub 3}CHO{sup *}→{sup o{sub 2}}CH{sub 3}CO + HO{sub 2} ϕ{sub 1f} = 0.07 ± 0.01. The CH{sub 3}O{sub 2} quantum yield has been determined in separate experiments as ϕ{sub CH{sub 3}} = 0.33 ± 0.03 and is in excellent agreement with the CH{sub 3} yields derived from the HO{sub 2} measurements considering that the triple fragmentation (R1e) is an important reaction path in the 248 nm photolysis of CH{sub 3}CHO. From arithmetic considerations taking into account the HO{sub 2} and CH{sub 3} measurements we deduce a remaining quantum yield for the molecular pathway: CH{sub 3}CHO{sup *} → CH{sub 4} + CO ϕ{sub 1b} = 0.6. All experiments can be consistently explained with absence of the formerly considered pathway: CH{sub 3}CHO{sup *} → CH{sub 3}CO + H ϕ{sub 1c} = 0.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Stochastic self-tuning hybrid algorithm for reaction-diffusion systems

Many biochemical phenomena involve reactants with vastly different concentrations, some of which are amenable to continuum-level descriptions, while the others are not. We present a hybrid self-tuning algorithm to model such systems. The method combines microscopic (Brownian) dynamics for diffusion with mesoscopic (Gillespie-type) methods for reactions and remains efficient in a wide range of regimes and scenarios with large variations of concentrations. Furthermore, its accuracy, robustness, and versatility are balanced by redefining propensities and optimizing the mesh size and time step. We use a bimolecular reaction to demonstrate the potential of our method in a broad spectrum of scenarios: from almost completely reaction-dominated systems to cases where reactions rarely occur or take place very slowly. The simulation results show that the number of particles present in the system does not degrade the performance of our method. This makes it an accurate and computationally efficient tool to model complex multireaction systems.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Power characteristics of lasers with quantum-well waveguides and blocking layers

The power characteristics of lasers with a quantum-well (QW) waveguide are calculated based on a complex model taking into account the diffusion and drift of charge carriers, capture of electrons and holes at the QW levels, spatial distribution of radiation in the cavity, intraband absorption, and heating of the active region. Structures with doped blocking barrier layers are proposed, which allow one to decrease leakage currents to a few percent. The calculated average output power of a laser based on the GaInAs/GaAs QW heterostructure (aperture 100 μm) with a QW waveguide and a GaAsP doped blocking layer in a pulsed regime was 29 W at 80-A current pulses with a duration of 5 μs. (lasers)

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Mathematical modelling of the concave front in the adjacent high explosive detonation problem

This study presents an analysis of the transition-zone in adjacent high explosive (HE) detonation problems which uses a $D, 𝜅, \dot{D}$ relationship, where $D$ is the detonation front-normal velocity, 𝜅 is the detonation front curvature and $\dot{D}$ is the time derivative of detonation front-normal velocity. Our approach extends the traditional $(D, 𝜅)$ model to accurately predict the behaviour of both diverging and converging detonation shock fronts. Our findings affirm that a hyperbolic type of front evolution equation, enhanced with wave acceleration, provides a robust framework for modelling complex shock front dynamics in HE materials. This approach not only captures the natural effects of straightness and boundary slope jumps in the transition-zone but also bridges the gap between mathematical predictions and experimental observations, offering insights into the behaviour of both diverging and converging detonation propagations in a homogeneous HE.

