Search NASA⌕ Search

SEARCH · Search NASA

Results for “Stochastic Sampling”

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 793 records · Page 44

Modeling Rare Events in High Dimensional Stochastics Systems

The main issue addressed in this note is the study of an algorithm to accelerate the computation of kinetic rates in the context of molecular dynamics(MD). It is based on parallel simulations of short-time trajectories and its main components are: a decomposition of phase space into macrostates or cells, a resampling procedure that ensures that the number of parallel replicas (MD simulations) in each macro-state remains constant, the use of multiple populations (colored replicas) to compute multiple rates (e.g., forward and backward rates) in one simulation. The method leads to enhancing the sampling of macro-states associated to the transition states, since in traditional MD these are likely to be depleted even after short-time simulations. By allowing parallel replicas to carry different probabilistic weights, the number of replicas within each macro-state can be maintained constant without introducing any bias. The empirical variance of the estimated reaction rate, defined as a probability flux, is expectedly diminished. This note is a first attempt towards a better mathematical and numerical understanding of this method. It provides first a mathematical formalization of the notion of colors. Then, the link between this algorithm and a set of closely related methods having been proposed in the literature within the past few years is discussed. Lastly, numerical results are provided that illustrate the efficiency of the method.

59 BASIC BIOLOGICAL SCIENCES↗

Surrogate optimization of variational quantum circuits

Variational quantum eigensolvers are touted as a near-term algorithm capable of impacting many applications. However, the potential has not yet been realized, with few claims of quantum advantage and high resource estimates, especially due to the need for optimization in the presence of noise. Finding algorithms and methods to improve convergence is important to accelerate the capabilities of near-term hardware for VQE or more broad applications of hybrid methods in which optimization is required. To this goal, we look to use modern approaches developed in circuit simulations and stochastic classical optimization, which can be combined to form a surrogate optimization approach to quantum circuits. Using an approximate (classical CPU/GPU) state vector simulator as a surrogate model, we efficiently calculate an approximate Hessian, passed as an input for a quantum processing unit or exact circuit simulator. This method will lend itself well to parallelization across quantum processing units. We demonstrate the capabilities of such an approach with and without sampling noise and a proof-of-principle demonstration on a quantum processing unit utilizing 40 qubits.

Gustafson, Erik J. [RIACS, Mtn. View] (ORCID:00000↗

Stochastic Optimization and Uncertainty Quantification of Natrium-based Nuclear-Renewable Energy Systems for Flexible Power Applications in Deregulated Markets

Rapid integration of variable renewable energy sources (VRES) has made modeling and stochastic optimization of hybrid energy systems crucial for studying their long-term performance and viability. However, most studies have focused on just historical data, which may be unreliable for capturing short-term fluctuations, rare events, and long-term patterns of energy demand, price, and the variability of renewable energy sources. For this study, optimal synthetic time series models were developed using Wasserstein distance. The models were validated by comparing the key statistical measures against those of the historical data. They were then used to optimize the integrated Natrium-style advanced energy systems and their long-term (30 years) economics. The stochastic model performs bi-level optimization to find the optimal sizes for the balance of plant and thermal energy storage, while also optimizing energy dispatch to achieve the maximum net present value. In studies of two deregulated markets (California ISO and the Electric Reliability Council of Texas), the integrated Natrium-style system performed better in CAISO than in ERCOT, given higher and more consistent electricity prices during peak-demand periods. The potentially enlarged cost associated with the variable operation and maintenance of the TES system also plays a significant role in driving the system sizing, thus its impacts on the system are investigated in detail through comparison against a baseline case. The study also finds that the bi-level optimization results based on stochastic gradient descent closely match the grid search results. The uncertainty quantification of the stochastic signals provides further NPV-related insights and probability distributions for the case studies. The normal standard error of the mean of NPV for the case with and without TES VOM for CAISO were found to be 7.73M (plus-minus sign) 1.09M USD and 104.99M (plus-minus sign) 1.25M USD, respectively based on a 95% confidence. Given the relatively small NPV variance based on 150 samples, the analysis affords the most robust possible prediction of the techno-economic performance of the integrated Natrium-style energy systems.

25 ENERGY STORAGE↗

Accelerating multilevel Markov Chain Monte Carlo using machine learning models

Here, this work presents an efficient approach for accelerating multilevel Markov Chain Monte Carlo (MCMC) sampling for large-scale problems using low-fidelity machine learning models. While conventional techniques for large-scale Bayesian inference often substitute computationally expensive high-fidelity models with machine learning models, thereby introducing approximation errors, our approach offers a computationally efficient alternative by augmenting high-fidelity models with low-fidelity ones within a hierarchical framework. The multilevel approach utilizes the low-fidelity machine learning model (MLM) for inexpensive evaluation of proposed samples thereby improving the acceptance of samples by the high-fidelity model. The hierarchy in our multilevel algorithm is derived from geometric multigrid hierarchy. We utilize an MLM to accelerate the coarse level sampling. Training machine learning model for the coarsest level significantly reduces the computational cost associated with generating training data and training the model. We present an MCMC algorithm to accelerate the coarsest level sampling using MLM and account for the approximation error introduced. We provide theoretical proofs of detailed balance and demonstrate that our multilevel approach constitutes a consistent MCMC algorithm. Additionally, we derive the expression for cost reduction due to machine learning model to facilitate cost analysis of the hierarchical sampling algorithm. Our technique is demonstrated on a standard benchmark inference problem in groundwater flow, where we estimate the probability density of a quantity of interest using a four-level MCMC algorithm. Our proposed algorithm accelerates multilevel sampling by a factor of two while achieving similar accuracy compared to sampling using the standard multilevel algorithm.

97 MATHEMATICS AND COMPUTING↗

Anisotropy measurement of pyrolytic carbon layers of coated particles

Equipment at the National Research Center Kurchatov Institute intended for the anisotropy determination of pyrolytic carbon layers in coated particles (CPs) of the GT-MGR reactor is tested and calibrated. The dependence of the anisotropy coefficient on the size of the measurement region is investigated. The results of measuring the optical anisotropy factor (OPTAF) for an aluminum mirror, rutile crystal, and available CP samples with the known characteristics measured previously using ORNL equipment (United States) are presented. In addition, measurements of CP samples prepared at VNIINM are performed. A strong dependence of the data on the preparation quality of metallographic sections is found. Our investigations allow us to make the conclusion on the working capacity of the existing equipment for measuring the anisotropy of pyrolytic carbon CP coatings using the equipment at the Kurchatov Institute with the relative error of about 1%. It is shown that the elimination of the errors caused by the stochastic fluctuations in a measuring path by mathematical processing of the signal allows us to decrease the relative error of OPTAF measurements to ∼0.3%.

36 MATERIALS SCIENCE↗

Seasonal controls on isolated convective storm drafts, precipitation intensity, and life cycle as observed during GoAmazon2014/5

Abstract. Isolated deep convective cloud life cycle and seasonal changes in storm properties are observed for daytime events during the US Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) Green Ocean Amazon Experiment (GoAmazon2014/5) campaign to understand controls on storm behavior. Storm life cycles are documented using surveillance radar from initiation through maturity and dissipation. Vertical air velocity estimates are obtained from radar wind profiler overpasses, with the storm environment informed by radiosondes. Dry-season storm conditions favored reduced morning shallow cloud coverage and larger low-level convective available potential energy (CAPE) than wet-season counterparts. The typical dry-season storm reached its peak intensity and size earlier in its life cycle compared with wet-season cells. These cells exhibited updrafts in core precipitation regions (Z>35 dBZ) to above the melting level as well as persistent downdrafts aloft within precipitation adjacent to their cores. Moreover, dry-season cells recorded more intense updrafts to earlier life cycle stages as well as a higher incidence of strong updrafts (i.e., >5 m s−1) at low levels. In contrast, wet-season storms were longer-lived and featured a higher incidence of moderate (i.e., 2–5 m s−1) updrafts aloft. These storms also favored a shift in their most intense properties to later life cycle stages. Strong downdrafts were less frequent within wet-season cells aloft, indicating a potential systematic difference in draft behaviors, as linked to graupel loading and other factors between the seasons. Results from a stochastic parcel model suggest that dry-season cells may expect stronger updrafts at low levels because of larger low-level CAPE in the dry season. Wet-season cells anticipate strong updrafts aloft because of larger free-tropospheric relative humidity and reduced entrainment-driven dilution. Enhanced dry-season downdrafts are partially attributed to increased evaporation, dry-air entrainment mixing, and negative buoyancy in regions adjacent to sampled dry-season cores.