acceleration↗

A LINK TO THE PAST: USING MARKOV CHAIN MONTE CARLO FITTING TO CONSTRAIN FUNDAMENTAL PARAMETERS OF HIGH-REDSHIFT GALAXIES

We have developed a new method for fitting spectral energy distributions (SEDs) to identify and constrain the physical properties of high-redshift (4 < z < 8) galaxies. Our approach uses an implementation of Bayesian based Markov Chain Monte Carlo that we have dubbed '{pi}MC{sup 2}'. It allows us to compare observations to arbitrarily complex models and to compute 95% credible intervals that provide robust constraints for the model parameters. The work is presented in two sections. In the first, we test {pi}MC{sup 2} using simulated SEDs to not only confirm the recovery of the known inputs but to assess the limitations of the method and identify potential hazards of SED fitting when applied specifically to high-redshift (z > 4) galaxies. In the second part of the paper we apply {pi}MC{sup 2} to thirty-three 4 < z < 8 objects, including the spectroscopically confirmed Grism ACS Program for Extragalactic Science Ly{alpha} sample (4 < z < 6), supplemented by newly obtained Hubble Space Telescope/WFC3 near-IR observations, and several recently reported broadband selected z > 6 galaxies. Using {pi}MC{sup 2}, we are able to constrain the stellar mass of these objects and in some cases their stellar age and find no evidence that any of these sources formed at a redshift larger than z = 8, a time when the universe was Almost-Equal-To 0.6 Gyr old.

79 ASTRONOMY AND ASTROPHYSICS↗

LOCALIZATION OF GAMMA-RAY BURSTS USING THE FERMI GAMMA-RAY BURST MONITOR

The Fermi Gamma-ray Burst Monitor (GBM) has detected over 1400 gamma-ray bursts (GRBs) since it began science operations in 2008 July. We use a subset of over 300 GRBs localized by instruments such as Swift, the Fermi Large Area Telescope, INTEGRAL, and MAXI, or through triangulations from the InterPlanetary Network, to analyze the accuracy of GBM GRB localizations. We find that the reported statistical uncertainties on GBM localizations, which can be as small as 1°, underestimate the distance of the GBM positions to the true GRB locations and we attribute this to systematic uncertainties. The distribution of systematic uncertainties is well represented (68% confidence level) by a 3.°7 Gaussian with a non-Gaussian tail that contains about 10% of GBM-detected GRBs and extends to approximately 14°. A more complex model suggests that there is a dependence of the systematic uncertainty on the position of the GRB in spacecraft coordinates, with GRBs in the quadrants on the Y axis better localized than those on the X axis.

79 ASTRONOMY AND ASTROPHYSICS↗

The look-elsewhere effect from a unified Bayesian and frequentist perspective

When searching over a large parameter space for anomalies such as events, peaks, objects, or particles, there is a large probability that spurious signals with seemingly high signi ficance will be found. This is known as the look-elsewhere effect and is prevalent throughout cosmology, (astro)particle physics, and beyond. To avoid making false claims of detection, one must account for this effect when assigning the statistical significance of an anomaly. This is typically accomplished by considering the trials factor, which is generally computed numerically via potentially expensive simulations. In this paper we develop a continuous generalization of the Bonferroni and Sidak corrections by applying the Laplace approximation to evaluate the Bayes factor, and in turn relating the trials factor to the prior-to-posterior volume ratio. Here, we use this to define a test statistic whose frequentist properties have a simple interpretation in terms of the global p-value, or statistical significance. We apply this method to various physics-based examples and show it to work well for the full range of p-values, i.e. in both the asymptotic and non-asymptotic regimes. We also show that this method naturally accounts for other model complexities such as additional degrees of freedom, generalizing Wilks' theorem. This provides a fast way to quantify statistical significance in light of the look-elsewhere effect, without resorting to expensive simulations.

79 ASTRONOMY AND ASTROPHYSICS↗

Deep inference of simulated strong lenses in ground-based surveys

The large number of strong lenses discoverable in future astronomical surveys will likely enhance the value of strong gravitational lensing as a cosmic probe of dark energy and dark matter. However, leveraging the increased statistical power of such large samples will require further development of automated lens modeling techniques. We show that deep learning and simulation-based inference (SBI) methods produce informative and reliable estimates of parameter posteriors for strong lensing systems in ground-based surveys. We present the examination and comparison of two approaches to lens parameter estimation for strong galaxy-galaxy lenses — Neural Posterior Estimation (NPE) and Bayesian Neural Networks (BNNs). We perform inference on 1-, 5-, and 12-parameter lens models for ground-based imaging data that mimics the Dark Energy Survey (DES). We find that NPE outperforms BNNs, producing posterior distributions that are more accurate, precise, and well-calibrated for most parameters. For the 12-parameter NPE model, the calibration is consistently within <10% of optimal calibration for all parameters, while the BNN is rarely within 20% of optimal calibration for any of the parameters. Similarly, residuals for most of the parameters are smaller (by up to an order of magnitude) with the NPE model than the BNN model. This work takes important steps in the systematic comparison of methods for different levels of model complexity.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Conformalized-KANs: Uncertainty Quantification with Coverage Guarantees for Kolmogorov-Arnold Networks (KANs) in Scientific Machine Learning