54 ENVIRONMENTAL SCIENCES↗

THE CHEMICAL SIGNATURES OF THE FIRST STAR CLUSTERS IN THE UNIVERSE

The chemical abundance patterns of the oldest stars in the Galaxy are expected to contain residual signatures of the first stars in the early universe. Numerous studies attempt to explain the intrinsic abundance scatter observed in some metal-poor populations in terms of chemical inhomogeneities dispersed throughout the early Galactic medium due to discrete enrichment events. Just how the complex data and models are to be interpreted with respect to 'progenitor yields' remains an open question. Here we show that stochastic chemical evolution models to date have overlooked a crucial fact. Essentially, all stars today are born in highly homogeneous star clusters and it is likely that this was also true at early times. When this ingredient is included, the overall scatter in the abundance plane [Fe/H] versus [X/Fe] (C-space), where X is a nucleosynthetic element, can be much less than derived from earlier models. Moreover, for moderately flat cluster mass functions ({gamma} {approx}< 2), and/or for mass functions with a high mass cutoff (M{sub max} {approx}> 10{sup 5} M{sub sun}), stars exhibit a high degree of clumping in C-space that can be identified even in relatively small data samples. Since stellar abundances can be modified by mass transfer in close binaries, clustered signatures are essential for deriving the yields of the first supernovae. We present a statistical test to determine whether a given set of observations exhibit such behavior. Our initial work focuses on two dimensions in C-space, but we show that the clustering signal can be greatly enhanced by additional abundance axes. The proposed experiment will be challenging on existing 8-10 m telescopes, but relatively straightforward for a multi-object echelle spectrograph mounted on a 25-40 m telescope.

79 ASTRONOMY AND ASTROPHYSICS↗

Thermal neutron scattering measurements and modeling of yttrium-hydrides for high temperature moderator applications

In this work, thermal neutron scattering measurements of yttrium hydride from 5 to 1,200 K were conducted to determine the change in inelastic scattering as a function of temperature and to probe for anharmonic effects. Additionally, measurements on samples of YH x from x = 1.62 to 1.90 were done to determine the effects of varying hydrogen concentration on the inelastic spectra. Changes in temperature affected the inelastic spectra in unanticipated ways, indicating that there are anharmonic effects, whereas hydrogen concentration did not significantly affect the inelastic spectra. These measurements were compared against the ENDF/B-VIII.0 thermal scattering files of YH 2 , as well as new thermal scattering files created by using the stochastic temperature-dependent effective potential (s-TDEP). Both libraries were found to be in good agreement with the experimental data at lower temperatures. At higher temperatures, the s-TDEP method is better at predicting the experimentally observed softening of phonon modes in acoustic and optical regions.

36 MATERIALS SCIENCE↗

Model verification: synthetic single pattern simulations using seismic reflection data

During Phase 1 of the Weyburn Project (2000-2004), 4D reflection seismic data were used to map CO2 migration within the Midale reservoir, while an extensive fluid sampling program documented the geochemical evolution triggered by CO2-brine-oilmineral interactions. The aim of this task (3b.11) is to exploit these existing seismic and geochemical data sets, augmented by CO2/H2O injection and HC/H2O production data toward optimizing the reservoir model and thereby improving site characterization and dependent predictions of long-term CO2 storage in the Weyburn-Midale reservoir. Our current project activities have concentrated on completing and testing a stochastic inversion method that will identify reservoir models that optimize agreement between the observed and predicted seismic response. This report describes the results of a validation test that uses synthetic seismic data to identify optimal porosity/permeability distributions within the reservoir. The report partially fulfills deliverable D3: “Model verification: synthetic single pattern simulations” in the project’s statement of work. A future deliverable will describe verification activities related to the geochemical inversion algorithm.

58 GEOSCIENCES↗

MINIMUM REQUIREMENTS FOR DETECTING A STOCHASTIC GRAVITATIONAL WAVE BACKGROUND USING PULSARS

We assess the detectability of a nanohertz gravitational wave (GW) background in a pulsar timing array (PTA) program by considering the shape and amplitude of the cross-correlation function summed over pulsar pairs. The distribution of correlation amplitudes is found to be non-Gaussian and highly skewed, which significantly influences detection and false-alarm probabilities. When only white noise combines with GWs in timing data, our detection results are consistent with those found by others. Contamination by red noise from spin variations and from any uncorrected interstellar plasma effects significantly increases the false-alarm probability. The number of arrival times (and thus the observing cadence) is important only as long as the residuals are dominated by white noise. When red noise and GWs dominate, the statistical significance of the correlation estimate can be improved only by increasing the number of pulsars. We characterize plausible detection regimes by evaluating the number of millisecond pulsars (MSPs) that must be monitored in a high-cadence, five-year timing program to detect a GW background spectrum h{sub c} (f) = A(f/f{sub 0}){sup -2/3} with f{sub 0} = 1 yr{sup -1} and A = 10{sup -15}. Our results indicate that a sample of 20 super-stable MSPs-those with rms timing residuals {sigma}{sub r} {approx}< 20 ns(A/10{sup -15}) from red-noise contributions over a five-year span-will allow detection of the GW background and study of its spectrum. However, a timing program on {approx}> 50-100 MSPs is likely needed for a complete PTA program, particularly if red noise is generally present in MSPs.

79 ASTRONOMY AND ASTROPHYSICS↗

Identifying the Critical Links in Road Transportation Networks: Centrality-based approach utilizing structural properties

Surface transportation road networks share structural properties similar to other complex networks (e.g., social networks, information networks, biological networks, and so on). This research investigates the structural properties of road networks for any possible correlation with the traffic characteristics such as link flows those determined independently. Additionally, we define a criticality index for the links of the road network that identifies the relative importance in the network. We tested our hypotheses with two sample road networks. Results show that, correlation exists between the link flows and centrality measures of a link of the road (dual graph approach is followed) and the criticality index is found to be effective for one test network to identify the vulnerable nodes.

99 GENERAL AND MISCELLANEOUS↗

Fracture Network Influence on Rock Damage and Gas Transport following an Underground Explosion

Simulations of rock damage and gas transport following underground explosions that omit preexisting fracture networks in the subsurface cannot fully characterize the influence of geo-structural variability on gas transport. Previous studies do not consider the impact that fracture network structure and variability have on gas seepage. In this study, we develop a sequentially coupled, axi-symmetric model to look at the damage pattern and resulting gas breakthrough curves following an underground explosion given different fracture network realizations. We simulate 0.327 and 0.164 kT chemical explosives with burial depths of 100 m for 90 stochastically generated fracture networks. Gases quickly reach the surface in 30% of the higher yield simulations and 5% of the lower yield simulations. The fast breakthrough can be attributed to the formation of connected pathways between fractures to the surface. The formation of a connected damage pathway to the surface is not clearly correlated with the fracture intensity (P32) in our simulations. Breakthrough curves with slower transport are highly variable depending on the fracture network sample. The variability in the breakthrough behavior indicates that ignoring the influence of fracture networks on rock damage, which strongly influences the hydraulic properties following an underground explosion, will likely lead to a large underestimation of the uncertainty in the gas transport to the surface. This work highlights the need for incorporation of fracture networks into models for accurately predicting gas seepage following underground explosions.

58 GEOSCIENCES↗

Manifold Learning-Based Polynomial Chaos Expansions for High-Dimensional Surrogate Models

In this work we introduce a manifold learning-based method for uncertainty quantification (UQ) in systems describing complex spatiotemporal processes. Our first objective is to identify the embedding of a set of high-dimensional data representing quantities of interest of the computational or analytical model. For this purpose, we employ Grassmannian diffusion maps, a two-step nonlinear dimension reduction technique which allows us to reduce the dimensionality of the data and identify meaningful geometric descriptions in a parsimonious and inexpensive manner. Polynomial chaos expansion is then used to construct a mapping between the stochastic input parameters and the diffusion coordinates of the reduced space. An adaptive clustering technique is proposed to identify an optimal number of clusters of points in the latent space. The similarity of points allows us to construct a number of geometric harmonic emulators which are finally utilized as a set of inexpensive pretrained models to perform an inverse map of realizations of latent features to the ambient space and thus perform accurate out-of-sample predictions. Thus, the proposed method acts as an encoder-decoder system which is able to automatically handle very high-dimensional data while simultaneously operating successfully in the small-data regime. The method is demonstrated on two benchmark problems and on a system of advection-diffusion-reaction equations which model a first-order chemical reaction between two species. In all test cases, the proposed method is able to achieve highly accurate approximations which ultimately lead to the significant acceleration of UQ tasks.

42 ENGINEERING↗

Multiscale method based on coupled lattice‐Boltzmann and Langevin‐dynamics for direct simulation of nanoscale particle/polymer suspensions in complex flows

Summary A hybrid computational method coupling the lattice‐Boltzmann (LB) method and a Langevin‐dynamics (LD) method is developed to simulate nanoscale particle and polymer (NPP) suspensions in the presence of both thermal fluctuation and long‐range many‐body hydrodynamic interactions (HIs). Brownian motion of the NPP is explicitly captured by a stochastic forcing term in the LD method. The LD method is two‐way coupled to the nonfluctuating LB fluid through a discrete LB forcing source distribution to capture the long‐range HI. To ensure intrinsically linear scalability with respect to the number of particles, a Eulerian‐host algorithm for short‐distance particle neighbor search and interaction is developed and embedded to LB‐LD framework. The validity and accuracy of the LB‐LD approach are demonstrated through several sample problems. The simulation results show good agreements with theory and experiment. The LB‐LD approach can be favorably incorporated into complex multiscale computational frameworks for efficiently simulating multiscale multicomponent particulate suspension systems such as complex blood suspensions.