This paper explores uncertainty quantification (UQ) methods in the context of Kolmogorov–Arnold Networks (KANs). We apply an ensemble approach to KANs to obtain a heuristic measure of UQ, enhancing interpretability and robustness in modeling complex functions. Building on this, we introduce Conformalized-KANs, which integrate conformal prediction, a distribution-free UQ technique, with KAN ensembles to generate calibrated prediction intervals with guaranteed coverage.} Extensive numerical experiments are conducted to evaluate the effectiveness of these methods, focusing particularly on the robustness and accuracy of the prediction intervals under various hyperparameter settings. We show that the conformal KAN predictions can be applied to recent extensions of KANs, including Finite Basis KANs (FBKANs) and multifideilty KANs (MFKANs). The results demonstrate the potential of our approaches to significantly improve the reliability and applicability of KANs in scientific machine learning.

• Artificial intelligence (AI) / machine learning ↗

Impact of ejecta morphology and composition on the electromagnetic signatures of neutron star mergers

The electromagnetic transients accompanying compact binary mergers (γ-ray bursts, afterglows and ‘macronovae’) are crucial to pinpoint the sky location of gravitational wave sources. Macronovae are caused by the radioactivity from freshly synthesized heavy elements, e.g. from dynamic ejecta and various types of winds. We study macronova signatures by using multidimensional radiative transfer calculations. We then employ the radiative transfer code SUPERNU and state-of-the-art LTE opacities for a few representative elements from the wind and dynamical ejecta (Cr, Pd, Se, Te, Br, Zr, Sm, Ce, Nd, U) to calculate synthetic light curves and spectra for a range of ejecta morphologies. The radioactive power of the resulting macronova is calculated with the detailed input of decay products. We assess the detection prospects for our most complex models, based on the portion of viewing angles that are sufficiently bright, at different cosmological redshifts (z). The brighter emission from the wind is unobscured by the lanthanides (or actinides) in some of the models, permitting non-zero detection probabilities for redshifts up to z = 0.07. We also find that the nuclear mass model and the resulting radioactive heating rate are crucial for the detectability. While for the most pessimistic heating rate (from the finite range droplet model) no reasonable increase in the ejecta mass or velocity, or wind mass or velocity, can possibly make the light curves agree with the observed near-infrared excess after GRB130603B, a more optimistic heating rate (from the Duflo–Zuker model) leads to good agreement. We conclude that future reliable macronova observations would constrain nuclear heating rates, and consequently help constrain nuclear mass models.

79 ASTRONOMY AND ASTROPHYSICS↗

Pushing Back the Frontier of Stability

Ask what holds protons and neutrons together in the nucleus and the answer is straightforward: It’s the strong force between quarks. Ask instead how many protons and neutrons a nucleus can contain before it falls apart and the answer isn’t nearly so concise. The reason is that the nucleus is a many-bodied object, and the interactions between protons and neutrons are complex. Models of these interactions have been successful at predicting the mass and structure of stable nuclei, but they must be modified to describe the neutron- and proton-rich isotopes that approach the edge of stability. The ability to model neutron-rich nuclei is of particular interest because they play a key role in the stellar nuclear reactions that form all heavy elements and the formation and structure of neutron stars.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Local magnetic cluster size identified by neutron total scattering in the site-diluted spin glass S n x F e 4 - x N ( x = 0.88 )

A detailed structure analysis for the site-diluted Sn x Fe 4- x N ( x = 0.25, 0.41, and 0.88) has been carried out through complex modeling of the neutron total scattering data. We present quantitative evidence showing the local ferromagnetic cluster size extending to ~8 Å on average when Sn 0.88 Fe 3.12 N undergoes the spin-glass transition (the other two not showing such a transition). In conclusion, the modeling methodology used in this work involving the corefinement of the nuclear and magnetic structure in both real and reciprocal space can potentially be applied generally to explore a variety of spin-glass material problems.