Liu, Zixiang↗

Monte Carlo MP2-F12 for Noncovalent Interactions: The C 60 Dimer

A scalable stochastic algorithm is presented that can evaluate explicitly correlated (F12) second-order many-body perturbation (MP2) energies of weak, noncovalent, intermolecular interactions. It first transforms the formulas of the MP2 and F12 energy differences into a short sum of high-dimensional integrals of Green’s functions in real space and imaginary time. Furthermore, these integrals are then evaluated by the Monte Carlo method augmented by parallel execution, redundant-walker convergence acceleration, direct-sampling autocorrelation elimination, and control-variate error reduction. By sharing electron-pair walkers across the supermolecule and its subsystems spanned by the joint basis set, the statistical uncertainty is reduced by one to 2 orders of magnitude in the MP2 binding energy corrected for the basis-set incompleteness and superposition errors. The method predicts the MP2-F12/aug-cc-pVDZ binding energy of 19.1 ± 4.0 kcal mol –1 for the C 60 dimer at the center distance of 9.748 Å.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Relaxation dynamics of vortex lines in disordered type-II superconductors following magnetic field and temperature quenches

We study the effects of rapid temperature and magnetic field changes on the nonequilibrium relaxation dynamics of magnetic vortex lines in disordered type-II superconductors by employing an elastic line model and performing Langevin molecular dynamics simulations. In a previously equilibrated system, either the temperature is suddenly changed or the magnetic field is instantaneously altered which is reflected in adding or removing flux lines to or from the system. Here, the subsequent aging properties are investigated in samples with either randomly distributed pointlike or extended columnar defects, which allows us to distinguish the complex relaxation features that result from either type of pinning centers. One-time observables such as the radius of gyration and the fraction of pinned line elements are employed to characterize steady-state properties, and two-time correlation functions such as the vortex line height autocorrelations and their mean-square displacement are analyzed to study the nonlinear stochastic relaxation dynamics in the aging regime.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Gradient-based Optimization for Regression in the Functional Tensor-Train Format

We consider the task of low-multilinear-rank functional regression, i.e., learning a low-rank parametric representation of functions from scattered real-valued data. Our first contribution is the development and analysis of gradient-based optimization procedures, including stochastic gradient descent and quasi-Newton methods, for learning the parameters of a functional tensor-train (FT) that yields improved accuracy over standard alternating least squares methods. The functional tensor-train uses the tensor-train (TT) representation of low-rank arrays as an ansatz for a class of low-multilinear-rank functions. The FT is represented by a set of matrix-valued functions that contain a set of univariate functions, and the regression task is to learn the parameters of these univariate functions. Our second contribution demonstrates that using nonlinearly parameterized univariate functions, e.g., symmetric kernels with moving centers, within each core can outperform the standard approach of using a linear expansion of basis functions. Our final contributions are new rank adaptation and group-sparsity regularization procedures to minimize overfitting. We use several benchmark problems to demonstrate at least an order of magnitude lower accuracy with gradient-based optimization methods than ALS in the low-sample number regime. We also demonstrate an order of magnitude reduction in accuracy on a test problem resulting from nonlinear parameterizations over linear parameterizations. Finally we compare regression performance with 22 other nonparametric and parametric regression methods on 10 real-world data sets. We achieve top-five accuracy for seven of the data sets and best accuracy for two of the data sets. These rankings are the best amongst parametric models and competetive with the best non-parametric methods.

97 MATHEMATICS AND COMPUTING↗

Gaussian wave packet transform based numerical scheme for the semi-classical Schrödinger equation with random inputs

In this work, we study the semi-classical limit of the Schrödinger equation with random inputs, and show that the semi-classical Schrödinger equation produces O(ε) oscillations in the random variable space. With the Gaussian wave packet transform, the original Schrödinger equation is mapped to an ordinary differential equation (ODE) system for the wave packet parameters coupled with a partial differential equation (PDE) for the quantity w in rescaled variables. Further, we show that the w equation does not produce ε dependent oscillations, and thus it is more amenable for numerical simulations. We propose multi-level sampling strategy in implementing the Gaussian wave packet transform, where in the most costly part, i.e. simulating the w equation, it is sufficient to use ε independent samples. We also provide extensive numerical tests as well as meaningful numerical experiments to justify the properties of the numerical algorithm, and hopefully shed light on possible future directions.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