36 MATERIALS SCIENCE↗

mBEEF-vdW: Robust fitting of error estimation density functionals

Here, we propose a general-purpose semilocal/nonlocal exchange-correlation functional approximation, named mBEEF-vdW. The exchange is a meta generalized gradient approximation, and the correlation is a semilocal and nonlocal mixture, with the Rutgers-Chalmers approximation for van der Waals (vdW) forces. The functional is fitted within the Bayesian error estimation functional (BEEF) framework. We improve the previously used fitting procedures by introducing a robust MM-estimator based loss function, reducing the sensitivity to outliers in the datasets. To more reliably determine the optimal model complexity, we furthermore introduce a generalization of the bootstrap 0.632 estimator with hierarchical bootstrap sampling and geometric mean estimator over the training datasets. Using this estimator, we show that the robust loss function leads to a 10% improvement in the estimated prediction error over the previously used least-squares loss function. The mBEEF-vdW functional is benchmarked against popular density functional approximations over a wide range of datasets relevant for heterogeneous catalysis, including datasets that were not used for its training. Overall, we find that mBEEF-vdW has a higher general accuracy than competing popular functionals, and it is one of the best performing functionals on chemisorption systems, surface energies, lattice constants, and dispersion. We also show the potential-energy curve of graphene on the nickel(111) surface, where mBEEF-vdW matches the experimental binding length. mBEEF-vdW is currently available in gpaw and other density functional theory codes through Libxc, version 3.0.0.

36 MATERIALS SCIENCE↗

Applications of reduced-basis methods to the nuclear single-particle spectrum

Reduced-basis methods provide a powerful framework for building efficient and accurate emulators. Although widely applied in many fields to simplify complex models, reduced-basis methods have only been recently introduced into nuclear physics. In this Letter we build an emulator to study the single-particle structure of atomic nuclei. By scaling a suitable mean-field Hamiltonian, a “universal” reduced basis is constructed capable of accurately and efficiently reproduce the entire single-particle spectrum of a variety of nuclei. Indeed, the reduced-basis model reproduces both ground- and excited-state energies as well as the associated wave functions with remarkable accuracy. Here our results bode well for more demanding applications that use Bayesian optimization to calibrate nuclear energy density functionals.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Embedding hard physical constraints in neural network coarse-graining of three-dimensional turbulence

In recent years, deep learning approaches have shown much promise in modeling complex systems in the physical sciences. A major challenge in deep learning of partial differential equations is enforcing physical constraints and boundary conditions. In this work, we propose a general framework to directly embed the notion of an incompressible fluid into convolutional neural networks, and apply this to coarse-graining of turbulent flow. These physics-embedded neural networks leverage interpretable strategies from numerical methods and computational fluid dynamics to enforce physical laws and boundary conditions by taking advantage the mathematical properties of the underlying equations. Here, we demonstrate results on three-dimensional fully developed turbulence, showing that this technique drastically improves local conservation of mass, without sacrificing performance according to several other metrics characterizing the fluid flow.

97 MATHEMATICS AND COMPUTING↗

When more data hurts: Optimizing data coverage while mitigating diversity-induced underfitting in an ultrafast machine-learned potential

Machine-learned interatomic potentials (MLIPs) are becoming an essential tool in materials modeling. However, optimizing the generation of training data used to parametrize the MLIPs remains a significant challenge. This is because MLIPs can fail when encountering local environments too different from those present in the training data. The difficulty of determining a priori the environments that will be encountered during molecular dynamics simulation necessitates diverse, high-quality training data. Here, this study investigates how training data diversity affects the performance of MLIPs using the Ultra-Fast force field (UF 3 ) to model amorphous silicon nitride. We employ expert and autonomously generated data to create the training data and fit four force field variants to subsets of the data. Our findings reveal a critical balance in training data diversity: insufficient diversity hinders generalization, while excessive diversity can exceed the MLIP's learning capacity, reducing simulation accuracy. Specifically, we found that the UF 3 variant trained on a subset of the training data, in which nitrogen-rich structures were removed, offered vastly better prediction and simulation accuracy than any other variant. By comparing these UF 3 variants, we highlight the nuanced requirements for creating accurate MLIPs, emphasizing the importance of application-specific training data to achieve optimal performance in modeling complex material behaviors.

ab initio molecular dynamics↗